Lazarus

Miscellaneous => Other => Topic started by: Webdrifter on August 19, 2020, 09:37:02 am

Title: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 19, 2020, 09:37:02 am
Hello,

We all know that (one of) the main problem of Free Pascal is it's shortish of programmers that can maintain and develop the language, it's tools and its eco-system. And we all know this has all to do with it's popularity. Which has all to do with it's limited eco-system and it's pace of development. In short, Pascal (and by that we as a community) are in a catch 22 situation. 

I believe that Pascal can in the end only survive when we can break the catch 22. And that this is only possible by giving more attention in our strategy and putting more effort in attracting more programmers towards Pascal (preferable those from the Python community because Pascal already fulfills many of the needs they have).

Of course this should start with getting more attention from the programmers community as a whole.
Ranking in indexes like the TIOBE-index has everything to do with this.
It's just like in politics. People like to get on the bandwagon of fast growing parties, even when they are still small.

The TIOBE programming community index is a measure of popularity of programming languages, created and maintained by the TIOBE Company based in Eindhoven, the Netherlands. Maintainers specify that the TIOBE index is "not about the best programming language or the language in which most lines of code have been written", but do claim that the number of web pages may reflect the number of skilled engineers, courses and jobs worldwide.

See here the way they works: https://www.tiobe.com/tiobe-index/programming-languages-definition/

They deduct their rankings from the pages found in major search engines (mainly Google but not all language version off Google and thus forgetting counties like the Netherlands).

So to get a better ranking we can try to chance their policy and/or adept to it.....

For instance, they split the pascal community up into 2 languages:
-  Delphi/Object Pascal
-  Pascal

This splitting up results in the fact that they both end up with a lower ranking!!!

End guess what they don't even rate Free pascal. I asked them why.
The response is that Free Pascal is not defined in Wikipedia as a programming language but as a compiler.
And that the same count for Lazarus because it is defined as a development environment.

They also stated that explicitly for that purpose Delphi has changed the definition of Delphi in Wikipedia into:
"Delphi is a programming language and an integrated development environment".

By doing so they have effectively split up the Pascal community in the rankings (Which is dumb).
And page hits concerning their development environment are counted as hits for their language (Which is clever).

So what could we do?

Change the name(s) of "Free Pascal" and/or "Lazarus" ?

Change the definition(s) of "Free Pascal" and/or "Lazarus" in Wikipedia?

Have Delphi change it's definition in Wikipedia in a way that the split up in the Tiobe rankings get avoided?

At least it seems wise to start a discussion about this among ourselves and get into action in conference with Delphi and Tiobe representatives.

And of course we need an entire strategy and task-force to get us noticed better out there.....
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Zoran on August 19, 2020, 10:33:15 am
At least it seems wise to start a discussion about this among ourselves and get into action in conference with Delphi and Tiobe representatives.

There has been a long discussion about it (https://forum.lazarus.freepascal.org/index.php/topic,13771.msg372563.html#msg372563).
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 19, 2020, 10:43:18 am
OK. I'll read that.

In the meantime I had some extra contact with a Tiobe representative (Paul Jansen). He advised me to at least change the definition in Wikipedia on webpage: https://en.wikipedia.org/wiki/Free_Pascal

It now says:
"Free Pascal Compiler (FPC) is a compiler for the closely related programming-language dialects Pascal and Object Pascal. It is free software released under the GNU General Public License, with exception clauses that allow static linking against its runtime libraries and packages for any purpose in combination with any other software license.

It supports its own Object Pascal dialect, as well as the dialects of several other Pascal family compilers to a certain extent, including those of Turbo Pascal, Delphi, and some historical Macintosh compilers. The dialect is selected on a per-unit (module) basis, and more than one dialect can be used per program."

It would like to advise to change it in:
"Free Pascal is modern dialect of the Pascal programming language (just as it's closely related "cousin" Delphi). The Free Pascal Compiler (FPC) compiles the closely related programming-language dialects Pascal and Object Pascal. It is free software released under the GNU General Public License, with exception clauses that allow static linking against its runtime libraries and packages for any purpose in combination with any other software license.

It supports its own Object Pascal dialect (Free Pascal), as well as the dialects of several other Pascal family compilers to a certain extent, including those of Turbo Pascal, Delphi, and some historical Macintosh compilers. The dialect is selected on a per-unit (module) basis, and more than one dialect can be used per program."

We also can and also have to do something about the split up into communities.

Because when we address Free Pascal as a language with it's own separate entry in Wikipedia it will also be counted in the Tiobe-index as separate language. We have to prevent this! For instance: Object Pascal and Delphi are separate entries in Wikipedia but they made a redirection in wikipedia from the term "Delphi programming language" towards the page "Object Pascal" in which they mentioned that Delphi stems from Object Pascal, together with a link to the "Delphi (Software)" page in Wikipedia. This makes it merge for Tiobe.

We should do the same (A redirection from "Free Pascal Programming language" towards the "Object Pascal" page. Make a link in the "Object Pascal' page towards our own "Free Pascal" page in Wikipedia . And also make links in our own "Free Pascal'page towards the "Delphi" en the "Object Pascal" page.

Next we should try to find a way to get the "Pascal (programming language)" page and the "Object Pascal" page merged, the same way.

Unfortunately this won't help us to not get the web pages for Lazarus included in our ranking.

But it is start.

PS: I won't make these changes in Wikipedia by myself, because I wish to leave this to the Pascal community and it's leaders. But when asked by them, I can see what I can do.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Thaddy on August 19, 2020, 11:41:38 am
Even that is not correct or do justice.
FreePascal is also ISO compliant.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on August 19, 2020, 01:17:49 pm
It would like to advise to change it in:
"Free Pascal is modern dialect of the Pascal programming language (just as it's closely related "cousin" Delphi). The Free Pascal Compiler (FPC) compiles the closely related programming-language dialects Pascal and Object Pascal. It is free software released under the GNU General Public License, with exception clauses that allow static linking against its runtime libraries and packages for any purpose in combination with any other software license.

It supports its own Object Pascal dialect (Free Pascal), as well as the dialects of several other Pascal family compilers to a certain extent, including those of Turbo Pascal, Delphi, and some historical Macintosh compilers. The dialect is selected on a per-unit (module) basis, and more than one dialect can be used per program."

Free Pascal is not a dialect, it is a compiler. Period. If the Tiobe creators are too stubborn to realize that their approach is flawed then this is not our problem.

Unfortunately this won't help us to not get the web pages for Lazarus included in our ranking.

But it is start.

No, it is not. It's the approach used by Tiobe that is flawed.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: 440bx on August 19, 2020, 02:05:15 pm
Free Pascal is not a dialect, it is a compiler. Period.
That's not really convincing.  Free Pascal is a compiler, no doubt about that.  It is a compiler for a dialect of Pascal, no doubt about that either.   Given those two facts, it seems quite reasonable to state that Free Pascal is a Pascal dialect (obviously a compiler for that specific dialect.)

If the Tiobe creators are too stubborn to realize that their approach is flawed then this is not our problem.
I certainly don't think that TIOBE's method of measuring the usage of Pascal is accurate but, their ranking can definitely influence the decision of which language a _new_ programmer wishes to learn.


Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 19, 2020, 02:33:22 pm
I totally agree with 440bx. In the end it's all a matter of semantics. Of course we can be stubborn about it. but what would that do for the Pascal community we are all a part of. I believe it worth being savvy about it. We are developers. So why can't we be as creative in this as in our programming?

For instance why not call this the (Free) Pascal forum instead of the Lazarus forum? That would get us a lot of extra hits and thus extra "popularity" on the index. Of course Lazarus would remain a large part of the forum and we would also have to link from the Lazarus site to the forum.

End to get even more creative. I believe that Lazarus is THE most important reason for using Pascal at this moment. Even so Lazarus could be a great tool for Python programmers if only they could use it to code Python. For some it might even be enough if they could input Python even if it outputs Pascal code. That would not only bring them in contact with Pascal (Which is a language that could suit most Python programmers well), it would also create many extra hits towards the renamed  "Pascal" Forum.

And it would attach them with the Pascal community. Which would be great because many of them are scholars and academics with the knowledge and time to help enhance the Pascal eco-system.

And even more creative! How hang up are we with our sentiments on the name "Lazarus". Renaming it "Pascal" would greatly enhance the number of hits and automatically create a nice link in Wikipedia between our RAD and the Pascal language and thus also between Free Pascal and Pascal.

It's all about being creative... :D       
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 19, 2020, 03:20:46 pm
Even so Lazarus could be a great tool for Python programmers if only they could use it to code Python. For some it might even be enough if they could input Python even if it outputs Pascal code.
Go ahead, write it as a package (or find someone who will, which I guess is what you are trying to do with this post). Then put it into the online package manager.

Quote
How hang up are we with our sentiments on the name "Lazarus".
VERY


Besides, if I google "Pascal" Lazarus is the 8th result
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 19, 2020, 03:45:02 pm
At least it seems wise to start a discussion about this among ourselves and get into action in conference with Delphi and Tiobe representatives.

There has been a long discussion about it (https://forum.lazarus.freepascal.org/index.php/topic,13771.msg372563.html#msg372563).

I have just finished reading the thread you suggested. And I read a lot of whining there, but very little positive creative contributions that could help the community. Are we all really getting that old? Because with an attitude like that we will be getting no further than dying out as a community together with us old-farts dying ourselves..... 
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 19, 2020, 03:59:39 pm
Even so Lazarus could be a great tool for Python programmers if only they could use it to code Python. For some it might even be enough if they could input Python even if it outputs Pascal code.
Go ahead, write it as a package (or find someone who will, which I guess is what you are trying to do with this post). Then put it into the online package manager.

Quote
How hang up are we with our sentiments on the name "Lazarus".
VERY


Besides, if I google "Pascal" Lazarus is the 8th result

- In no way was my intention as you suggest above. It was just an idea. It was truly only about hoping to get new blood into the community (for which we need ideas and not barriers on the road). Since that is the only thing that can make it flourish in a way that you all get a better pascal eco-system. I see it all as a means to a very important end (unless nobody cares about a better Pascal eco-system off-course).

- I understand how especially people like you will have their sentiments about things, but how far are you willing to go with that if it frustrates getting what you probably care for even more....

- I don't understand you remark on googling "pascal". Since clearly the problem is that people don't do that often enough, for one of the reasons being because they Google "Lazarus" instead.

_ I would greatly applaud ideas from your side, especially because in these quarter's your voice carries far (and rightly so I might add).


Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: lainz on August 19, 2020, 06:00:27 pm
I doubt any user can do nothing to increase the popularity of pascal, but some things can be
- Writing articles on how you use Pascal at work
- Making packages and put them on the Online Package Manager

At university I did not learn Pascal, but was previous knowledge. At university is teached C and Java, JavaScript in the last course I take this year.

So Pascal is not even being teached, there are paid courses that are 'learn free pascal'? Not books, but modern interactive courses with online tools.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Handoko on August 19, 2020, 06:04:25 pm
@Webdrifter

I appreciate you want to make Pascal eco-system better. I care and we all care because all of us here are using it. But the problem is not as simple as care or willingness.

I am not a member of the development team and I cannot represent the community. I just a hobbyist who use Pascal. I just want to say if you have any idea to make it better, just do it. No need to wait for the others.

Maybe you haven't known the community is currently lack of man power. The development team is not paid for developing it, they have their own life. They have family, kids and job. The documentation, especially the wiki need some massive work. And there is a long list of bugs in the bugtracker. As a not very skilled person here, not much help I can do. I can only spend some of my time answering newbies' questions in the forum.

You seem have many ideas. Some sound good. If you think that is the right thing to do, go ahead. About the Pascal-Python bridge, that is a nice thing. There were several discussions recently talking about it.

And you said you can improve the Wikipedia, that's great. I wish I can do it myself, unfortunately I can't. My English is not good enough for formal writing.

About the TIOBE, I heard some users here personally wrote to TIOBE explaining that they're wrong. If you can, you should do it too, to add some pressure to them.

I have a blog about graphics. I was thinking to add some Pascal graphics programming tutorials there. I'm working on it now.

Lets do what we can to make Pascal great.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 19, 2020, 07:14:57 pm
Even so Lazarus could be a great tool for Python programmers if only they could use it to code Python. For some it might even be enough if they could input Python even if it outputs Pascal code.
Go ahead, write it as a package (or find someone who will, which I guess is what you are trying to do with this post). Then put it into the online package manager.

Quote
How hang up are we with our sentiments on the name "Lazarus".
VERY


Besides, if I google "Pascal" Lazarus is the 8th result

- In no way was my intention as you suggest above. It was just an idea. It was truly only about hoping to get new blood into the community (for which we need ideas and not barriers on the road). Since that is the only thing that can make it flourish in a way that you all get a better pascal eco-system. I see it all as a means to a very important end (unless nobody cares about a better Pascal eco-system off-course).

Which is absolutely fine. Great even.

The name part, is in the end a matter of opinion. I made mine clear. That is just my opinion, no idea what other team members think.

As for the rest, if I may, I expect that you did post that not merely as "nice to mention", but rather "seeking ways for actual achievement"

Now if something should be achived, there a small numbers of ways (Including combination of the below)
1) Do it yourself
2) Have someone else do it. That splits into
a) pay them
b) give them some other motivation (e.g. ask a  favour, if you are owed one)
c) pray and hope someone will just do it. (If no one has yet, and the idea is not completely revolutionary new, then the question begs why someone would do now, if they have not in the past.)
3) Lobbying (hoping for 2c)


Quote
- I understand how especially people like you will have their sentiments about things, but how far are you willing to go with that if it frustrates getting what you probably care for even more....
Well if this is about the name. It is not just my decision. Its for the team. My vote, if it came that far, is clear. Other than that I will do no lobbying, or other work to make it happen. But you are free (and welcome) to lobby for it, and try to reach us many people as you can.

As for the meaning of the name FPC / FreePascal => that is for the fpc team. (I am part of the Lazarus team)

As for other things (communicating with TIOBE, working on wikipedia, blogging, tweeting, or finding people who do so): I am fine if that happens. I have no objections to that. But I will not spent my time on it. (Not the reason why I joined the project, there are simply to many things, and I can only to some of them)

The last paragraph is kind of what prompted my previous reply: You need to find people that have time and willingness. Getting a better rank is good, and I guess no one will object.
My experience is that a sentence like "Something should be done" (in your case: add python), will lead to everyone thinking => great idea, wouldn't mind, wait for someone => and in the end no-one does it (this has happen countless times, but maybe you are more lucky)

So basically I wanted to say: According to my experience you need to put in more effort or it is not going to happen.
There are a lot of suggestions what could be added. Very few are picked up by existing contributors. Because those people are already flooded with work.

As for my suggestion "online package manager": Again my opinion. OPM can be added without the team needing to agree.
If you (assuming someone would be there to implement it) wanted it to be shipped with the default installer => needs team agreement.
As for, would I agree? Well maybe later, that is if it proves really popular on OPM, then maybe.
(Btw there is Pas2Js ... just to mention - its not impossible)


I am not saying your ideas can/should not happen. I am only pointing out what I have observed to happen to similar movements.
And it be a shame if this would go the same way....


Quote
- I don't understand you remark on googling "pascal". Since clearly the problem is that people don't do that often enough, for one of the reasons being because they Google "Lazarus" instead.
Sorry that was unclear....

About renaming Lazarus to include Pascal => I wanted to see, how much google already associates those terms.

Also adding the other way, if I search for "Lazarus", both the IDE and free pascal are in the top results. So google associates Lazarus<>Pascal.
Which may mean renaming, will not increase the amount of pages found.

Also
+"pascal progamming"  gives 185000 results
+"lazarusl progamming" 41000
reduce the overlaps, and renaming would maybe add 10k to 20k?

Compare that to the Millions of pages other language have, it wont do much of a difference.
Well, yes: every little helps.

But the point is, a lot more is needed, that a name change would bring (not to mention that many people associate the name, and could be lost if it changed).
Changing the name would be one of the most dramatic steps one could do (and still the gain would be so small....)
So IMHO that is certainly not a step that stand anywhere at the top of the list.


Quote

_ I would greatly applaud ideas from your side, especially because in these quarter's your voice carries far (and rightly so I might add).

Well the main issue is: We are programmers not publicists. (Even writing documentation could need extra help).

We have a twitter account, and a developer blog. But none of us, feels like writing anything there.

Anyone can create their own blog, twitter, facebook.
Follow the project:
- Interesting info here on the forum?
- svn commit that might be worth talking about (ask on the mail list, if the committer is willing to give you background, and then make a story of it)
- fpc / release notes
- just explore what exists, pick random features and write about them

If we can get a handful of active blogs/twitter (i.e. at least one or more posts per week, with interesting content) => I would guess that should attract people.

If you look at the fact that people retweet our (boring) "new release" tweets... Well imagine a well written twitter account.
Only, none of us has the time to do this. So find people who can.


Add to that: Better press releases.
One of the biggest German papers. https://www.heise.de/developer/meldung/Entwicklungsumgebung-Lazarus-2-0-fuehrt-Pascal-zu-JavaScript-Transpiler-ein-4303977.html
I do not know who mailed them. But I think the text is exactly what was mailed. They did not change it.
So if we had someone who does better press release announcements. And feeds them to more papers....


Anyway, it all needs more contributors (in that case not for code-writing, but article-writing).
So the question is: How to get them?

If you have time, you could start a blog, or pre-draft an announcement for a newspaper. Once you have something, if you ask who would be willing to work with you, that may attract more people (hopefully, I really do not know).


There may be many more ways.

My 2 cents.



Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Windsurfer on August 19, 2020, 09:56:12 pm
I have a lot of sympathy for the seemingly revolutionary views expressed here. If a simple tweak to wikipedia could effect significant change to the situation, why not?

I switched a whole team from VB to Delphi when it first came out, and would have switched to Lazarus when it became available and stable had I still been managing a software team.

However, I did not discover fpc and Lazarus until the mid 90s. I suspect that this was partly to do with the name. IIRC fpc could be found when searching for pascal, but it did not then mention Lazarus. There was no way anyone used to RAD programming would want go backwards.

Had Lazarus had a tag added like 'Pascal RAD IDE' so that it always appeared as 'Lazarus - Pascal RAD IDE',  I and many others might have found it sooner, and search engines would also have found it.

Psychologists talk about reaction to change as following a sequence of Denial, Resistance, Exploration and Commitment. They also warn about it taking time.  Please keep thinking and talking about it.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 19, 2020, 11:34:40 pm
I have a lot of sympathy for the seemingly revolutionary views expressed here. If a simple tweak to wikipedia could effect significant change to the situation, why not?

But will it? Or might it not be so simple? If it is all that important, why doesn't the TIOBE guy maintain a few exceptions, and why did he change his algorithm wrt Pascal several times in the last 2 years?

Most likely it will just progress to the next artificial hurdle, and even if it wouldn't, it won't bring in the hordes, as it is only a minor additional decision maker.

Nice to have? Sure. But IMHO vastly overrated.  The same for SEO by name change, which are only temporary effects at best to an already known brand.  Maybe not perfect, but it might take years to get the new brands up in the search engine rankings.

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 20, 2020, 12:42:54 am
However, I did not discover fpc and Lazarus until the mid 90s.
That is an achievement....
Lazarus was only created around 1999 / 2000.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 20, 2020, 12:43:53 am
However, I did not discover fpc and Lazarus until the mid 90s.
That is an achievement....
Lazarus was only created around 1999 / 2000.

And used GTK on windows till what, end of 2003?
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: dbannon on August 20, 2020, 02:50:46 am
While I am no fan of badge engineering, I do think Webdrifter has a point.  If its about naming, and doing naming correctly can make a difference, a discussion is a good idea.  I suspect if we, the FPC, Object Pascal, Lazarus, LCL, (and Delphi?) users, agreed on a simple set of names it could make the products we love more recognizable.  Trouble is, these things are all subsets of each other.

The Free Pascal Compiler is a compiler, obviously. But "Free Pascal" does sound to me like a programing language (sorry PascalDragon). In fact a Pascal dialect, one of several dialects supported by the FPC.  Object Pascal is clearly a Language (or dialect), definitely not a substitute for "Free Pascal" because we have lots of code written for the FPC that is not OO code.  Similarly, many of us think we are writing Lazarus code, no, its Free Pascal, maybe Object Pascal, maybe Pascal code written using a Lazarus IDE and 'linked' to the Lazarus Component Library.  And I guess Delphi calls its underlying code "Delphi" ?  And would not consider changing it.

I suggest that one term be picked, even if its not perfect.  is that term "Free Pascal" ?  If so, it should appear in the "about" boxes we pop up and in marketing literature.  Mention Lazarus as well by all means but Free Pascal should be the eye catching tag.  Right now, my application says "...using Lazarus and FPC", maybe I will change that to say, hmm, -

"A Free Pascal application built using FPC and Lazarus"

That approach definitely conflicts with what Wikipedia says. We have several Wikipedia pages, the most prominent, "Free Pascal" starts off defining the "Free Pascal Compiler" and perhaps needs to be renamed ?  (them's fighting words!).  We also have an "Object Pascal" and "Pascal Programming Language".   If we are to fit into the existing Wikipedia model, my about box should say -

"A Pascal application built using Free Pascal and Lazarus"

And, for what its worth, github identifies my code as "Pascal".  Would we convince Tiobe that there is one Pascal to Rule the World ?  Things like Delphi and Free Pascal are just manifestations of the great Pascal Family ?

(Yes, I know Tiobe is a flawed measure but people like to refer to it.  We could invent our own measure, perhaps one that excludes languages that don't use "begin ... end" but nobody would care !)

Davo
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: TRon on August 20, 2020, 03:38:44 am
I stopped using Free Pascal for a long time now. These days I program in Generics Pascal  :-X
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: trev on August 20, 2020, 04:37:41 am
There may be many more ways.

As previously mentioned, the Wiki (https://wiki.freepascal.org/Main_Page) needs love and the best part is that ANYONE can edit the Wiki to update out of date material and/or add new material. It is also well represented in G-o-o-g-l-e search results. I've spent a year on the Wiki fixing template loops, categorising uncategorised pages, linking orphaned pages, deleting dead pages (eg foreign language pages which are wholly in English) and translating eg the German AVR and ARM tutorial pages (among others) into English.

Apart from that, I've been concentrating on the Wiki's macOS Portal and creating new macOS-specific content pages with example code. I also, with help from Martin without whom it would never have happened, updated the Wiki's long static main page to help make the content more accessible, especially via the Platform and Thematic Portals that are now listed there.

Another avenue for ANYONE to contribute is logging bugs. Most of those that I've logged have been resolved and where they haven't been resolved for one reason or another, I've published workarounds or patches on the Wiki.

I, along with a number of others around here, contribute monthly to a developer's Patreon account (and have in the past donated Apple hardware). Again, something ANYONE can do.

Finally, the reality is that Pascal is no longer being taught formally in most universities and this of course affects the number of new Pascal programmers. Whether as a consequence or not, there are now very few jobs for Pascal programmers. This is not to say that Pascal is dying as some do, but rather that it is now a road much less travelled by programmers. It is a niche, if you like. Nothing wrong with that. The consequence though is that one needs to adjust one's expectations and not be deluded into believing that the reduction in the popularity of Pascal can be arrested or even reversed. That's about as likely as a snowball surviving an Australian summer bushfire.

Before Embarcadero managed to come up with the much promised and much delayed macOS compiler, there was an opportunity that lasted at least two years for Lazarus and Free Pascal to own that market space. It didn't. Consider why it didn't. I think the answer is obvious. There's the largely ineffective (imnsho) Free Pascal and Lazarus Foundation. There's the volunteer group of Pascal developers. There's the volunteer group of Lazarus developers. There's no strategy, direction or plan. Nothing wrong with that - this is a hobby for most people and such people work on what interests them and they do this in their spare time. This is perfectly understandable. Nothing is going to change that. That's what I'm doing after all.

So, to Webdrifter, do what interests you. Propose things if necessary and others may well join you. Just don't expect others to do it for you or nothing is likely to happen. I wish you luck.

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Blade on August 20, 2020, 08:03:00 am
So to get a better ranking we can try to chance their policy and/or adept to it.....

For instance, they split the pascal community up into 2 languages:
-  Delphi/Object Pascal
-  Pascal

This splitting up results in the fact that they both end up with a lower ranking!!!

The Object Pascal and Pascal wikipedia does show many of the dialects (including Free Pascal/Lazarus), so I'm not sure (sarcasm) how the TIOBE staff was blind to this, and decided the dialects were all separate languages (which makes no sense whatsoever).
 
I have e-mailed TIOBE too.  What I got out of it was a lot of ignorance from them about what a programming language is and contradictions.  It appears that their index is aligned to their business interests and bias within their company.  I got the impression that they will definitely put their thumb on the scale, regardless of what any data says, because TIOBE is also selling products that support specific programming languages.  If you look through what they sell, it becomes more obvious.  Unfortunately, the average person doing Google searches about language popularity has no idea that the TIOBE index is highly questionable and takes it as the gospel, so it's shaping and corrupting public opinion (which aligns with their products).

However, I do agree that Free Pascal/Lazarus should make it clear that it falls under the umbrella of Object Pascal and link to that Wiki page as well. 
At university I did not learn Pascal, but was previous knowledge. At university is teached C and Java, JavaScript in the last course I take this year.
So Pascal is not even being teached, there are paid courses that are 'learn free pascal'? Not books, but modern interactive courses with online tools.

This is flawed thinking, that we have to be careful about.  That is, what affects us personally or is going on in our area/country is therefore the reality for everyone and everywhere.  We shouldn't do that.  Many countries do more actively teach Pascal/Object Pascal and use Delphi.  Say what we will about how Embarcadero has managed the Delphi product, but they do push the Academic Edition of RAD Studio very hard.  There is also no limit of installations on the same network.  This year, Turkey's minister of education bought 1 million Delphi licenses for their students.  By contrast, I'm not quite sure why the Free Pascal and Lazarus foundation is not making a push or some type of outreach towards being an alternative or being used in schools.  Especially in very poor communities and countries.

Various countries around the world are still teaching Pascal/Object Pascal.  Russia has PascalABC (free open-source), which has been taught in their school system for over 18 years.  Check YouTube, and you will see many Russian videos on it (sadly they are nearly all in the Russian language only).  If they every got their act together about producing English manuals and English videos, they might be more popular.  South African schools purposely removed Java and use Delphi.  Pascal/Object Pascal or Delphi has a long history in the school systems of many countries such as: Brazil, Bosnia, Romania, Costa Rica, Croatia, France, Italy, Jamaica, Libya, Moldova, Serbia, Tunisia, Ukraine, Czech republic and Vietnam.

The issue is that in the USA, we have major global corporations backing their pet languages.  Oracle with Java, Microsoft with C#, Apple with Objective-C/Swift, etc...  So the market is crowded.  These huge companies will stoop to some very low tactics, like claiming every language that is not theirs is  "dead" or should not be learned.  This insane thinking seems to be pushed heavily into the minds of too many of their language users and advocates.

What might also be effective for Free Pascal/Lazarus is a new YouTube series from Lazarus experts about this flavor of Object Pascal.  Delphi based videos, by contrast, are everywhere on YouTube.  Even PascalABC has more videos out than Free Pascal/Lazarus, despite being lesser known.  The Free Pascal/Lazarus specific video series that I know of are SchoolFreeware and Devstructor, which are both good, but getting a bit old.

Links to a Free Pascal/Lazarus YouTube series can be pushed by the website and forum members on their personal websites and social media.  The SchoolFreeware and Devstructor series are nice, but from a branding perspective and website promotion, probably Free Pascal/Lazarus having their own video series (at least an introductory to intermediate one) would be more effective.  And of course the video series should make it clear that it's teaching Object Pascal and the IDE is Lazarus, to stop that bit of confusion.

Something else that I noticed is that existing free PDFs teaching Free Pascal/Lazarus appear to not be getting any push from this website or the Wikipedia.  Correct me if I'm wrong about that.  The Lazarus Handbook is excellent, very extensive, and thorough.  But, also pushing "getting your feet wet" type PDFs could help too with general public awareness. 

http://www.copperwood.com/pub/FreePascalFromSquareOne.pdf
(Free Pascal From Square One)

http://code.sd/startprog/StartProgUsingPascal.pdf
(Start Programming Using Object Pascal)

https://castle-engine.io/modern_pascal_introduction.pdf
(Modern Object Pascal Introduction for Programmers)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on August 20, 2020, 10:05:11 am
Something else that I noticed is that existing free PDFs teaching Free Pascal/Lazarus appear to not be getting any push from this website or the Wiki.  Correct me if I'm wrong about that.  The Lazarus Handbook is excellent, very extensive, and thorough.  But, also pushing "getting your feet wet" type PDFs could help too with general public awareness.

You mean like here (https://wiki.freepascal.org/Pascal_and_Lazarus_Books_and_Magazines)? Though that page might be promoted a bit better...
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Blade on August 20, 2020, 10:24:04 am
Something else that I noticed is that existing free PDFs teaching Free Pascal/Lazarus appear to not be getting any push from this website or the Wiki.  Correct me if I'm wrong about that.  The Lazarus Handbook is excellent, very extensive, and thorough.  But, also pushing "getting your feet wet" type PDFs could help too with general public awareness.

You mean like here (https://wiki.freepascal.org/Pascal_and_Lazarus_Books_and_Magazines)? Though that page might be promoted a bit better...

Excuse me for the confusion, I meant on the Free Pascal Wikipedia page (https://en.wikipedia.org/wiki/Free_Pascal (https://en.wikipedia.org/wiki/Free_Pascal)), which I think casuals are much more likely to see.  The Wiki of Free Pascal would likely be visited by active users of Free Pascal/Lazarus.  But besides that, maybe show the introductory PDFs on the Lazarus Forum index page.  I don't have the answers on the best ways for promotion, but do think something should be done to help promote awareness.  The narrative is being pushed that Pascal died with Turbo Pascal or in the 80s/90s, even from programmers of all people, which is both a bit ridiculous and sad.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 10:25:42 am
@Martin_fr

I agree with all you say. But in the end before all else it all comes down to bringing new blood to the community who have the motivation and time to spent in benefit off the community. The Python community flourishes because they where able to do that. This to my opinion is mostly because they tapped into the world of High-schools, Universities and innovators. The python support for Machine Learning gave this an extra boost. Since there lays their interest and a buck load of money in the market.

With the right strategy Pascal could act the same, not only towards machine learning but also towards other upcoming technologies. Off course for that we need more developers. 

See here the catch22...

I believe we can only break this vicious circle if we start creating more attention and desire, for pascal and its eco-system, on the internet.

With other words it's all a matter of marketing and salesmanship.

Like you I am not a salesman either. I'm like many programmers more of a slightly Asperger type of person. However through the years I have learned that promotion is part off every success and thus should not be ignored. It's unfortunately something we have to live with....

Since we also have short capacity off marketing capabilities, I was searching for methods to improve the situation with relatively easy methods with possibly the largest effect...

Improving the ranking seemed one off them.

Changing the Wikipedia pages could be a start.
I'm willing to put some effort in that, if the community give me the up-vote for that.

I had an elaborate e-mail dialog with Paul Jansen from Tiobe. Which is easy for me since I am dutch.
I believe he is an OK guy with good intentions.
He acknowledges the shortages of their method. and seems open for ideas.
Maybe we could help them improve their methods (and tools?).

However, for now we will just have to accept reality and go with how they work.
For now instance. We could maybe put meta tags and standard sentences containing ALL the right combination of words on EVERY webpage we create, starting with the freepascal and lazarus website and our forums and personal blogs.

The right words for us would be:

"DwScript programming"
"Object Pascal programming"
"Delphi programming"
"Delphi.NET programming"
"Pascal programming"

Don't no weather just something like "DwScript Object Delphi Delphi.NET Pascal programming" would also do the trick.

This could help boost the number of hits in search engines a little.

However do realize that nowadays "Python programming" gives 374.000.000 on google.com
Whereas "Pascal programming" gives 16.300.000 and "Delphi programming" gives 36.000.000, and "object pascal programming" gives 6.280.000. No idea how they take out the doubles.

But as you can see all is relative....
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Thaddy on August 20, 2020, 10:26:24 am
Never trust a wiki, not even ours.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 10:34:13 am
I have a lot of sympathy for the seemingly revolutionary views expressed here. If a simple tweak to wikipedia could effect significant change to the situation, why not?

But will it? Or might it not be so simple? If it is all that important, why doesn't the TIOBE guy maintain a few exceptions, and why did he change his algorithm wrt Pascal several times in the last 2 years?

Most likely it will just progress to the next artificial hurdle, and even if it wouldn't, it won't bring in the hordes, as it is only a minor additional decision maker.

Nice to have? Sure. But IMHO vastly overrated.  The same for SEO by name change, which are only temporary effects at best to an already known brand.  Maybe not perfect, but it might take years to get the new brands up in the search engine rankings.

As far as I understand the changes where not the results off a change in their methods but of the way how for instance Delphi was defined in the English Wikipedia.

I agree that changes have short and long term effect.

But I believe every effort in this direction is worth the while, as long as it holds in the long run without to much harm for the short term and doesn't cost to much effort compared to the gains.

It all starts however with a positive attitude towards the matter  ;)
 
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 10:47:36 am
Maybe somebody knows a SEO specialist who is willing to look into the matter? And give us some tips...
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 11:09:45 am
Maybe we should give the marketing of Pascal a more important position within our community?

Starting by giving this topic it's own entrance prominently positioned on the main page of our forum......

And creating a team that actively pursues this mission. Just in the way we have a development team.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 11:15:52 am
Maybe we can split long pages on our forum and our wiki's into two or more pages?

Thus with little extra effort creating much more pages, and thus hits on search engines.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 11:17:17 am
Anybody willing to brainstorm with me.... I dare you!
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 11:20:58 am
May we can make our forum and wiki's more friendly for use with mobile devices? Since youngsters seem to work with these more and more! Thus creating more traffic and attention. And thus in the end maybe more influx of young people?
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 11:29:20 am
Maybe some of the more knowledgeable people of our community could regularly write some articles on major forums where young programmers tend to go. On where Pascal and it's eco-system excels or could hand them tools that lack in other languages.

For instance I have always found that although Pascal does not excel on many field it tends to belong to the top contestors in many fields. Which in overview should maybe even label it as the best general programming language.

And I believe it's capability of supporting a great many of platforms is a huge salespitch for anybody interested in hypes like "the internet of things".
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 11:45:34 am
I know my children hardly search in forums and wikis, if at all.

When they want to know something they search on Youtube first.

I can't even find our own Free pascal and Lazarus channels on there!

Even Delphi shines by absence.

So maybe we should be much more prominently active on those platforms......

(Together with links towards our promoting articles, forums and wiki's off course)

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 20, 2020, 12:32:50 pm
If we rename, the new name must be "Bamboo"
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: dbannon on August 20, 2020, 12:37:18 pm
If we rename, the new name must be "Bamboo"

OK, I'll bite, why bamboo ?

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 20, 2020, 12:47:28 pm
If we rename, the new name must be "Bamboo"

OK, I'll bite, why bamboo ?

Delphi once planned to rename its products, somewhere in the D2009-DXE period. One of the suggested alternatives in the survey was Bamboo.

I assume the idea was something like grows quickly and still a quite strong building material.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 12:50:31 pm
If we rename, the new name must be "Bamboo"

OK, I'll bite, why bamboo ?

Delphi once planned to rename its products, somewhere in the D2009-DXE period. One of the suggested alternatives in the survey was Bamboo.

I assume the idea was something like grows quickly and still a quite strong building material.

Nice name, but will it help us up in the rankings......
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: nospambox on August 20, 2020, 12:56:21 pm
I have been programmer in Delphi / FPC since 1995. Private and corporate usage.
it is by far a very convenient eco system to make efficient programming, both at design but also at execution stages.

But why Delphi/FPC are so low in the ranking ? It is not "only" advertizing or the way the statistics are built. It is a problem of interoperability with some frameworks that are today a "must have" for a programmer.

Most libs, today, have moved from a former C implementation to a C++ implementation. In the old days, it was still acheivable to wrap the C library so it could be used with FPC/Delphi. It required some efforts, but community was taking care of that as well as some automation tools to convert H to PAS.

But with the move to C++, it is now nearly impossible to use the modern libs. Down the road, too many efforts to get it working (for wrapper developers) and very hard to use (for programmer).

So, all languages being able to inter operate with these frameworks are "winners". Of course, you have C++ native programming (not very friendly), ... but many others options (Python, ...). If you want neural network programming (Caffe, tensorflow), you need C++. You want vision system (OpenCV), you need C++. You want some "drivers" to operate prioprietary cards (like IO cards), you need C++.
Look at the top "libraries" that exist, what ever the purpose. In the most case, no option to run them from Delphi/FPC.

Only way in Pascal to use these is to incorporate a P4D layer and get a python layer programming inside Delphi/FPC. Crazy and a lot of limitations.

Down the road, if we want to increase ranking in reviews, we need to increase functionnalities so programmers get best of the world programming environment : ease at design time, error safe coding, fast to execute, muti tageting with cross compilation.... and able to connect to anything alreay existing. FPC/Delphi, covers very well the first points - probably best in class - but last point is becoming every day more and more a handicap that will kill this framework.

And I am very sad writting this.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 01:11:38 pm
@nospambox - Seems like good points you make. Would like to hear the opinion about this from the Pascal die-hard's of this forum. In the meantime if nobody picks up this challenge all that rests us is trying to pull in some new blood by better marketing.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 20, 2020, 01:15:16 pm
Nospambox:  I think the reasons are less technical, but not having a major company as backing (and no Embarcadero doesn't count).

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 02:02:49 pm
Nospambox:  I think the reasons are less technical, but not having a major company as backing (and no Embarcadero doesn't count).

Agree, but unless somebody knows how to land a company like that for the Pascal community, I guess that road is not worth our energy.

But maybe you have other suggestions that might cut it?
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 20, 2020, 02:14:18 pm
Nospambox:  I think the reasons are less technical, but not having a major company as backing (and no Embarcadero doesn't count).

Agree, but unless somebody knows how to land a company like that for the Pascal community, I guess that road is not worth our energy.

But maybe you have other suggestions that might cut it?

No, I think that World Dominations are the unicorn option, and choose to work mostly for actual users, not for potential users.

I don't mind doing a bit of publicity, but that should be about promoting what there is, not the shortest route to changing it.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Bi0T1N on August 20, 2020, 02:18:44 pm
There's the largely ineffective (imnsho) Free Pascal and Lazarus Foundation.
That's true. If you create a foundation you should a least maintain the website and your projects but that's obviously not the case (e.g. Attributes are implemented in mid 2019 (https://lists.freepascal.org/pipermail/fpc-announce/2019-July/000612.html) or just look at the awkward RoadMap (https://foundation.freepascal.org/projects/roadmap) page). Each project page should also highlight the advantage(s) of having this feature and thus might increases the willingness of donations.

May we can make our forum and wiki's more friendly for use with mobile devices?
That's indeed needed. For the wiki I've already mentioned this here (https://forum.lazarus.freepascal.org/index.php/topic,48831.msg356956.html#msg356956). Adding the wiki translation plug-in would also improve the overall user experience but seems the admin(s) don't care.
However, for the forum you've to wait for the final release of SMF 2.1 as this introduces (https://www.simplemachines.org/community/index.php?topic=530233.0) a responsive design but as you can see it's still at RC2 (https://www.simplemachines.org/community/index.php?topic=566669.0) after almost 6 years. :-X

My recommendations for the above points:
The people behind the Foundation should think about moving to Github Pages so everyone could suggest changes by simple pull requests. The same applies to the development of FPC/Lazarus even if I know that some people don't like Github but it's the #1 like Google is for searching the web. If you want to attract new people go to the place where they are. ;)
I mean, sure you can host your own Gitee/Gitlab instance but if the amount of volunteers is limited and the few don't even keep pace with updating/installing asked plug-ins or features why running another software?
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Handoko on August 20, 2020, 02:40:14 pm
Maybe somebody knows a SEO specialist who is willing to look into the matter? And give us some tips...

I'm not a specialist but because I build websites for my clients, I ever spent some time learning SEO.

These are some mayor things that affect a website's SEO score:
- The domain name contains the targeted keywords
- The URL contains the targeted keywords
- Backlink from .edu, .org, .gov domains
- Do follow link from high rank and traffic sites (cnn.com, reddit.com, etc)
- High quality and unique contents
- Update regularly
- Not link to bad neighborhoods (p*rn, s*x, drug, etc)

Nowadays with the fast growing of social media, search engines also put some weight on the social media mentions.

I personally think Free Pascal does a good job in SEO. At least this forum is the first item if you search using the keyword "pascal forum" using DuckDuckGo. I don't think renaming Free Pascal is good. It may sounds nice in human ears but it will ruin the previous built SEO. It can regain the score but it needs time and lots of efforts.

I personally think if we want to improve the SEO we should put more efforts on social media like Facebook, YouTube, Twitter. Also update and improve the wiki. The wiki has too many outdated information, not good for for both human and search engines.

We could maybe put meta tags and standard sentences containing ALL the right combination of words on EVERY webpage we create, starting with the freepascal and lazarus website and our forums and personal blogs.

Meta tags are often abused and Google no longer values meta tags.
https://support.seomanager.com/article/30-keywords-meta-field-is-no-longer-used-by-google

May we can make our forum and wiki's more friendly for use with mobile devices?

When I'm not sitting in front of my computer, I often visit this forum using my mobile phone. It does not look good on my phone.

When they want to know something the search on Youtube first.

Yes YouTube. Youngsters like YouTube, not my favorite but I also spend some time watching it.

I am glad to see some nice users post their tutorials on YouTube but I wish to see more:
https://wiki.freepascal.org/Lazarus_videos

If we rename, the new name must be "Bamboo"

No, don't use the name. I own a Bamboo, it is a brand name of a discontinued model of Wacom tablet.
https://en.wikipedia.org/wiki/Wacom#Discontinued

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 03:19:44 pm
I don't see any point in discussing names that destroy our rating or the mental connection with "Pascal".

It should help us up in the rating and strengthen the mental connection with "Pascal". 

Otherwise we might just as well start all over again with a new language.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 04:15:25 pm
@nospambox - I checked the future plans page: https://www.freepascal.org/future.html

And ques what? Linking with C++ code is right up there!

So maybe you can help them out with that one.

And thus help solve your own problems with Free Pascal / Lazarus).
And maybe learn some really neat stuff in the process.

I'm sure  the development team will be very glad with any really useful help they can get.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Webdrifter on August 20, 2020, 04:57:19 pm
However, for the forum you've to wait for the final release of SMF 2.1 as this introduces (https://www.simplemachines.org/community/index.php?topic=530233.0) a responsive design but as you can see it's still at RC2 (https://www.simplemachines.org/community/index.php?topic=566669.0) after almost 6 years. :-X

When I look for that topic on RC2 of SMF 2.1, I find: "The term "Release Candidate" means that the developers believe this version of the software is reliable enough to be installed and used on production websites. "

There are probably good reasons why we still don't use it. But maybe the administrators can explain to me why?

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: nospambox on August 20, 2020, 05:36:47 pm
@nospambox - I checked the future plans page: https://www.freepascal.org/future.html

And ques what? Linking with C++ code is right up there!

So maybe you can help them out with that one.

And thus help solve your own problems with Free Pascal / Lazarus).
And maybe learn some really neat stuff in the process.

I'm sure  the development team will be very glad with any really useful help they can get.

I know... But it is roadmapped for quite a while... And I searched the status in the dev branch. It seems not even started.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Blade on August 20, 2020, 08:42:07 pm
May we can make our forum and wiki's more friendly for use with mobile devices?

When I'm not sitting in front of my computer, I often visit this forum using my mobile phone. It does not look good on my phone.

When they want to know something the search on Youtube first.

Yes YouTube. Youngsters like YouTube, not my favorite but I also spend some time watching it.

I am glad to see some nice users post their tutorials on YouTube but I wish to see more:
https://wiki.freepascal.org/Lazarus_videos

I agree that for a short-term goal, simple promotional actions are good.  New Free Pascal/Lazarus focused YouTube training videos, update the Free Pascal Wikipedia with more training material, a website that looks good on mobile phones (though that might take some work), and more public advocacy by forum members.  These seem doable.

Part of the battle is public awareness.  We shouldn't have people that even though they are in the programming community, when someone says Pascal, they think it disappeared in the 90s.  "Oh yeah I used Turbo Pascal way back when, but isn't that language dead?"  "Ah no, go check out Free Pascal/Lazarus."
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: VTwin on August 21, 2020, 12:29:59 am
"REALbasic" seems to have a problem with their association with BASIC. First they tried renaming it "Real Studio", then "Xojo".

Maybe because of Dijkstra's famous words: "It is practically impossible to teach good programming to students that have had a prior exposure to BASIC: as potential programmers they are mentally mutilated beyond hope of regeneration."

Wikipedia: "Xojo has been listed several times among the top 150 programming languages as published by TIOBE".  :D

I like "Bamboo", but how about "Blaise" or "Blaze"? I'm not really advocating anything, I'm happily using "Lazarus" because it works.

Cheers,
VTwin
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: winni on August 21, 2020, 12:53:34 am
Hi!

They should have called  Xojo the "UnRealBasic".

"Lazarus - The whitest whites ever!"

How about that? Some Dada adds?
Brings more attention than "We do this and that".

Winni

PS.: "Lazarus - 1.999 $ cheaper than Xojo Pro Plus!"
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: TRon on August 21, 2020, 12:58:16 am
I like "Bamboo", ..
Every time I read that word my mind reads bamboozled instead  :)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: VTwin on August 21, 2020, 01:01:48 am
Hi!

They should have called  Xojo the "UnRealBasic".

"Lazarus - The whitest whites ever!"

How about that? Some Dada adds?
Brings more attention than "We do this and that".

Winni

PS.: "Lazarus - 1.999 $ cheaper than Xojo Pro Plus!"

:D I fell into the REALbasic trap when it was a good cross-platform alternative. Extracting myself from that mess was arduous.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PierceNg on August 21, 2020, 04:05:31 am
I have been programmer in Delphi / FPC since 1995. Private and corporate usage.

I remember the Windows shareware scene - FTP sites like cica.cica, aggregators like download.com, payment processors like ... can't remember their names lol. IMHO, a major reason for the vibrancy was that there was money to be made, not just by selling applications, but also by selling components, libraries, tools and such to the programmers building those applications in Delphi, VB, etc.

These days the world is much different. Try-before-you-buy still exists but now they are SaaS and APIs. And the heavily hyped languages especially those with corporate backing already come "batteries included" and have hordes of programmers, paid or not, producing yet more open source. People expect their programming language ecosystem to be free as in free beer.

Certainly Free Pascal and Lazarus are also batteries included, thanks to the developers who generously made and continue making those batteries. My own interest is in server-side web programming and Free Pascal has enough libraries to get going. Indeed as a newcomer the choices can be bewildering.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Blade on August 21, 2020, 05:44:07 am
SchoolFreeware might be updating their Free Pascal/Lazarus video series.  Not sure if the video was a one off for Windows 10 or are they planning more.  But it's good they are giving a shoutout.
https://youtu.be/05S7z8bmOjY (https://youtu.be/05S7z8bmOjY)

Aiming Free Pascal/Lazarus at schools seems like a winning strategy, as the language has a number of inherent advantages such as being easier to understand, for making GUIs, and being easy to get a hold of without any strings attached (like e-mail addresses or yearly licenses).

Something interesting about Pascal/Object Pascal videos on YouTube is a lot of them are non-English.  This might be indicative of the programming language's relatively greater use/popularity in other countries (or at least their school system).

A mention should also be made for Swinburne University's Object Pascal YouTube series
CodeCasts - Introduction to Programming in Pascal (https://www.youtube.com/watch?v=8maOxNc0Aio&list=PLdVESrjTNUXtU8zclRh9ovhstzWQAY05USwinburne)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on August 21, 2020, 09:54:45 am
The same applies to the development of FPC/Lazarus even if I know that some people don't like Github but it's the #1 like Google is for searching the web. If you want to attract new people go to the place where they are. ;)
I mean, sure you can host your own Gitee/Gitlab instance but if the amount of volunteers is limited and the few don't even keep pace with updating/installing asked plug-ins or features why running another software?

It was already decided that if we (FPC) move to Git that we'll use our own (https://gitlab.freepascal.org/fpc) GitLab instance with probably a GitHub mirror (https://github.com/fpc/). We are not putting our main repository into the sole hands of a cloud provider. This has the added benefit that we can have an internal repository that includes the full FPC history including those commits that included copyrighted code.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: trev on August 21, 2020, 10:28:31 am
A mention should also be made for Swinburne University's Object Pascal YouTube series
CodeCasts - Introduction to Programming in Pascal (https://www.youtube.com/watch?v=8maOxNc0Aio&list=PLdVESrjTNUXtU8zclRh9ovhstzWQAY05USwinburne)

Better links the Swinburne University of Technology videos for the whole series (including code attachments) is the listing I made on the Wiki: Free Pascal Video Tutorials (https://wiki.freepascal.org/Free_Pascal_videos#Swinburne_University_of_Technology_Free_Pascal_Tutorials).
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Blade on August 21, 2020, 12:04:22 pm
A mention should also be made for Swinburne University's Object Pascal YouTube series
CodeCasts - Introduction to Programming in Pascal (https://www.youtube.com/watch?v=8maOxNc0Aio&list=PLdVESrjTNUXtU8zclRh9ovhstzWQAY05USwinburne)

Better links the Swinburne University of Technology videos for the whole series (including code attachments) is the listing I made on the Wiki: Free Pascal Video Tutorials (https://wiki.freepascal.org/Free_Pascal_videos#Swinburne_University_of_Technology_Free_Pascal_Tutorials).

Yes, they are there.  I was more going on about English videos among the non-English ones out there.  Something to note, as with the SchoolFreeware and Devstructor series, it's been some years since these video series were created.  They all are quite good, so quality isn't the issue.  Would be nice to figure away to get more visibility for such free learning from the Object Pascal Wikipedia and Free Pascal Wikipedia.

When you look up Pascal, Object Pascal, Lazarus...  A lot of the new videos and series are non-English.  Take these for example:
Introdução ao Lazarus e Pascal (https://www.youtube.com/watch?v=tcjOMaU5k7c&list=PLOQqk9kh6vgfjHjzPvN9trixiMNOcsrag).  Brazilian Portuguese...  Which is fantastic. 
Curso de programación en Lazarus Project Free Pascal (https://www.youtube.com/watch?v=6brafIhXpUY&list=PL5t8pBBHgSOtkBPKe1aDrHMlAZ_zrajBd).  Another good looking one in Spanish.

Maybe finding a way to list these video series in different languages would be nice, if they aren't already.  And just saying some newer English ones would be great too.

Ultimately, it would be good to keep a strong push towards schools.  Without a huge global corporation pushing Object Pascal as its pet language, one of the other routes would be schools and academia, to help with popularity. 
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 21, 2020, 02:26:28 pm
The people behind the Foundation should think about moving to Github Pages so everyone could suggest changes by simple pull requests. The same applies to the development of FPC/Lazarus even if I know that some people don't like Github but it's the #1 like Google is for searching the web. If you want to attract new people go to the place where they are. ;)
I mean, sure you can host your own Gitee/Gitlab instance but if the amount of volunteers is limited and the few don't even keep pace with updating/installing asked plug-ins or features why running another software?

Presence on GitHub is certainly a marketing argument. In which way it will manifest itself, is however a good question....

In any case, despite the critical questions that follow below, creating a (yet another) github mirror, may help a little. So if it can be done with little effort, then it definitely should be done.

Keep in mind, it has been looked into already. And there are steps (e.g. developers have scripts for certain tasks) in the development/release process that need heavy work to be change.
So changing the main repro is a big task (bigger than you would think)



* More Google hits?
Well there is an FPC mirror already: https://github.com/graemeg/freepascal
And 2 Lazarus mirrors. So google should have the hits for those already.



*New users to trust the project more easily?
I have a feeling that may be true for some people, despite it is absolutely illogical.



* Pull request easier for the developer than patches?
Well not for me.
I already work with git, on my local pc, and applying a patch and comparing it in git is as easy as looking at a pull request.
I should say that for me, the online interface to compare pullrequests is no good (So my feedback comes by mail, no annotations via the online interface). I always pull the requests to my local repro, and compare them locally.
So patch/pull requests, is the exact same amount of work to me.



* More contributors?
I am not that sure... Yes at first glance I would think so too.
But people have reasoned for so many things (that they wanted to be done), with this and similar arguments. And not once has it happens. We had been promised huge amount of new users if we would only change the version to 1.0. We are at 2.0 now. Still waiting for the spike in new users (there have been new users, and maybe the rate has increased over time, but I did not see any increase correlated to the 1.0 date....

Keep in mind that writing the code, that you want to contribute is a lot more work and needs more knowledge and skills, than submitting a patch (or even submitting an entire copy of the modified file (which people have done, sigh))

And having the option of pull requests, does not change many other issues:
- You need to check first, if the feature/fix itself is acceptable (submitting codetools for whitespacehttps://en.wikipedia.org/wiki/Whitespace_(programming_language) (https://en.wikipedia.org/wiki/Whitespace_(programming_language)), may be rejected, even if your code has the finest quality)
- Dealing with the forth an back, until the submission has the format/quality required
- Dealing with delays until someone has time to answer.

The "delay" issue, is unfortunately a big problem.
Even if you are lucky and a developer picks up your issue and takes care of it, that developer could take weeks to answer. I have been at both ends (being the developer and being the contributor, the latter with various projects).

But worse, because the projects have a flat team structure, there is no-one who will "force assign" patches to a team member. (the team members are all volunteers, they only do work that they volunteer to do). So sometimes a patch comes along, and everyone thinks: This does not fall into the area of the project which I am working on.
Then the patch may be unattended for very very long.
That is a massive and really bad problem. No patch should ever be left without a reply. But it does happen. And GitHub is not going to change it.

But we can actually test it.
I run a github mirror myself: https://github.com/User4martin/lazarus
On this mirror, it is possible to make pull requests. However, the only person looking at those is me. So pull request for anything that does not fall into my area will be closed without even a review.
Generally, if you have a pull request for SynEdit or the Debugger, then you can make it on this mirror. (You should check that the feature itself is acceptable, and that you are looking at the correct place to implement it... Those pre-conditions also apply).
So far I had one. Which started with patches, and only became pull-requests when I pointed to the possibility. So the contributor existed without need for github.




Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 21, 2020, 05:18:28 pm
Quote
And GitHub is not going to change it.

GitHub-site is not open source.

If you want your own git server, with a nice web-site, you should use GitLab-code that is free and open source.

Fre;D
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 21, 2020, 05:49:43 pm
Quote
And GitHub is not going to change it.
GitHub-site is not open source.
Not the point (the above is quoted from my post). Also PascalDragon pointed out some plan.....

GitHub was only mentioned (by me) as mirror (and at some time "official mirror"). In that capacity, it is the same as Sourceforge, only at this moment in time a bit more popular.

A mirror on github may add (some amount, which could be some small, or not so small amount - only the future might show) of popularity.
As such it is a small piece of the puzzle. (Once it becomes possible within reasonable amount of work it should be added... Whenever that is...)

As for the big parts of the puzzle. I do not have the answer.


Just to make my self very clear:

I do not object to (many of) the small pieces in the puzzle.
I only wish to keep expectations within what I see reasonable (And well, if the future proves that I was wrong in some point... then I still was entitled to my opinion)

And point out, that some things my look trivial, but actually require a lot of work. Therefore, even if they are (or are believed to be) desirable => they may not happen as soon as hoped for. (Again simply help keeping expectations, in no way saying they should not happen, or should not be discussed...)


Point is creating publicity is a lot of work.

Also, as I pointed out, the existing team is dead busy. If they were not, then they had already done something....

So anyone who wants to have the situation improved, should start with those possibilities were they can create publicity without needing the team to do something. (The team cares, but again limited workload).

There are plenty of ways to help, with no need to take time of the team.
Prove yourself by acting on those ways.

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 21, 2020, 06:00:22 pm
Prove yourself by acting on those ways.

Maybe create a GitLab mirror of Graeme's Lazarus GitHub site.

It takes only 2 minutes (but you must be the administrator of the mirrored site, so only Graeme could do it).

Imho, it is good to have your project on both GitLab + GitHub, this to profit of the visibility of each.

Fre;D
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: howardpc on August 21, 2020, 06:03:25 pm
Maybe some of the more knowledgeable people of our community could regularly write some articles on major forums where young programmers tend to go.
...
May we can make our forum and wiki's more friendly for use with mobile devices?
The trouble with  "maybes" and all such good ideas is not the discussion of them, but the good and consistent implementation of them in the months and years ahead by "people" (or "we") who have the time and motivation. Often the most effective publicists tend to be rather annoying people. I think most forum users prefer to code rather than write blogs or documentation.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 21, 2020, 06:27:38 pm
Quote
Maybe create a GitLab mirror of Graeme's Lazarus GitHub site.

@ Martin, ooops, you did create your own GitHub Lazarus repo.

So, here what to do when you find 2 minutes of free time:

1) Create you own GitLab account.
   https://gitlab.com/users/sign_up

2) When done, go there and sign with the button "Sign with GitHub":
    https://gitlab.com/users/sign_in

3) Go on "My projects" and click on "New Project".

4) Click on "Import Project" then on "Import Github project".

5) Choose the projects you want to import.

6) When the project is imported, go to that GitLab project
    and select "Settings-Repository-Mirroring repositories" + "Expand".

7) Write the GitHub repository URL, choose "Pull" for Mirror direction.

Yep, all done, and your GitLab mirror will be automatically refreshed after each change done on your GitHub repo.

Fre;D

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 21, 2020, 06:28:36 pm
With regards to various posts in this thread concerning the look of the forum/wiki/webpage on mobile devices.....

The wiki and the forum are based on 3rd party software. I do not know what is available for each of them.
On top of that, I know that the server admins for the involved servers are busy people. Upgrading/Changing those installations takes serious time. It is not going to happen several times in short succession, just to test different scenarios.

The website is custom made. If it can be made more mobile friendly (keeping the existing design / not completely changing the desktop version), => patches are welcome. (Complete design changes need approval of the entire team. A lengthy process, and the need to convince plenty of people. Also too big changes to the design would mean loosing brand association, and therefore actually reducing popularity for at least some time.
https://svn.freepascal.org/cgi-bin/viewvc.cgi/trunk/?root=lazaruswebsite
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 21, 2020, 06:58:05 pm
The people behind the Foundation should think about moving to Github Pages so everyone could suggest changes by simple pull requests. The same applies to the development of FPC/Lazarus even if I know that some people don't like Github but it's the #1 like Google is for searching the web. If you want to attract new people go to the place where they are. ;)
I mean, sure you can host your own Gitee/Gitlab instance but if the amount of volunteers is limited and the few don't even keep pace with updating/installing asked plug-ins or features why running another software?

Presence on GitHub is certainly a marketing argument. In which way it will manifest itself, is however a good question....

A mirror would be nice publicity, but any US service can never be the primary mirror, since the US government constantly willy-nilly bans countries from any interaction with US companies.

Note that the often advocated  "easy" pull request concept usually only works with minor projects that are easily overviewable and with a very eager receiver.

However for non-trivial projects as FPC and Lazarus standards wrt submission of patches will be kept, and many patches are revised several times due to requirements of FPC/Lazarus devels which doesn't fit the pull request model.




Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 21, 2020, 07:00:49 pm
@ Martin, ooops, you did create your own GitHub Lazarus repo.

So, here what to do when you find 2 minutes of free time:

I did not stop the time. But with the waiting for the clone to finish, I am sure it took more than 2 minutes.

https://gitlab.com/martin_frb/lazarus

I do not know how often GitLab pulls.... (the github gets updated every few minutes, IIRC)


The gitlab currently has merge-request disabled (given that they only are for a limited set of the codebase, that should not be to big a limit).

Otherwise I would need to find, how I can manage them in my local git. (I.e. the github pull-requests are downloaded as branches)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 21, 2020, 07:13:19 pm
I did not stop the time. But with the waiting for the clone to finish, I am sure it took more than 2 minutes.
https://gitlab.com/martin_frb/lazarus

Yep, well done!

I do not know how often GitLab pulls.... (the github gets updated every few minutes, IIRC)
The gitlab currently has merge-request disabled (given that they only are for a limited set of the codebase, that should not be to big a limit).

Did you try 6) and 7) of my previous post?

For me it works perfectly the pulls from GitHub are automatic.
But I still use GitHub as "master" and dont do anything to GitLab, all is done automatically.

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 21, 2020, 07:15:59 pm
A mirror would be nice publicity, but any US service can never be the primary mirror, since the US government constantly willy-nilly bans countries from any interaction with US companies.
+1

Quote
Note that the often advocated  "easy" pull request concept usually only works with minor projects that are easily overviewable and with a very eager receiver.

However for non-trivial projects as FPC and Lazarus standards wrt submission of patches will be kept, and many patches are revised several times due to requirements of FPC/Lazarus devels which doesn't fit the pull request model.

The "online tools for merging" are no good for me (and I guess you and other in both teams)

Github (and afaik gitlab too) allow to pull them as a branch (you do get *all* pull requests / but you do get all other branches too). => That is a work able feature.
The branch is not much different from a patch. You can:
- compare to current checkout
- checkout and test
- rebase (a local copy) of it, to the latest master or fixes branch.
- and anything else....
- In case of changes/iterations the contributor adds more commits to the branch. (or even rebases it)

A pull request has one advantage:
It can have many commits.
That is, if there are several changes (like first refactor, then add the feature), you would need 2 separate patches. Ore one pull request with 2 commits.

As I said, I handle it all on my local machines, with access to all my favourite tools. Reviewing it in the IDE, running and testing it....

And that is why I wrote earlier: I do not use the online annotation system. (Which allows to attach notes to individual lines of code).
Those annotations only work in the web-browser. But I never open the pull request in the browser. (Except once applied, in order to close it (not sure if github realizes that it was merged, maybe it does).

---
- For the user that means, they can do a pull-request on the mirror.
- I can process it offline, and when I am happy directly commit (and push) it to the primary git server.
- Afaik, I can commit changes to the pull request, and the user can see them, and then act on them (eg comments)
  (But I need to test, that they can be pushed)
- If I give feedback, I will do so by mail. Or on the associated mantis report.



Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 21, 2020, 07:17:38 pm
Did you try 6) and 7) of my previous post?
Yes, I added that.

Ah, in the meantime, it has pulled again (3 minutes ago)

--------------
Btm, It works without needing access to my github. And no password either.

So anyone can create a clone like that.

Access to github was only needed, to automatically get the description and that.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 21, 2020, 07:27:43 pm
Btm, It works without needing access to my github. And no password either.

Yes, and this scares me a little, so GitLab inc knows now my GitHub password...

So anyone can create a clone like that.

Access to github was only needed, to automatically get the description and that.

Yes, but he would have the possibility to do automatic pulls only if he is the administrator of the mirrored GitHub repo.

Fre;D
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 21, 2020, 07:33:31 pm
off topic

Yes, and this scares me a little, so GitLab inc knows now my GitHub password...
I don't think it does.

The authorize gitlab process, is done in github.

At no point did I enter my github password in gitlab.  (and github should only have a hash of it, so it should not even be able to reveal it).
Gitlab however can then access all my repros (public and hidden), get my email, and all that..... :(

Once I revoke the permission (can be done on github), gitlab has no longer access.

Quote
So anyone can create a clone like that.

Access to github was only needed, to automatically get the description and that.

Yes, but he would have the possibility to do automatic pulls only if he is the administrator of the mirrored GitHub repo.

I added the pull, after revoking permissions.

Gitlab pulls in the very same way, like any person would do on their local PC.
After all, the repro is public.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Bi0T1N on August 21, 2020, 08:00:29 pm
The same applies to the development of FPC/Lazarus even if I know that some people don't like Github but it's the #1 like Google is for searching the web. If you want to attract new people go to the place where they are. ;)
I mean, sure you can host your own Gitee/Gitlab instance but if the amount of volunteers is limited and the few don't even keep pace with updating/installing asked plug-ins or features why running another software?

It was already decided that if we (FPC) move to Git that we'll use our own (https://gitlab.freepascal.org/fpc) GitLab instance with probably a GitHub mirror (https://github.com/fpc/).
I see but it's exactly what I wrote - your Gitlab instance is kinda outdated. Even if you don't show the version (https://gitlab.freepascal.org/help) (see here (https://about.gitlab.com/blog/2015/05/07/version-check/)) one can find out that the instance uses JavaScript Cookie v2.1.3 while the official Gitlab (https://gitlab.com/help) uses JavaScript Cookie v2.2.1 (check main.*.chunk.js).
This change (https://gitlab.com/gitlab-org/gitlab/-/commit/6c798506230b3ed251f9f9e708950b326674f4df) was made in May and went into Gitlab 12.9 and thus you run a version prior that. The recent (still supported) versions are 13.0, 13.1 and 13.2 which get important security fixes (https://about.gitlab.com/releases/categories/releases/) quite often. I could keep on with the Mantis Bugtracker 2.24.0 or the MediaWiki 1.3.7 but I think I've already made my point and showed that you should rethink the idea with hosting your own software - or maintain and update them ASAP (hours not weeks).
I also don't believe it'll be a good publicity if the server/software gets hacked and all the user data are exposed.

We are not putting our main repository into the sole hands of a cloud provider. This has the added benefit that we can have an internal repository that includes the full FPC history including those commits that included copyrighted code.
You still can mirror it to "your safe place" and everyone also has its local copy (at least if you checked out the repository via git clone once).

The people behind the Foundation should think about moving to Github Pages so everyone could suggest changes by simple pull requests. The same applies to the development of FPC/Lazarus even if I know that some people don't like Github but it's the #1 like Google is for searching the web. If you want to attract new people go to the place where they are. ;)
I mean, sure you can host your own Gitee/Gitlab instance but if the amount of volunteers is limited and the few don't even keep pace with updating/installing asked plug-ins or features why running another software?

Presence on GitHub is certainly a marketing argument. In which way it will manifest itself, is however a good question....

In any case, despite the critical questions that follow below, creating a (yet another) github mirror, may help a little. So if it can be done with little effort, then it definitely should be done.

Keep in mind, it has been looked into already. And there are steps (e.g. developers have scripts for certain tasks) in the development/release process that need heavy work to be change.
So changing the main repro is a big task (bigger than you would think)
You should probably rewrite all these "old" scripts as the design is bad if it needs svn and cannot run on other directory structures easily. Nowadays your scripts run automatically if triggered by commit/user action on your Github/Gitlab/Gitee instance. Therefore the release process happens almost automatically and needs only some checks to verify that everything went well. Sure, it might need some time to rewrite your scripts but the benefits of it are totally worth it.


* Pull request easier for the developer than patches?
Well not for me.
I already work with git, on my local pc, and applying a patch and comparing it in git is as easy as looking at a pull request.
I should say that for me, the online interface to compare pullrequests is no good (So my feedback comes by mail, no annotations via the online interface). I always pull the requests to my local repro, and compare them locally.
So patch/pull requests, is the exact same amount of work to me.
It depends on your view and workflow but in general it's easier to accept small patches like this (https://bugs.freepascal.org/view.php?id=37049), this (https://bugs.freepascal.org/view.php?id=37616) or that one (https://bugs.freepascal.org/view.php?id=36949). There is not really a need to check it out locally as the review process can happen directly through your browser (additionally with CI you and the contributor see directly when creating the PR if it breaks something).
With hub (https://hub.github.com/) a contributor don't need to leave the console at all to fork and create a pull request. The reviewer can also comment on specific lines and give suggestions, advices or hints.
In my opinion these things are big advantages for both - reviewer and contributor - compared to the current workflow with issues/patches.

* More contributors?
I agree that moving to e.g. Github won't bring 100 new developers but it will bring some publicity (blogs could write about the move) and that might lead to a few people who try it out and might stick to it. And after some time they might start to add their "missing feature".

But we can actually test it.
I run a github mirror myself: https://github.com/User4martin/lazarus
I don't think its a good idea to have several places where one could submit pull requests. It should be ONE official place used by everyone.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 21, 2020, 08:11:24 pm
off topic
Yes, and this scares me a little, so GitLab inc knows now my GitHub password...
I don't think it does.
The authorize gitlab process, is done in github.

Ooops, so in my previous post:
Quote
7) Write the GitHub repository URL, choose "Pull" for Mirror direction and write your GitHub password.
"write your GitHub password" is totally wrong, I will directly change my GitHub password and will remove that from my post.

off topic
Gitlab pulls in the very same way, like any person would do on their local PC.
After all, the repro is public.
Yes, and after it can be used for "marketing" using all the features that GitLab inc offers to promote the project.

Fre;D
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: VTwin on August 22, 2020, 01:39:34 am
Again, I am not advocating. I think it is great to have multiple mirrors on GitHub/GitLab/SourceForge, but if the goal is to get new users aren't they just going to find whatever pops up by Googling Free Pascal or Lazarus?
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 22, 2020, 01:52:54 am
Again, I am not advocating. I think it is great to have multiple mirrors on GitHub/GitLab/SourceForge, but if the goal is to get new users aren't they just going to find whatever pops up by Googling Free Pascal or Lazarus?

I think that the goal is to get new users using the "marketing" feature, discussion, tags of each GitHub and GitLab power.

By the way, there was some advice about the country of the server that should not be USA, so GitHub and GiLab are not welcome.

But, afaik, SourceForge has also his servers in USA, so to be logical, fpc should not have his first host server in SourceForge too.
Title: Re: The Ranking of Free Pascal in the Tiobe-index
Post by: Kays on August 22, 2020, 03:44:28 am
I would not care too much about representation on many repository sites, but being represented where “our” potential users are most likely looking.

If you are looking for more developers (who don’t need or want to use the software), then syncing with a couple popular repository hosts could make sense, I don’t know, but I’m primarily concerned about finding, attracting new users using our great compiler that compiles the best programming language :D.

I think, ten years ago I discovered fpc by a simple query
Code: Bash  [Select][+][-]
  1. apt-cache search pascal compiler
Therefore I think it’s important to support and maintain packages for all major OS. Currently, Ubuntu (https://changelogs.ubuntu.com/changelogs/pool/universe/f/fpc/fpc_3.0.4+dfsg-23/changelog)/Debian, FreeBSD, and OpenBSD all list 3.0.4 as the latest version. That version’s over two years old. sid already has 3.2.0 (https://packages.debian.org/sid/fpc-3.2.0), though, but that doesn’t count.
Title: Re: The Ranking of Free Pascal in the Tiobe-index
Post by: dbannon on August 22, 2020, 04:29:38 am
.... Currently, Ubuntu (https://changelogs.ubuntu.com/changelogs/pool/universe/f/fpc/fpc_3.0.4+dfsg-23/changelog)/Debian, FreeBSD, and OpenBSD all list 3.0.4 as the latest version. That version’s over two years old. sid already has 3.2.0 (https://packages.debian.org/sid/fpc-3.2.0), though, but that doesn’t count.

In Ubuntu's case thats because Ubuntu has a policy of not changing the packages in released series.  FPC320 was released after the feature freeze for U20.04 so, FPC304 was all that was available.  Its thier way of ensuring compatibility of the packages in release for the life of that release. I imagine other distros have similar approaches. Nothing the FPC/Lazarus teams can do about that.

I have a PPA with FPC320 (and Lazarus 2.0.10) for U18.04 and U20.04 that I use to build a my app, also in a PPA. Its not hard but users who choose to use things like that do so at their own risk.

Davo
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: trev on August 22, 2020, 06:17:56 am
I would have thought the easiest people to convert to Lazarus + FPC would be those who are using Delphi ;)

I started with Delphi v1 to write tax calculators for a commercial legal publisher to put on CDs. In the early 2000s I bought Kylix so I could port my Windows router utility to FreeBSD (runs Linux binaries). In 2007 I acquired a Mac mini and wanted to port the router utility to it which is when I stumbled on Lazarus + FPC. Then for more than a decade I was back programming in C on FreeBSD and Solaris for a free access legal publisher.

On retiring I made the rather expensive mistake of buying Delphi 10.1 because it now catered for macOS, except it only had a 32 bit compiler and just as I was about to upgrade, the Embarcadero CEO announced no support without current maintenance which included no more resetting of licence counts thereby depriving one of the "permanent/perpetual" Delphi licence (subsequently walked back). At that point I rediscovered Lazarus + FPC which had improved substantially since my last experience.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Thaddy on August 22, 2020, 07:39:55 am
trev, Lazarus was created by Delphi users, and FPC by TP users.....

But by now, there are developers who never touched either...
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on August 22, 2020, 12:42:19 pm
The same applies to the development of FPC/Lazarus even if I know that some people don't like Github but it's the #1 like Google is for searching the web. If you want to attract new people go to the place where they are. ;)
I mean, sure you can host your own Gitee/Gitlab instance but if the amount of volunteers is limited and the few don't even keep pace with updating/installing asked plug-ins or features why running another software?

It was already decided that if we (FPC) move to Git that we'll use our own (https://gitlab.freepascal.org/fpc) GitLab instance with probably a GitHub mirror (https://github.com/fpc/).
I see but it's exactly what I wrote - your Gitlab instance is kinda outdated. Even if you don't show the version (https://gitlab.freepascal.org/help) (see here (https://about.gitlab.com/blog/2015/05/07/version-check/)) one can find out that the instance uses JavaScript Cookie v2.1.3 while the official Gitlab (https://gitlab.com/help) uses JavaScript Cookie v2.2.1 (check main.*.chunk.js).
This change (https://gitlab.com/gitlab-org/gitlab/-/commit/6c798506230b3ed251f9f9e708950b326674f4df) was made in May and went into Gitlab 12.9 and thus you run a version prior that. The recent (still supported) versions are 13.0, 13.1 and 13.2 which get important security fixes (https://about.gitlab.com/releases/categories/releases/) quite often. I could keep on with the Mantis Bugtracker 2.24.0 or the MediaWiki 1.3.7 but I think I've already made my point and showed that you should rethink the idea with hosting your own software - or maintain and update them ASAP (hours not weeks).
I also don't believe it'll be a good publicity if the server/software gets hacked and all the user data are exposed.

Our decision is not up to discussion. Either that or we'll simply stay with SVN which we don't mind.

Also the GitLab instance was set up around a year ago as a test balloon. We'll update it should we do the final migration.


* Pull request easier for the developer than patches?
Well not for me.
I already work with git, on my local pc, and applying a patch and comparing it in git is as easy as looking at a pull request.
I should say that for me, the online interface to compare pullrequests is no good (So my feedback comes by mail, no annotations via the online interface). I always pull the requests to my local repro, and compare them locally.
So patch/pull requests, is the exact same amount of work to me.
It depends on your view and workflow but in general it's easier to accept small patches like this (https://bugs.freepascal.org/view.php?id=37049), this (https://bugs.freepascal.org/view.php?id=37616) or that one (https://bugs.freepascal.org/view.php?id=36949). There is not really a need to check it out locally as the review process can happen directly through your browser (additionally with CI you and the contributor see directly when creating the PR if it breaks something).
With hub (https://hub.github.com/) a contributor don't need to leave the console at all to fork and create a pull request. The reviewer can also comment on specific lines and give suggestions, advices or hints.
In my opinion these things are big advantages for both - reviewer and contributor - compared to the current workflow with issues/patches.

Our workflow won't change with a move to Git. We'll still review locally and ensure that nothing breaks, because we do not use CI.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 22, 2020, 01:38:15 pm
Quote
Quote
Keep in mind, it has been looked into already. And there are steps (e.g. developers have scripts for certain tasks) in the development/release process that need heavy work to be change.
So changing the main repro is a big task (bigger than you would think)
You should probably rewrite all these "old" scripts as the design is bad if it needs svn and cannot run on other directory structures easily.

Yes. Worse, they will have to be significantly enhanced, as GIT's mergetracking is inferior to SVNs.

Quote
Nowadays your scripts run automatically if triggered by commit/user action on your Github/Gitlab/Gitee instance.

CVS already had pre/post-commit hooks, such things are already in place for 20 years.

Quote
Therefore the release process happens almost automatically and needs only some checks to verify that everything went well. Sure, it might need some time to rewrite your scripts but the benefits of it are totally worth it.

The trouble is that when you actually want to write anything for it, and get stuck (as in said defective merge tracking), all advocates suddenly take a giant step back and only mutter things "you shouldn't want that".

Quote

* Pull request easier for the developer than patches?
Well not for me.
I already work with git, on my local pc, and applying a patch and comparing it in git is as easy as looking at a pull request.
I should say that for me, the online interface to compare pullrequests is no good (So my feedback comes by mail, no annotations via the online interface). I always pull the requests to my local repro, and compare them locally.
So patch/pull requests, is the exact same amount of work to me.

The very small commits are not the problem anyway. So that benefit is definitely not for the developer. Maybe for the submitter, but more often than not that ease is paired with a dropping of submission quality (like people reformatting/renaming/recasing). An extreme is the serbod chm (https://github.com/serbod/lazhelp) repository, which took me several evenings to mine anything from it. The fixes were good, but it was totally unusable, and couldn't be submitted piecemeal.

Too many unexperienced developers (either in absolute terms, or just inexperienced in working in teams where they don't call the shots), just go ahead, and only at submission time they start thinking.  That is a big problem with any VCS, and it won't be solved by a sliver of web over it.

It is an issue of spirit of teamwork/cooperation, and that is not fixed by tools.

Quote
It depends on your view and workflow but in general it's easier to accept small patches like this (https://bugs.freepascal.org/view.php?id=37049), this (https://bugs.freepascal.org/view.php?id=37616) or that one (https://bugs.freepascal.org/view.php?id=36949). There is not really a need to check it out locally as the review process can happen directly through your browser (additionally with CI you and the contributor see directly when creating the PR if it breaks something).

(Personally I prefer a good diff tool over the webjunk, CI per request takes too long for trivial changes. CI is a good principle, but not universal)

Quote
With hub (https://hub.github.com/) a contributor don't need to leave the console at all to fork and create a pull request. The reviewer can also comment on specific lines and give suggestions, advices or hints.
In my opinion these things are big advantages for both - reviewer and contributor - compared to the current workflow with issues/patches.

Note that those are just fixing of problems of GIT that makes simple patch creation apparently difficult.

Quote
* More contributors?
I agree that moving to e.g. Github won't bring 100 new developers but it will bring some publicity (blogs could write about the move) and that might lead to a few people who try it out and might stick to it. And after some time they might start to add their "missing feature".

I don't expect it to. Interested people will find a project, and are usually searching for live content, rather than autogenerated pages. We do bad there too (as in updating websites etc), but that won't change on github.

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 22, 2020, 02:46:53 pm
And here we are (as expected (at least as I somehow expected))

4 pages into the thread.
And not a single man-hour gained

That is no contributor has come up, and pledged an hour (or more) to do any work that would create publicity.
That is any work, that was not already planed before hand.

As said, existing contributors already have TODO list for years to come. And as far as I can tell, new contributors have not yet been sourced as a consequence of this thread. Nor does it look as if such resourcing is about to happen.

Still, at least some ideas were brought up. Unfortunately they are now hidden between lots of other messages.
Take away all the suggestions, that require work from people whose todo is already full. I.e, reduce to ideas that new independent contributors could do.

If anyone had pledged time, I would encourage that person to start gleaning those ideas, and put them on a "How to contribute" wiki page. (Existing or New? And may get extended with info for conventional contribution, i.e. patches, docs, ...).
But no one has pledged time.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: lainz on August 22, 2020, 03:03:33 pm
And here we are (as expected (at least as I somehow expected))

4 pages into the thread.
And not a single man-hour gained

That is no contributor has come up, and pledged an hour (or more) to do any work that would create publicity.
That is any work, that was not already planed before hand.

As said, existing contributors already have TODO list for years to come. And as far as I can tell, new contributors have not yet been sourced as a consequence of this thread. Nor does it look as if such resourcing is about to happen.

Still, at least some ideas were brought up. Unfortunately they are now hidden between lots of other messages.
Take away all the suggestions, that require work from people whose todo is already full. I.e, reduce to ideas that new independent contributors could do.

If anyone had pledged time, I would encourage that person to start gleaning those ideas, and put them on a "How to contribute" wiki page. (Existing or New? And may get extended with info for conventional contribution, i.e. patches, docs, ...).
But no one has pledged time.

I agree, well I have a lot of bugs to fix for BGRAControls, any help is welcome  :)
https://github.com/bgrabitmap/bgracontrols/issues
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: korba812 on August 22, 2020, 05:21:21 pm
And here we are (as expected (at least as I somehow expected))

4 pages into the thread.
And not a single man-hour gained

That is no contributor has come up, and pledged an hour (or more) to do any work that would create publicity.
That is any work, that was not already planed before hand.
A little less conversation, a little more action, please...  :P
https://www.youtube.com/watch?v=WWVMXLSS1cA (https://www.youtube.com/watch?v=WWVMXLSS1cA)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 22, 2020, 05:29:02 pm
And here we are (as expected (at least as I somehow expected))
...
If anyone had pledged time,...
But no one has pledged time.

I think, like Lainz said, that the active contributors-developers have enough work and will still continue to contribute, even without pledged time.

What is missing (imho) is contributors-marketing, gurus in publicity, kings of social network, like in each serious commercial factory.

But I fear that there are very few fpc developers (and developers in general) that have those genes.

Fre;D
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: asdf1337 on August 22, 2020, 05:45:39 pm
Sorry but I really have to disagree with some points. >:D


* Pull request easier for the developer than patches?
Well not for me.
I already work with git, on my local pc, and applying a patch and comparing it in git is as easy as looking at a pull request.
I should say that for me, the online interface to compare pullrequests is no good (So my feedback comes by mail, no annotations via the online interface). I always pull the requests to my local repro, and compare them locally.
So patch/pull requests, is the exact same amount of work to me.
It depends on your view and workflow but in general it's easier to accept small patches like this (https://bugs.freepascal.org/view.php?id=37049), this (https://bugs.freepascal.org/view.php?id=37616) or that one (https://bugs.freepascal.org/view.php?id=36949). There is not really a need to check it out locally as the review process can happen directly through your browser (additionally with CI you and the contributor see directly when creating the PR if it breaks something).
With hub (https://hub.github.com/) a contributor don't need to leave the console at all to fork and create a pull request. The reviewer can also comment on specific lines and give suggestions, advices or hints.
In my opinion these things are big advantages for both - reviewer and contributor - compared to the current workflow with issues/patches.

Our workflow won't change with a move to Git. We'll still review locally and ensure that nothing breaks, because we do not use CI.
That's one of the problems. You should not work in 2020 with methods from end of 1990. That's mostly the mistake done by old men as they slept for 20 years because everything always worked and 'never change a running system' and they didn't noticed that the world around them has changed. Now they cannot keep up with the youngsters anymore and think like 'nah I'm too old to learn something and anyway my retirement is soon'.
It's the same for company's which want to work 'agile' but still assigns several different projects to the same person - that won't work.

Quote
Quote
Keep in mind, it has been looked into already. And there are steps (e.g. developers have scripts for certain tasks) in the development/release process that need heavy work to be change.
So changing the main repro is a big task (bigger than you would think)
You should probably rewrite all these "old" scripts as the design is bad if it needs svn and cannot run on other directory structures easily.

Yes. Worse, they will have to be significantly enhanced, as GIT's mergetracking is inferior to SVNs.
Do you know about git branch --contains <hash>?
Gitlab also shows on each commit all branches/tags which contain the commit.
You cannot keep your old scripts when switching to another system. I also cannot run my Windows cmd.exe commands on Linux without adapting to the new behaviour and functionality of the operating system. I've to learn and adapt my thinking to the new environment but will be happy after I did so ;)

Maybe for the submitter, but more often than not that ease is paired with a dropping of submission quality (like people reformatting/renaming/recasing).
Well, if there is a defined standard for formatting in your project you can tell your tools about and everything will be formatted like this. But that would mean that everyone uses a defined formatting scheme but seems every FPC developers uses it's own (sometimes even different schemes in one file). Sorry but how should the submitter know which one is the right one to use? You cannot blame others for your fault.

(Personally I prefer a good diff tool over the webjunk, CI per request takes too long for trivial changes. CI is a good principle, but not universal)
Have you ever tried Gitlab WebIDE? It's pretty decent since some releases. I use it a lot to search/read/edit stuff in my company. The Github WebIDE is shit compared with it.
And it doesn't matter how long the CI runs - I'm sure you can also config it to run only tests which might be affected. Computing time doesn't matter and if you run it in parallel it doesn't consume that much time anyway. Its much cheaper than man hours.

Note that those are just fixing of problems of GIT that makes simple patch creation apparently difficult.
Please explain what exactly is so difficult in doing:

---------

I can understand if young people (students) don't want to work with old methods as they often do it to get some experience which might help them to find a job after graduating. All the leading company's in the digital area don't use ancient methods so where is the benefit of a student to learn long deprecated ways which disappear more and more in all company's? So those people contribute to projects which use modern methods (git, CI, etc).
Just read some job offers for 'newcomers': nobody asks that they knew svn, submitting patches by hand, running ancient computers which appeared before they were born, supporting operating system deprecated a long time ago, ...
So all these people are lost right now.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Handoko on August 22, 2020, 05:52:30 pm
I can understand if young people (students) don't want to work with old methods as they often do it to get some experience which might help them to find a job after graduating.

Sorry if I am wrong. I don't think young students are capable to join in to the developer team. An experience programmer can reach the goal, as long as there is a way to it.

I myself made several attempts to join in the developer team, but I failed. I visited the bugtracker, looking for some bug that seemed easy to me. If I really fixed some bugs, I would contact and request to be a member of the developer team. Unfortunately, I'm not skill enough none of the bugs I can fix. To be a member of the developer team, s/he needs to prove s/he is a capable person.

My 2c
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: asdf1337 on August 22, 2020, 06:03:30 pm
I can understand if young people (students) don't want to work with old methods as they often do it to get some experience which might help them to find a job after graduating.

Sorry if I am wrong. I don't think young students are capable to join in to the developer team. An experience programmer can reach the goal, as long as there is a way to it.
Why do you think so? Underestimating young people skills? Thinking that old men are better by law? You don't need to have a degree to be a (good) programmer. The best guys often don't have a degree at all and still fu*k the ones with a degree in all areas. Sometimes because they do it 'by heart' and not money.
Just to name a few: George Hotz, Tavis Ormandy, the guys from LulzSec/Anonymous, people running drugs darknet sites are below 30 (unfortunately they got arrested so you knew the age)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on August 22, 2020, 06:05:25 pm
Sorry but I really have to disagree with some points. >:D

You can disagree all you want. It's us core devs that need to work with this day in, day out, not you and we have processes that have shown in the past years that they work for us even if they might be "last century" for you.

The more everyone is nagging here I'm inclined to simply stay with SVN for the core repo and keep using git-svn on the client side. ::)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Handoko on August 22, 2020, 06:06:15 pm
Thinking that old men are better by law?

I never say that.

You post too fast without fully read my post.

So you're young and you think you're worthy? Goto  the bugtracker, pick some bugs and fix them. I you managed to do it, then I will bow on you. Prove it. Words are cheap, works are real.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 22, 2020, 06:13:15 pm
@ asdf1337: I agree totally with all of your points but this is a other story and has no cure.

To come with something constructive, I propose to every fpc user-lover, young and old, gurus and beginners, to give his part for the publicity of fpc.

It dont need to be a big mountain, but it could be some publication on social network, a advice in other computing forum, a YouTube Pascal site, wear a t-shirt with "I love Pascal", ...

If every fpc user do it, sure next month Pascal is number one in Tiobe.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: asdf1337 on August 22, 2020, 06:15:42 pm
I myself made several attempts to join in the developer team, but I failed. I visited the bugtracker, looking for some bug that seemed easy to me. If I really fixed some bugs, I would contact and request to be a member of the developer team. Unfortunately, I'm not skill enough none of the bugs I can fix.
Often its a longer process until someone grants you more rights and responsibility. You don't get it by just fixing a few simple bugs.
However, you don't need to be in the 'core team' to write your opinion about a new feature/bug etc. You might also help someone else with his patch without being in the 'core team'. That's what it is about to be in a community.
But after proving that you're really trying to work and improve the software you should be added to the team - maybe not as a developer but as a 'runner' who supports others and has a vote for decisions.
So I personally appreciate your work (even small work is work) and also the work of everyone else. :)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Handoko on August 22, 2020, 06:19:43 pm
I believe all people here want to make Lazarus/FPC better.

I could be wrong, but I believe we should not make the 'exam' to join the the development team easy. Don't feel sad if you can't be in the team, there still many things we can help.

I really appreciate you want to make it better.

~peace~
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: asdf1337 on August 22, 2020, 06:21:39 pm
However, you don't need to be in the 'core team' to write your opinion about a new feature/bug etc. You might also help someone else with his patch without being in the 'core team'.
Well, I think that's also a good point which could be improved:
Is there any place where people could come together and then start working on something? Like fixing a bug, maybe a single person isn't capable or motivated enough to work alone on it but two together could solve it.
Discord? Gitter? Microsoft Teams? What is there else?
In my opinion the bugtracker, mail and IRC are too deprecated and not comfortable enough to share code/links/suggestions (some of them also don't allow private conversations).
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Handoko on August 22, 2020, 06:28:14 pm
+1
That's a good idea. Need to do some research.

Anyone here know Blender 3D? They grow very fast which I believe one of the cause is Blender accept Google summer code. Is it any possible if Lazarus/FPC to do so?
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: lainz on August 22, 2020, 06:33:01 pm
I'm not sure if the position in the TIOBE index is bad at all, for example Kotlin (position 29) is below Delphi/Object Pascal (position 27). And is widely used in the Android world, I personally use it for making commercial Android apps at work.

TypeScript is at position 47 and is widely used for web development, for example Angular apps.

So we're talking of 2 big languages that are below Pascal  :)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 22, 2020, 06:34:58 pm
Do you know about git branch --contains <hash>?

And what happens if you had a huge merge conflict and edited it?

Afaik, git compares if 2 commits have the same (or similar) patch.
If that patch differs too much, then there is no info, that it was a merge (cherry pick) from another branch. (never mind which exact commit)

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Handoko on August 22, 2020, 06:35:50 pm
I'm not sure if the position in the TIOBE index is bad at all, for example Kotlin (position 29) is below Delphi/Object Pascal (position 27). And is widely used in the Android world, I personally use it for making commercial Android apps at work.

TypeScript is at position 47 and is widely used for web development, for example Angular apps.

So we're talking of 2 big languages that are below Pascal  :)

I think people care too much about the number.

I personally won't pick the tool just because it is no 1.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 22, 2020, 06:42:15 pm
I visited the bugtracker, looking for some bug that seemed easy to me. If I really fixed some bugs, I would contact and request to be a member of the developer team.

If you have a bug, you are interested in, you should ask for pointers where to start looking.

For example, this debugger related issue should be one of the easier issues. https://bugs.freepascal.org/view.php?id=37581
Then if someone would tell you in which unit/function the "ignored params" are accessed, and in which function they are ignored -> you would have a starting point.
Depending on what area a bug is on, different team members can advice. Though to reach more team members you need to ask on the mail list.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: asdf1337 on August 22, 2020, 06:44:41 pm
Do you know about git branch --contains <hash>?

And what happens if you had a huge merge conflict and edited it?

Afaik, git compares if 2 commits have the same (or similar) patch.
If that patch differs too much, then there is no info, that it was a merge (cherry pick) from another branch. (never mind which exact commit)

You should never edit/amend a commit once it has been sent to the outer world. The merge conflict will create a new commit and thus the history doesn't change!
Cherry pick also creates a reference to the original commit and I'm pretty sure it gets tracked by that. There was a link somewhere which showed that Gitlab maintains several releases (or use the different Linux versions + LTS version) at the same time and all of them contain different features but you're able to keep track of all the stuff and merge needed commits into older branches/tags so it's possible. (And Gitlab/Linux us huge compared to tiny FPC)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 22, 2020, 06:45:00 pm
off topic...

Where is the "official" fpc svn server?

I use this to synchronize:

Code: Pascal  [Select][+][-]
  1. svn checkout https://svn.freepascal.org/svn/fpc/branches/fixes_3_2

Afaik, that svn server is in USA.
Would it not be time to have fpc server in Europe?



Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 22, 2020, 06:45:14 pm
Sorry but I really have to disagree with some points. >:D

That's fine, we'll see what you come up with  >:D

So patch/pull requests, is the exact same amount of work to me.

Good, so then just send patches :-)

That's one of the problems. You should not work in 2020 with methods from end of 1990. That's mostly the mistake done by old men as they slept for 20 years because everything always worked and 'never change a running system' and they didn't noticed that the world around them has changed.

Note that about 3/4 of the devels have GIT and some also Mercurial experience (I only very minor). They still were hesitant. They (and me too) remember how the CVS to SVN conversion went. But that was a necessary one, this time around benefits are smaller, and used VCS featureset is higher.

Now they cannot keep up with the youngsters anymore and think like 'nah I'm too old to learn something and anyway my retirement is soon'.

Oh please, come back when you can shave and some real experience under your belt:-)

It's the same for company's which want to work 'agile' but still assigns several different projects to the same person - that won't work.

If you have 4 users in a new project it is different. If you have two repos with 40000-50000 commits each, and another half of that in the pre 2005 CVS history for a total of 25 years of VCS info. With say an hundred historical committers of which you at least want to keep recording the name, accidentally copyrighted code that must be blotted out, links between repos that must be honoured (svn external),  merge related info, file renaming that was recorded as (copy+delete) in the early days and the list goes on and on.

And once you start working with the conversions, new issues will pop up, and somebody will have to keep going after those. The CVS->SVN migration took an year, and then another year at a lower rate. (e.g. things that only pop up with major branches) before everything settled down. And I think this conversion will go slower because GIT is more complicated, and the advantages mostly theoretic for most users.

That is a reason why this is going so glacial, while most core members think GIT might be better, the current GIT<->SVN bridge already gives most advantages (like private branches etc), so the motivation to push through the whole transition pains is low it seems.

Quote
Yes. Worse, they will have to be significantly enhanced, as GIT's mergetracking is inferior to SVNs.
Do you know about git branch --contains <hash>?

Several models were tested, none worked. Afaik the problem is that cherry picking info is not complete if the master branch also contains revs merged from more forward branches. There was some scheme of recording some info while merging, but I lacked the GIT detail knowledge to fully understand it other than the management summary: there is no direct replacement, and it will require a mix of procedures (stuffing metadata in commit msgs) and software to get a list of revs available to merge from master to a fixes branch.

Quote
Gitlab also shows on each commit all branches/tags which contain the commit.

This is not info for a commit, but getting a (certain) list of commits. IOW identifying a list of hashes to merge. (or at least a list that contains all candidates that you can then could filter yourself in more detail using contains).

Since you know gitlab that well, it should be piece of cake to point that out?

Quote
You cannot keep your old scripts when switching to another system. I also cannot run my Windows cmd.exe commands on Linux without adapting to the new behaviour and functionality of the operating system. I've to learn and adapt my thinking to the new environment but will be happy after I did so ;)

You are mixing multiple different issues here.  Rewriting for a new system is to be expected, but for non trivial VCS systems eats a lot of time.

Worse is when features that are heavily used now (like the whole release engineering that relies heavily on merge tracking) simply isn't in a workable state.

Another annoying issues are SVN externals, server dictated configuration for e.g. CRLF handling, and hashes vs global revisions.

Hashes are not just more unwieldy, they are not sequential, so in more cases extra context (like date) must be added. (e.g. bugtracker). I really hate this.

Quote
Maybe for the submitter, but more often than not that ease is paired with a dropping of submission quality (like people reformatting/renaming/recasing).
Well, if there is a defined standard for formatting in your project you can tell your tools about and everything will be formatted like this.

Yeah. What current is in central SVN/GIT is the defined standard. Just don't mess with it unnecessarily.

Quote
But that would mean that everyone uses a defined formatting scheme but seems every FPC developers uses it's own (sometimes even different schemes in one file). Sorry but how should the submitter know which one is the right one to use? You cannot blame others for your fault.

We are not a bureaucracy, and the contributors are not (supposed  to be) children. Not everything must be locked down and nailed shut.

And the rule is simple: don't run formatters, and stay in the same style as the file. It has been communicated that way for 25 years. So when people then run formatters it is their own fault. And don't try to turn it on us, that is childish.

Quote
(Personally I prefer a good diff tool over the webjunk, CI per request takes too long for trivial changes. CI is a good principle, but not universal)
Have you ever tried Gitlab WebIDE? It's pretty decent since some releases. I use it a lot to search/read/edit stuff in my company. The Github WebIDE is shit compared with it.

No. I tried some stuff a few time, and found it cumbersome. Many pages are painful because they assume a monitor wider than its height, and I use my main programming system monitor in portrait mode. 

The webapps are generally worse in making use of available space IMHO and much slower to navigate (keyboard navigation is often impossible)

Quote
And it doesn't matter how long the CI runs - I'm sure you can also config it to run only tests which might be affected.

Sure, specially for non compiler commits. It is a long term plan to break up the testsuite (which is not tick-all-boxes CI, but could be a basis for it) in problem domains.

But of course, that also requires manpower. And if it is a 3rd party Java or .NET CI tool, customization needs somebody fluent in that to do all the work.  etc etc, it always boils down to manpower.

And it is not just us. It seems every major C++ project also uses its own build and CI tool

Quote
Computing time doesn't matter and if you run it in parallel it doesn't consume that much time anyway. Its much cheaper than man hours.

How many parallel Amiga's do you know?  Keep in mind this is not an x86 only project, and the slowest one is the rate determining step. OTOH, x86,x86_64 and some more popular ARM on Windows,Linux,OS X would already tick many boxes.

Well, more complete CI (beyond the nightly run of the testsuite there already is) doesn't automatically save that much programmer time, since most issues get caught anyway, or wouldn't be with a run on each commit either.

It only saves narrowing down to the exact daily commit and those cases are already somewhat rare.

Quote
Note that those are just fixing of problems of GIT that makes simple patch creation apparently difficult.
Please explain what exactly is so difficult in doing:
  • git clone (or git pull if already cloned master branch)
  • git checkout -b my_new_feature
  • git add -p (even simpler via GUI)
  • git commit
  • git push origin
  • create a merge request

versus

[li]fill in details and push submit.
[/li][/list]

Quote
I can understand if young people (students) don't want to work with old methods as they often do it to get some experience which might help them to find a job after graduating.

This is normal. We didn't like Cobol too either. But the difference seems to be that back then, it was an accepted solution that if you want to change things, that you needed to at least get your feet wet in the project, so that you could actually judge the outcome of modernization in a balanced way, and even then you were expected to do the bulk of the work/conversion.

And that is the thing that is different nowadays. Back then, hotheads that would reorganize the project in five minutes on the back of a beercoaster were automatically considered to be newbs and ridiculed. Usually they got frustrated, but quite some eventually restarted work in a more constructive way, and in time would become developers. (myself included, just ask Florian or Jonas for examples where I relentlessly pushed for short term fixes in the dos rtl. Luckily I'm more flexible and less argumentative now  ;))

However nowadays in the current paranoia-driven society, everybody sees conspiracies everywhere, and by defaults seem to assume he himself is right without actually checking the details or doing the work. And worse, such behaviour is wide accepted in all levels of society.

We are not dinosaurs, just a tad careful due to the scale of the project, and actually want to have firm grip and plan on things before we start. Because when you start you are committed, and if it goes wrong, it goes beyond a few days hard work to fix it, as it is with smaller projects.

Anyway, on the risk that I'm repeating myself: the devil is in the details and the scale with such endeavours. Also the advocated benefits are often for simple, ideal cases, which we might never reach

And if you really want to have some decision power here, get cracking and make developer. There is no power without responsibility (well, Boris John son and Donald Trump excluded)

Quote
Just read some job offers for 'newcomers': nobody asks that they knew svn, submitting patches by hand, running ancient computers which appeared before they were born, supporting operating system deprecated a long time ago, ...

Yeah I know. In job adds you always must have 5 years experience in a project that is only 3 years public :-)

People always advertise with shiny future, legacy free projects, but save for a few new squeaky clean startups, on the first day the "but first some work must be done on the current range on projects". 

Quote
So all these people are lost right now.

Then they are probably lost anyway, iow not our audience.

Also I think it is wrong to advocate Lazarus/FPC too prominent as all-or-nothing. I think more promotion should be done for 2nd language.

(so you are now doing web development in PHP? What do you use for a quick application? Let me tell you about this fantastic, free language..)  O:-)

p.s. I'm not in favor of the GIT migration, as you might have guessed. At least not at this point in time. It is still to raw, and several workflow scenarios seem to exist. But more importantly, even if GIT is better, I doubt that the conversion is actually worth the benefits.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: lucamar on August 22, 2020, 06:50:09 pm
I personally won't pick the tool just because it is no 1.

Since the first three as of this writing are C, Java and Python (in that order) that's good for your health :P
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Handoko on August 22, 2020, 06:55:24 pm
Oh, I'm too old  :'(

You can't teach an old dog new tricks.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 22, 2020, 06:58:31 pm
I personally won't pick the tool just because it is no 1.

Since the first three as of this writing are C, Java and Python (in that order) that's good for your health :P

Those are languages, not tools. IMHO the worst thing you can do is overfocus on the language while doing tool selection. Quite often other factors are more important to actually achieve your goals.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on August 22, 2020, 07:07:18 pm
Do you know about git branch --contains <hash>?

And what happens if you had a huge merge conflict and edited it?

Afaik, git compares if 2 commits have the same (or similar) patch.
If that patch differs too much, then there is no info, that it was a merge (cherry pick) from another branch. (never mind which exact commit)

You should never edit/amend a commit once it has been sent to the outer world. The merge conflict will create a new commit and thus the history doesn't change!
Cherry pick also creates a reference to the original commit and I'm pretty sure it gets tracked by that. There was a link somewhere which showed that Gitlab maintains several releases (or use the different Linux versions + LTS version) at the same time and all of them contain different features but you're able to keep track of all the stuff and merge needed commits into older branches/tags so it's possible. (And Gitlab/Linux us huge compared to tiny FPC)

This is not about editing/amending a published commit. This is about merging commits from trunk to fixes. Due to progressing on trunk this might lead to merge conflicts (e.g. the line after the fix was changed, now it doesn't apply cleanly anymore and the commit that changed that line can't be merged, because it's part of a bigger feature - yes, those things happen, I had one of those when preparing 3.2). In SVN I adjust the commit manually and the merge is still recorded in the mergeinfo so that one can see that, yes, that revision is merged. Your advice for Git to check for the hash will fail then as that will be different.

off topic...

Where is the "official" fpc svn server?

I use this to synchronize:

Code: Pascal  [Select][+][-]
  1. svn checkout https://svn.freepascal.org/svn/fpc/branches/fixes_3_2

Afaik, that svn server is in USA.
Would it not be time to have fpc server in Europe?

Did you even bother to check where the IP address is? The server resolves to 85.222.228.11 and a geolocation check (e.g. here (https://www.iplocation.net/ip-lookup)) will show you that it is located in the Netherlands (the Wiki for example is located on a different server in Germany).
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 22, 2020, 07:12:06 pm
off topic...

Where is the "official" fpc svn server?

I use this to synchronize:

Code: Pascal  [Select][+][-]
  1. svn checkout https://svn.freepascal.org/svn/fpc/branches/fixes_3_2

Afaik, that svn server is in USA.
Would it not be time to have fpc server in Europe?

Did you even bother to check where the IP address is? The server resolves to 85.222.228.11 and a geolocation check (e.g. here (https://www.iplocation.net/ip-lookup)) will show you that it is located in the Netherlands (the Wiki for example is located on a different server in Germany).

Oooops, my very bad.

So, perfect, I will sleep good tonight.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: asdf1337 on August 22, 2020, 07:18:56 pm
Quote
This is not info for a commit, but getting a (certain) list of commits. IOW identifying a list of hashes to merge. (or at least a list that contains all candidates that you can then could filter yourself in more detail using contains).

Since you know gitlab that well, it should be piece of cake to point that out?
Just picked a random project from Gitlab:
https://gitlab.com/feuertiger/feuertiger/-/commit/e0199cef7b089fdef2e1c140e9ed1ca65f66e810
-> shows a big list of branches and tags which contain this specific commit
https://gitlab.com/feuertiger/feuertiger/-/commit/30b1b737e5260d2668fd4cd3190a4725434f0039
-> show only dev as it's just commited
https://gitlab.com/feuertiger/feuertiger/-/compare/v1.0.9-renovate-cypress-4-x.2e0d14c7.0...dev
-> show all commits from current dev which are not part of the tag v1.0.9-renovate-cypress-4-x.2e0d14c7.0
-> could click on the commit and select 'cherry-pick' and put it into my branch e.g. for the next release (appropriate permissions implied)

Quote
Hashes are not just more unwieldy, they are not sequential, so in more cases extra context (like date) must be added. (e.g. bugtracker). I really hate this.
If you switch to Gitlab then use their issue system. You can reference issues/merge requests simply by the number and a reference is added everywhere.
Maybe you can important the old reports from Mantis (not sure if that important way is supported, others are)

Quote
Sure, specially for non compiler commits. It is a long term plan to break up the testsuite (which is not tick-all-boxes CI, but could be a basis for it) in problem domains.
Go by using DUnitX once it compiles with FPC. Attributes are supported now :)
It also has a new PR to support Gitlabs CI parsing (based on JUnit XML) to show all changes made by a MR/commit.

Quote
How many parallel Amiga's do you know?  Keep in mind this is not an x86 only project, and the slowest one is the rate determining step. OTOH, x86,x86_64 and some more popular ARM on Windows,Linux,OS X would already tick many boxes.
Just let the Gitlab Runners run in docker, then you can run all in parallel on one machine. I assume it doesn't need to be a super computer to execute it fast :D
Guess you can emulate Amiga with QEMU - however you only need to run the Amiga CI jobs when something as changed regarding Amiga. No need to run it when a change was made specific to ARM.

Quote
    svn diff >file.patch
    drag file into bugreport

  • fill in details and push submit.
I don't have any commit message, creation date or any other info I could need to recall why I did the change like I did. And importantly I need to leave my console window and probably also need to create an issue while creating directly a merge requests is easier and provides the same information.[/list]
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 22, 2020, 07:27:29 pm
What disturbed me is that I was thinking that the official svn server of fpc was in USA.

But, no it stay in EU.

Now for svn vs git, personally I feeling good with both systems.

For people that prefers GiHub or GitLab, the solution of Martin is perfect: mirrored with only pull synchro (not enable pull request) on GitHub and GitLab servers.

Those will be "read only", no pull request, only issues and comment and given for convenience.

And if you want to give patches, just do it using svn.

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: asdf1337 on August 22, 2020, 07:29:14 pm
Do you know about git branch --contains <hash>?

And what happens if you had a huge merge conflict and edited it?

Afaik, git compares if 2 commits have the same (or similar) patch.
If that patch differs too much, then there is no info, that it was a merge (cherry pick) from another branch. (never mind which exact commit)

You should never edit/amend a commit once it has been sent to the outer world. The merge conflict will create a new commit and thus the history doesn't change!
Cherry pick also creates a reference to the original commit and I'm pretty sure it gets tracked by that. There was a link somewhere which showed that Gitlab maintains several releases (or use the different Linux versions + LTS version) at the same time and all of them contain different features but you're able to keep track of all the stuff and merge needed commits into older branches/tags so it's possible. (And Gitlab/Linux us huge compared to tiny FPC)

This is not about editing/amending a published commit. This is about merging commits from trunk to fixes. Due to progressing on trunk this might lead to merge conflicts (e.g. the line after the fix was changed, now it doesn't apply cleanly anymore and the commit that changed that line can't be merged, because it's part of a bigger feature - yes, those things happen, I had one of those when preparing 3.2). In SVN I adjust the commit manually and the merge is still recorded in the mergeinfo so that one can see that, yes, that revision is merged. Your advice for Git to check for the hash will fail then as that will be different.
It's the same if you do it the correct way.
It just creates another commit on top which says that there was a merge conflict. The underlying commits won't be touched and thus are tracked.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 22, 2020, 07:34:29 pm
Quote
ings happen, I had one of those when preparing 3.2). In SVN I adjust the commit manually and the merge is still recorded in the mergeinfo so that one can see that, yes, that revision is merged. Your advice for Git to check for the hash will fail then as that will be different.
It's the same if you do it the correct way.
It just creates another commit on top which says that there was a merge conflict. The underlying commits won't be touched and thus are tracked.

Aaaand, we are back with the universal "you shouldn't need that" GIT solution to all problems. Seriously, they should just throw away the manual, and print that on a single sheet in big letters.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 22, 2020, 07:45:16 pm
@asdf1337

I have a good job for you.

fpc main server will stay using svn.
There are too much problems to do the conversion to git with very little advantage.

But it would be great if somebody (you) take care (administrator) of the fpc GitHub and GitLab mirrored site.
For example, if somebody does a pull request, you transform this into a svn patch and sent it to fpc-svn.

You may also promote fpc via the marketing features of GitHub and GitLab, update the wikis, ...

Thanks for this!

Fre;D


Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: trev on August 23, 2020, 09:52:41 am
I've updated the  Wiki press kit (https://wiki.freepascal.org/Press_Kit) except for the Delphi entry which looks like it dates from around the time of the abomination that was Delphi 8:

Quote
Delphi is still a superior tool for writing Windows applications. Borland wants to make their flagship product the ideal tool to develop .NET applications. This is done without full backward compatibility. Whether this a good idea is highly doubted by many people in the Pascal world. Delphi will be influential for the Pascal language for the forseeable time, though.

Someone who uses both Lazarus + FPC and a current Delphi should do this. It shouldn't take you long ;)

I can't claim this was an extra hour of work because I've been updating the Wiki almost daily for more than  a year  O:-)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Handoko on August 23, 2020, 10:08:34 am
Very informative, thank you.

Could we add CodeTyphon? We can mention it as FPC + a fork of Lazarus IDE. Although I know some may dislike CT, but they really done some hardwork especially the automate compile scripts.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 23, 2020, 10:57:25 am
Very informative, thank you.

Could we add CodeTyphon? We can mention it as FPC + a fork of Lazarus IDE.

It is not a fork, as despite what they say, as it is still being synchronized it seems.

Quote
Although I know some may dislike CT, but they really done some hardwork especially the automate compile scripts.

And never submitted it back, or cooperated in any other way. We don't own them anything, and more importantly, I don't see the benefit of mentioning them here.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Bi0T1N on August 23, 2020, 12:24:14 pm
I don't expect it to. Interested people will find a project, and are usually searching for live content, rather than autogenerated pages. We do bad there too (as in updating websites etc), but that won't change on github.
I would help to improve the website by fixing typos and other things if it would be put e.g. on Github. Just browse to here (https://www.freepascal.org/news.html) and you'll notice that there are two entries for November 28th, 2017 and February 15th, 2017 (one with i instead of y).
Please have a look at the Rust Blog (https://blog.rust-lang.org/) which uses a modern style (mobile friendly) and additionally it is fully editable (https://github.com/rust-lang/blog.rust-lang.org) through Github. In theory everyone could write an article (about the language, an own application, library, ...) which would then go "live" if accepted by the maintainer(s).

Maybe for the submitter, but more often than not that ease is paired with a dropping of submission quality (like people reformatting/renaming/recasing).
Well, if there is a defined standard for formatting in your project you can tell your tools about and everything will be formatted like this. But that would mean that everyone uses a defined formatting scheme but seems every FPC developers uses it's own (sometimes even different schemes in one file). Sorry but how should the submitter know which one is the right one to use? You cannot blame others for your fault.
To be honest, he's not totally wrong. In my opinion the core team should define a fixed set of rules which has to be applied for all new/changed code by all developers. I know that it doesn't make sense to reformat everything with a tool now as it'd destroy the history of each single change but you could always improve for the future.

Just as an example a few different styles:
https://github.com/graemeg/freepascal/commit/440e120cd0c95d28672b36a8a70e707cbc3b0df9#diff-8d490ea2a1d1e3e94c84742e00e4db76R43
Code: Pascal  [Select][+][-]
  1. function pass_typecheck: tnode; override;

https://github.com/graemeg/freepascal/commit/d3c7b17400c8f45534ece229abb91185848907d1#diff-21423c6288a9f69dea018dd8cb159769R39
Code: Pascal  [Select][+][-]
  1. function use_generic_mul32to64: boolean;override;

https://github.com/graemeg/freepascal/commit/da4334d5a0cebaeb0748dce74df9d846d4655fe3#diff-a03decce807dcf8290230d6e0817e928R36
Code: Pascal  [Select][+][-]
  1. procedure yyerror ( msg : String );

https://github.com/graemeg/freepascal/commit/ea4e23403df380e6997a19ba1e5112d198695517#diff-b99ddf3ee6857eca64431f4e2d045bb3R1437
Code: Pascal  [Select][+][-]
  1. constructor Create(TheOwner: TObject); virtual;

Just small but obvious differences which shows that even the core team doesn't have agreed on a common style so how should the contributor know which style he should use?
Or can everyone bring its own style into FPC? Will adapt my patches then :D
Code: Pascal  [Select][+][-]
  1. pRoCeDuRe AdumbExample(       out ANumber       :       String       );

Is there any place where people could come together and then start working on something? Like fixing a bug, maybe a single person isn't capable or motivated enough to work alone on it but two together could solve it.
Discord? Gitter? Microsoft Teams? What is there else?
In my opinion the bugtracker, mail and IRC are too deprecated and not comfortable enough to share code/links/suggestions (some of them also don't allow private conversations).
I think that's a pretty good idea as it's more fun to work in a team and exchange ideas. :)
I'm only familiar with Gitter from Indy (https://gitter.im/IndySockets/Indy) so far - besides the public chat you also have the opportunity to message someone privately.

We are not a bureaucracy, and the contributors are not (supposed  to be) children. Not everything must be locked down and nailed shut.

And the rule is simple: don't run formatters, and stay in the same style as the file. It has been communicated that way for 25 years.
Sorry but that's not a rule. ::) Every project should have a defined set of rules (commit message layout, coding style, dir structure, allowed set of language extensions (e.g. only C++11 or newer), conventions for naming functions/parameters, etc).
Just click here (https://github.com/graemeg/freepascal/blob/440e120cd0c95d28672b36a8a70e707cbc3b0df9/compiler/i8086/n8086cal.pas#L35) and you see that someone failed to follow your above rule.
Code: Pascal  [Select][+][-]
  1.           procedure pop_parasize(pop_size:longint);override;
  2.           procedure extra_interrupt_code;override;
  3.           procedure extra_call_ref_code(var ref: treference);override;
  4.           function do_call_ref(ref: treference): tcgpara;override;
  5.         public
  6.           function pass_typecheck: tnode; override;
That's why you have tools for that because nobody is perfect. ;)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on August 23, 2020, 01:42:52 pm
Quote
Sure, specially for non compiler commits. It is a long term plan to break up the testsuite (which is not tick-all-boxes CI, but could be a basis for it) in problem domains.
Go by using DUnitX once it compiles with FPC. Attributes are supported now :)
It also has a new PR to support Gitlabs CI parsing (based on JUnit XML) to show all changes made by a MR/commit.

The amount of tests of our testsuite that can be done using a framework like DUnitX or fcl-fpcunit is minuscle. The majority of tests are compiler tests (some are also tested whether they run) where a test can fail to compile. Or even should fail to compile. Also we already have fcl-fpcunit and are using that for tests where it makes sense (e.g. the Rtti unit or Generics.Collections).

Quote
How many parallel Amiga's do you know?  Keep in mind this is not an x86 only project, and the slowest one is the rate determining step. OTOH, x86,x86_64 and some more popular ARM on Windows,Linux,OS X would already tick many boxes.
Just let the Gitlab Runners run in docker, then you can run all in parallel on one machine. I assume it doesn't need to be a super computer to execute it fast :D
Guess you can emulate Amiga with QEMU - however you only need to run the Amiga CI jobs when something as changed regarding Amiga. No need to run it when a change was made specific to ARM.

First of there is no official port of QEMU for the old M68k processors, only for Coldfire and even then I don't know whether that supports the specific Amiga chipsets. Thus emulators like WinUAE would need to be used.

Also it's not trivial to determine if something affected Amiga or not. Okay, the changes that are solely in e.g. the aarch64 directory of the compiler or RTL won't affect Amiga, but everything else can have effects on any target and we already had build errors on some of the smaller platforms due to some changes that looked alright.

I don't expect it to. Interested people will find a project, and are usually searching for live content, rather than autogenerated pages. We do bad there too (as in updating websites etc), but that won't change on github.
I would help to improve the website by fixing typos and other things if it would be put e.g. on Github. Just browse to here (https://www.freepascal.org/news.html) and you'll notice that there are two entries for November 28th, 2017 and February 15th, 2017 (one with i instead of y).
Please have a look at the Rust Blog (https://blog.rust-lang.org/) which uses a modern style (mobile friendly) and additionally it is fully editable (https://github.com/rust-lang/blog.rust-lang.org) through Github. In theory everyone could write an article (about the language, an own application, library, ...) which would then go "live" if accepted by the maintainer(s).

The Free Pascal website is available in SVN (https://svn.freepascal.org/cgi-bin/viewvc.cgi/?root=fpchtml) as well. One can provide patches already if someone is enthusiastic enough.

Maybe for the submitter, but more often than not that ease is paired with a dropping of submission quality (like people reformatting/renaming/recasing).
Well, if there is a defined standard for formatting in your project you can tell your tools about and everything will be formatted like this. But that would mean that everyone uses a defined formatting scheme but seems every FPC developers uses it's own (sometimes even different schemes in one file). Sorry but how should the submitter know which one is the right one to use? You cannot blame others for your fault.
To be honest, he's not totally wrong. In my opinion the core team should define a fixed set of rules which has to be applied for all new/changed code by all developers. I know that it doesn't make sense to reformat everything with a tool now as it'd destroy the history of each single change but you could always improve for the future.

The styles differ mainly between the involved components, namely Compiler, RTL and packages. Also things would be easier if Lazarus would support project specific settings for CodeTools, because I have a different set for my own work, alas my feature request (https://bugs.freepascal.org/view.php?id=18517) for that was denied.

Just small but obvious differences which shows that even the core team doesn't have agreed on a common style so how should the contributor know which style he should use?
Or can everyone bring its own style into FPC? Will adapt my patches then :D
Code: Pascal  [Select][+][-]
  1. pRoCeDuRe AdumbExample(       out ANumber       :       String       );

One needs to adhere to the surrounding style with especially the style for the compiler becoming more and more documented and fixed over time. New code (especially new packages) best adheres to one of the common Delphi styles.

Is there any place where people could come together and then start working on something? Like fixing a bug, maybe a single person isn't capable or motivated enough to work alone on it but two together could solve it.
Discord? Gitter? Microsoft Teams? What is there else?
In my opinion the bugtracker, mail and IRC are too deprecated and not comfortable enough to share code/links/suggestions (some of them also don't allow private conversations).
I think that's a pretty good idea as it's more fun to work in a team and exchange ideas. :)
I'm only familiar with Gitter from Indy (https://gitter.im/IndySockets/Indy) so far - besides the public chat you also have the opportunity to message someone privately.

I don't use systems like Discord, Microsoft Teams, Zoom, etc. Though that Gitter one does read interesting. We'd have to check whether it's possible to run that with a private instance as well...
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: howardpc on August 23, 2020, 01:54:04 pm
... I can't claim this was an extra hour of work because I've been updating the Wiki almost daily for more than  a year  O:-)

trev, You are one of the rather limited number of people who not only broadcast good ideas about Free Pascal and Lazarus publicity and documentation, but actually spend significant chunks of time doing something constructive to improve the (until recently woeful) situation.
I think I speak for many in recording our heartfelt thanks to you.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 23, 2020, 02:20:14 pm
Please have a look at the Rust Blog (https://blog.rust-lang.org/) which uses a modern style (mobile friendly)

That's not modern, that is 1997 html 1 design. A clickable list of urls to the actual topics. It looks like something from a failed html->txt converter.

Quote
and additionally it is fully editable (https://github.com/rust-lang/blog.rust-lang.org) through Github. In theory everyone could write an article (about the language, an own application, library, ...) which would then go "live" if accepted by the maintainer(s).

Yeah. Just like in theory everybody could submit patches and well about anything to the documentation.

Quote
Well, if there is a defined standard for formatting in your project you can tell your tools about and everything will be formatted like this.

Till the next newb creates an inflammatory thread on the forum to change the official style since it is outdated, or some feature gets added, or Embarcadero does something stupid again etc etc. It will never stop.

This may work with new languages with all new codebases directly in the advocated style, but with a 50 years old multi dialect language with many styles in circulation, AND leaning on submissions for expansions in many aspects of the project, it is different. Just be a bit flexible, being so rigid makes you look old  O:-)

Quote
But that would mean that everyone uses a defined formatting scheme but seems every FPC developers uses it's own (sometimes even different schemes in one file). Sorry but how should the submitter know which one is the right one to use? You cannot blame others for your fault.

The scheme of the surrounding code, period. If there are differences they are probably not that important. Our prime objective is to produce a pascal compiler, not nail every theoretical style possibility down.

Quote
To be honest, he's not totally wrong. In my opinion the core team should define a fixed set of rules which has to be applied for all new/changed code by all developers. I know that it doesn't make sense to reformat everything with a tool now as it'd destroy the history of each single change but you could always improve for the future.

Mixing style formatting with real changes makes reviewing also difficult, so that is why we refrain from that. And we accept minor discrepancies. It is no big deal. I'm so used to it, that I had to look really hard to actually see them in your posted urls.

Quote
Or can everyone bring its own style into FPC? Will adapt my patches then :D

You miss the central point, FPC is not about style. And trying to start own syntax is violates the rules, so won't be accepted, or roughly reformatted or whatever.

Is there any place where people could come together and then start working on something? Like fixing a bug, maybe a single person isn't capable or motivated enough to work alone on it but two together could solve it.
Discord? Gitter? Microsoft Teams? What is there else?
In my opinion the bugtracker, mail and IRC are too deprecated and not comfortable enough to share code/links/suggestions (some of them also don't allow private conversations).

There is some facebook that was active for some time, though shared with Delphi. But usually all these services rise and wane fairly quickly (if not in existence, then in popularity every few years).  I still have hyves,linkedin, orkut,ning and whatever accounts.

I would advise against Discord despite having an account since they can require you to enter a phonenumber for mandatory 2FA, and you don't want that kind of trouble for something you want to advocate as an official resource.

Anyway, do we really need more places to whine about how people really would like to contribute if we did X and Y and Z ? I think there are enough already. And nothing much ever comes from it anyway. People with that mind set will never contribute, since they prove them selves to be fundamentally incapable of cooperating (or anything but their own terms, which is the same thing).

If you want to gain useful experience for a corporate career, I suggest to tackle that. Learn to cooperate, learn to adapt, learn to bear responsibility and most of all DO PRODUCTIVE WORK. Everybody can stand to the side and find something to complain about, but that is not productive.

Don't overfocus on tools that will be in decline in 5 years and forgotten in 10

Quote
We are not a bureaucracy, and the contributors are not (supposed  to be) children. Not everything must be locked down and nailed shut.

And the rule is simple: don't run formatters, and stay in the same style as the file. It has been communicated that way for 25 years.
Sorry but that's not a rule. ::)

Of course it is. It is even in the imperative form, which is like a minimal requirement for a rule. :)

Quote
Every project should have a defined set of rules (commit message layout, coding style, dir structure, allowed set of language extensions (e.g. only C++11 or newer), conventions for naming functions/parameters, etc).

No, why? Not everybody is so narrowminded, and likes maintaining such standards for little purpose. While most programmers are control-freaks, that isn't an excuse to wallow in it.

Quote
That's why you have tools for that because nobody is perfect. ;)

Yeah, and using tools is perfect and the discussion will stop after 50 years of style wars. Dream on!
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 23, 2020, 02:29:09 pm
There is not only Tiobe in this world.

Here Pascal is 11th of the top 40:
https://raygun.com/blog/programming-languages/

And to promote Pascal, everybody may give some tips on various site, for example:
https://www.quora.com/What-are-some-popular-applications-written-in-Pascal

Come on, make some noise.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 23, 2020, 02:39:18 pm
Quote
Here Pascal is 11th of the top 40:
https://raygun.com/blog/programming-languages/

Ooops, I did not read their comment about Object Pascal.
Stupid ragun, he dont know that Pascal will never die.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 23, 2020, 03:04:53 pm
additionally it is fully editable (https://github.com/rust-lang/blog.rust-lang.org) through Github. In theory everyone could write an article (about the language, an own application, library, ...) which would then go "live" if accepted by the maintainer(s).

We have a wiki, everyone can write, and it goes live immediately.
And a forum, everyone can post...

Our website is avail on svn, so patches can be made. But of course, as in the github example, patches must be accepted. Do not expect that just any proposal will make it. The website is less of a "could be changed", but more of a "is the change really really necessary?"
Yet, I pointed out earlier, someone wants to contribute towards the website being mobile friendlier (without changing content, design, or concept of layout other than absolutely required) => come forward. (I advice to discuss first)

You think you have good ideas what we should tweet about? Well, come forward. You will have to meet all of the teams expectations. So there might be a bit of finding the exact ground to stand on.... Also, if your ideas are acceptable, you wont necessary get access to the account, but you can then send the text proposals, and we would use them.
Or start a twitter account of your own, about Lazarus...

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 23, 2020, 04:16:10 pm
Helllo.

About:

https://wiki.freepascal.org/Projects_using_Lazarus

Do you agree to re-name or have a new wiki named:

Code: Pascal  [Select][+][-]
  1. https://wiki.freepascal.org/Projects_using_fpc

And maybe add sub-menu "Graphic/Lazarus", "Graphic/fpGUI" "Graphic/MSEgui", "Console", "Libraries".

So console, fpGUI, MSEgui applications and libraries could be added?

Fre;D


Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 23, 2020, 04:26:13 pm
So long as the old page becomes a redirect. (it matters to some, it costs nothing)

IMHO it should be "front end" or "user interface", not "graphics".
And probably "UI/LCL", instead of lazarus.

Though the entire splitting into group, while necessary, should not be overdone....
There are to many grouping criteria. And the importance of each really differs to everyone. And we do not have a database to serve them all.....
Otherwise there would be: Avail for which OS, license, last updated, ....

--EDIT
You mean: https://wiki.freepascal.org/Projects_using_Lazarus
(uppercase "L")

It seems the wiki is case sensitive.



On second thought
Quote
And maybe add sub-menu "Graphic/Lazarus", "Graphic/fpGUI" "Graphic/MSEgui", "Console", "Libraries".
Either I misunderstand you, in which case please point it out to me....

Or, does that somehow conflict with existing options.
I.E. an FPGUI app can fall into other existing categories?

And in case you do mean "graphics" as in apps to process images, then why subsect them into widgetset?



Btw, if moving is to be done.

There is a move button, but possible only avail for wiki admins. So unless you see this, let an admin do it. That will afaik keep the history.




If this needs further talk: Please start a new thread. Thanks
Title: Re: The Ranking of Free Pascal in the Tiobe-index
Post by: Kays on August 23, 2020, 04:29:13 pm
[…] Yeah. Just like in theory everybody could submit patches and well about anything to the documentation. […]
Yes, in theory everybody could tweak the documentation by submitting patches, but documentation has to come from the developers. I simply can’t document anything I don’t have no clue about. My past patches for the docs have primarily been about fixing mistakes, not amending documentation of new features.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: asdf1337 on August 23, 2020, 08:45:56 pm
Quote
ings happen, I had one of those when preparing 3.2). In SVN I adjust the commit manually and the merge is still recorded in the mergeinfo so that one can see that, yes, that revision is merged. Your advice for Git to check for the hash will fail then as that will be different.
It's the same if you do it the correct way.
It just creates another commit on top which says that there was a merge conflict. The underlying commits won't be touched and thus are tracked.

Aaaand, we are back with the universal "you shouldn't need that" GIT solution to all problems. Seriously, they should just throw away the manual, and print that on a single sheet in big letters.
There is a reason why it's Github, Gitee and Gitlab and not SVNHub or SVNLab. ;)
Just accept that git is better and has won. Nobody except some old white men will remember SVN in a few years - nobody will cry about that. Period.

fpc main server will stay using svn.
There are too much problems to do the conversion to git with very little advantage.
If someone is against something, he will always find something which prevents it.

But it would be great if somebody (you) take care (administrator) of the fpc GitHub and GitLab mirrored site.
For example, if somebody does a pull request, you transform this into a svn patch and sent it to fpc-svn.

You may also promote fpc via the marketing features of GitHub and GitLab, update the wikis, ...
Sorry I do not have the time to convert things to add it to ancient software (svn).

To be honest, he's not totally wrong.
I'm always right - period. ;D
But thanks for giving some examples - looks like I'm not the only one seeing the issues with code formatting.

The amount of tests of our testsuite that can be done using a framework like DUnitX or fcl-fpcunit is minuscle.
DUnitX is also open-source. Feel free to send patches to adapt it to the needs of FPC. The guy seems to invite people nicely (https://github.com/VSoftTechnologies/DUnitX/issues) to submit patches.

First of there is no official port of QEMU for the old M68k processors, only for Coldfire and even then I don't know whether that supports the specific Amiga chipsets. Thus emulators like WinUAE would need to be used.
Shouldn't be problem, you still can run several VMs with it in parallel.

The styles differ mainly between the involved components, namely Compiler, RTL and packages. Also things would be easier if Lazarus would support project specific settings for CodeTools, because I have a different set for my own work, alas my feature request for that was denied.
Have you heard about EditorConfig (https://editorconfig.org/)? Not going to fix everything but at least the indentation of begin/end blocks eg.

This may work with new languages with all new codebases directly in the advocated style, but with a 50 years old multi dialect language with many styles in circulation, AND leaning on submissions for expansions in many aspects of the project, it is different. Just be a bit flexible, being so rigid makes you look old
omfg - telling me to be flexible but being a stubborn yourself? ::)
It's simply a mistake that nobody in the beginning defined a coding standard. Not my failure as I wasn't involved in this mistake - just here now to tell you that someone made a mistake and seems nobody feels responsible to solve it!

If you want to gain useful experience for a corporate career, I suggest to tackle that. Learn to cooperate, learn to adapt, learn to bear responsibility and most of all DO PRODUCTIVE WORK. Everybody can stand to the side and find something to complain about, but that is not productive.
Nobody who is really doing productive work works with methods from 1990. Especially in software development.

Don't overfocus on tools that will be in decline in 5 years and forgotten in 10
Then please switch to git - svn will be forgotten when the old men died.

The website is less of a "could be changed", but more of a "is the change really really necessary?"
Yet, I pointed out earlier, someone wants to contribute towards the website being mobile friendlier (without changing content, design, or concept of layout other than absolutely required) => come forward. (I advice to discuss first)
For me that sounds like: nothing is going to be changed!

[…] Yeah. Just like in theory everybody could submit patches and well about anything to the documentation. […]
Yes, in theory everybody could tweak the documentation by submitting patches, but documentation has to come from the developers. I simply can’t document anything I don’t have no clue about. My past patches for the docs have primarily been about fixing mistakes, not amending documentation of new features.
Agrred, if you contribute something you should also take care its getting documented properly. Otherwise its useless if nobody knows about it or knows how to use it.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 23, 2020, 09:45:47 pm
The website is less of a "could be changed", but more of a "is the change really really necessary?"
Yet, I pointed out earlier, someone wants to contribute towards the website being mobile friendlier (without changing content, design, or concept of layout other than absolutely required) => come forward. (I advice to discuss first)
For me that sounds like: nothing is going to be changed!

Well no, not nothing ...
But also, not just every idea and every request.

To clarify, as there are the fpc and lazarus website: this is about https://www.lazarus-ide.org/


To start with, the current version came from a contributor. And we changed to it. So it can be done.


However, it went through a long process. Many team members had clear expectations on different parts of the site. And after that implementation details, such as browser compatibility took another huge effort (Well yes, we only tested all the browsers back then, we don't keep up.. Lack of manpower)

So be clear what you want:

- Compatibility with mobile browsers. That is a technical issue, a bug. You provide a patch. We check that it works. Done.
   No changes to layout or content... Except if there is a "MinWidth: 4000px", that needs to be replaced...


- Different Content: Please propose what you want. Then it will be reviewed.
  However, we feel that most content should go to the wiki.
  + Easier for the contributor
  + Less work to review for us (We rather implement fixes and features to the code base)
  + More maintainable if it is likely to change over time.
If you do feel there is content, that despite this should go to the website, feel free to ask. If there are good reasons...
Minor changes such as adding a link to another portal on our wiki have occurred in the past, and mostly been accepted without any fuzz.


- Different Design.  (Other than implicitly dictated by technical reasons to ensure its working on all devices)
  That depends how well you can argue your reasons. But from my experience, it is the least likely to be accepted for change.
  Design is highly subjective. I am aware there are all kind of guidelines. By different, often self promoted, "experts" of the field. But as I said: all kinds. Just because you found an expert saying design must now be X, does not mean that there is not another expert saying the opposite.
  So as I said: highly subjective. And the subjective opinion that counts, is that of the team. (If everyone in the team, likes your idea... ok)


Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: lainz on August 23, 2020, 09:53:39 pm
Where is the repository with the sources of the Lazarus website / forum?

So we can play with it and try to make it mobile friendly.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: lainz on August 23, 2020, 09:56:48 pm
Here is the sourcecode of the Packages (OPM) website
https://github.com/lazarusccr/LazarusPackagesRepository

I did it some years ago, with an old version of Bootstrap, and doesn't look very mobile friendly, plus it doesn't look like the Lazarus website, it works but the style should be changed as well...
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 23, 2020, 10:18:48 pm
web
https://svn.freepascal.org/cgi-bin/viewvc.cgi?view=revision&root=lazaruswebsite

svn
https://svn.freepascal.org/svn/lazaruswebsite/trunk
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 23, 2020, 10:20:33 pm
Here is the sourcecode of the Packages (OPM) website
https://github.com/lazarusccr/LazarusPackagesRepository

Actual site: http://packages.lazarus-ide.org/
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: lainz on August 23, 2020, 10:29:00 pm
Here is the sourcecode of the Packages (OPM) website
https://github.com/lazarusccr/LazarusPackagesRepository

Actual site: http://packages.lazarus-ide.org/

Yes is the same that's on the sourcecode. There was a previous version that was made with material design theme, but it was a bit slower (not the theme, but the code) so I finally used bootstrap and angularjs v1 for the actual site. But I want to use material design again if I have time to change it, there is for example bootstrap material that looks really good on mobile devices (for the packages site only, not for the lazarus site).

web
https://svn.freepascal.org/cgi-bin/viewvc.cgi?view=revision&root=lazaruswebsite

svn
https://svn.freepascal.org/svn/lazaruswebsite/trunk

Thanks. I will take a look, I need to install PHP or some WAMP server (I'm on Windows) to test it since it seems that uses PHP.

A first look seems that it was not designed for mobile, since it lacks of the "meta viewport".
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on August 23, 2020, 10:44:51 pm
Thanks. I will take a look, I need to install PHP or some WAMP server (I'm on Windows) to test it since it seems that uses PHP.

Please start a new thread, for further discussions on the subject (if/when needed). Thanks
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: lainz on August 23, 2020, 10:54:37 pm
Thanks. I will take a look, I need to install PHP or some WAMP server (I'm on Windows) to test it since it seems that uses PHP.

Please start a new thread, for further discussions on the subject (if/when needed). Thanks

Ok.

Here it is
https://forum.lazarus.freepascal.org/index.php/topic,51174.msg375259/topicseen.html#new
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Bi0T1N on August 24, 2020, 12:17:24 pm
I don't expect it to. Interested people will find a project, and are usually searching for live content, rather than autogenerated pages. We do bad there too (as in updating websites etc), but that won't change on github.
I would help to improve the website by fixing typos and other things if it would be put e.g. on Github. Just browse to here (https://www.freepascal.org/news.html) and you'll notice that there are two entries for November 28th, 2017 and February 15th, 2017 (one with i instead of y).
Please have a look at the Rust Blog (https://blog.rust-lang.org/) which uses a modern style (mobile friendly) and additionally it is fully editable (https://github.com/rust-lang/blog.rust-lang.org) through Github. In theory everyone could write an article (about the language, an own application, library, ...) which would then go "live" if accepted by the maintainer(s).

The Free Pascal website is available in SVN (https://svn.freepascal.org/cgi-bin/viewvc.cgi/?root=fpchtml) as well. One can provide patches already if someone is enthusiastic enough.
Didn't knew about that but I also have never really browsed the SVN as I prefer the Github mirror from Graeme Geldenhuys.

Is there any place where people could come together and then start working on something? Like fixing a bug, maybe a single person isn't capable or motivated enough to work alone on it but two together could solve it.
Discord? Gitter? Microsoft Teams? What is there else?
In my opinion the bugtracker, mail and IRC are too deprecated and not comfortable enough to share code/links/suggestions (some of them also don't allow private conversations).
I think that's a pretty good idea as it's more fun to work in a team and exchange ideas. :)
I'm only familiar with Gitter from Indy (https://gitter.im/IndySockets/Indy) so far - besides the public chat you also have the opportunity to message someone privately.

I don't use systems like Discord, Microsoft Teams, Zoom, etc. Though that Gitter one does read interesting. We'd have to check whether it's possible to run that with a private instance as well...
Gitter is fully open-source, see here (https://gitlab.com/gitlab-org/gitter/webapp). However, if you host own software please keep the software always up-to-date!

Please have a look at the Rust Blog (https://blog.rust-lang.org/) which uses a modern style (mobile friendly)

That's not modern, that is 1997 html 1 design. A clickable list of urls to the actual topics. It looks like something from a failed html->txt converter.
Come on, please browse the full webpage and not just the link I gave as an example.
It's a simple but awesome and responsive design (without any fancy crap). Nothing like this probably existed in 1997. Especially because you don't need to know HTML at all to write an article. This also reduces the obstacles for a proficient writer. :)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 27, 2020, 02:28:26 pm
Quote
Aaaand, we are back with the universal "you shouldn't need that" GIT solution to all problems. Seriously, they should just throw away the manual, and print that on a single sheet in big letters.
There is a reason why it's Github, Gitee and Gitlab and not SVNHub or SVNLab. ;)
Just accept that git is better and has won. Nobody except some old white men will remember SVN in a few years - nobody will cry about that. Period.

Even then there is a choice between doing an haphazard migration now and waiting a few years till they finally get their act together. (and I'm not even so sure about that, it might be that github 10 years in the future will be like sf now).

Maybe then its users will actually be able to give to the point details, instead of just banging on about how popular it is and you shouldn't want anything else.

Quote
fpc main server will stay using svn.
There are too much problems to do the conversion to git with very little advantage.
If someone is against something, he will always find something which prevents it.

That is a gross oversimplification. But it also works the other way, people sometimes overlook the details if they have a real big drive for something.

Anyway, in my case there are multiple levels of against it. Most have been said. My original position was that I always thought it was too much work for benefit, but assumed git was a netto plus (which I mainly considered to be better merging and the more seamless offline work of a DVCS). With  the losing of global revisions and more cmdline actions for a small fix as main productivity sacrifices/problems.

There was a majority of devels in favour of looking into GIT and a migration about an year ago. The wiki page about the migration is mostly my initial list of SVN features used that needed equivalents in GIT.

Some had been (allegedly) fixed, like per repo CRLF handling (older GIT needed manual client configuration, which was a risk, one of the reasons we migrated from CVS to SVN). 

Losing global rev still hurts, and IMHO guids are unwieldy but maybe with something webbased linked to it, it will work out, and strictly speaking global revs also always needed branch as context in some cases, though that usually wasn't the problem, with only a handful of SVN branches (trunk,fixes) in multi-user use.

That left (to my surprise) the merge history stuff. I never expected a newer VCS to have a problem there, it was only on my list to have the alternative investigated before rather than after the migration.

But now we are a full year further, with all kinds of advocates writing lengthy posts and mails on it, and still we are not further than stuffing it manually in commit messages, and then writing software to string it together and that that is normal because GIT is so popular and you shouldn't want to have it any different way.

That is not progress, that is regression.

But it would be great if somebody (you) take care (administrator) of the fpc GitHub and GitLab mirrored site.
For example, if somebody does a pull request, you transform this into a svn patch and sent it to fpc-svn.

This won't work. A very large percentage of the patches are tossed back to the users for rewrite/cleanup. This is much harder with push/pull requests and at that point the history is toast anyway. The ones that apply cleanly and unmodified (my guess: about a third) are the trivial ones that work with any VCS system.

It is no use saying you accept pull requests if you have to reject all of them, except from some usual suspects that could do it either way. Maybe a merge request size limit for pull requests would put some onus on the submitter to only submit doable requests(iow only for trivial one line stuff), but that would probably require gitlab hacking etc again.

Also, the cloud stuff never will be anything but a mirror.  The central server will still be locally owned gitlab setup.

This may work with new languages with all new codebases directly in the advocated style, but with a 50 years old multi dialect language with many styles in circulation, AND leaning on submissions for expansions in many aspects of the project, it is different. Just be a bit flexible, being so rigid makes you look old
omfg - telling me to be flexible but being a stubborn yourself? ::)

As said I'm pretty lax as it comes to coding styles. That I say no to a hopeless project like rearrange everything stylewise,  write long documents to nail down the coding style to every last detail, build an automated custom framework to enforce it and then keep supporting it and discussing it ad infinitum doesn't mean I'm automatically an old stubborn ass in this.

It is simply pragmatism. At the very least before you start converting 10M lines of projects worth to such setup it should be fully production tested in a smaller but not trivial project.

And as far as stubbornness wrt GIT goes, I do know the project intimately after 20 years, and I studied GIT at least somewhat and tracked the discussions for 5 years+. I also come with arguments that are more than a simple popularity contest, or idealist advocacy copied right from the web. So please get off your high horse and do the same.

Quote
It's simply a mistake that nobody in the beginning defined a coding standard. Not my failure as I wasn't involved in this mistake - just here now to tell you that someone made a mistake and seems nobody feels responsible to solve it!

No, you are declaring it a mistake to justify your urgent need to nail down codestyle and converters. That is a choice not a decree that fell down from the heavens.

And I don't agree with that choice.

Quote
If you want to gain useful experience for a corporate career, I suggest to tackle that. Learn to cooperate, learn to adapt, learn to bear responsibility and most of all DO PRODUCTIVE WORK. Everybody can stand to the side and find something to complain about, but that is not productive.
Nobody who is really doing productive work works with methods from 1990. Especially in software development.

Hahaha, you would be surprised what old crap you encounter outside a few shiny new tech giants.


Quote
The website is less of a "could be changed", but more of a "is the change really really necessary?"
For me that sounds like: nothing is going to be changed!

Very few serious things have been suggested. Only knee-jerk suggestions to copy and paste fashionable setups without any looking into how FPC/Lazarus is organized and what its needs are. Basically redrawing the whole infrastructure on the back of a beercoasters in 5 minutes flat, while dozens work with a multitude of scenarios with the current setup, where processes and setup have matured together for several years. Replacing one piece is already a big undertaking.

Proposals often also conflict, e.g. arguing to use 3rd party services and arguing the very next message that a problem can be fixed with (server) software modification/customization.

It is hard to take such comments seriously, specially in cases when additional prodding doesn't get any more details, and only expands the magnitude of the changes rather than rein them in  (like making a discussion about VCS into an infrastructure of enforcing code styles)

I dared you to come up with a functional mergetracking setup for GIT, and you came back with a quickly looked up command (which didn't solve it) and then quietly sidestepped the issue from then on, and continued on the tired old "it is more popular" "you are old" trajectories. Typical.

Agreed, if you contribute something you should also take care its getting documented properly. Otherwise its useless if nobody knows about it or knows how to use it.

You don't always know the full scope. Major functionality is often created in trajectories of an year and longer, and then some more before it comes into endusers hand.

Like the codeformatting discussion, such oversimplified mantras often don't work. But still I think there could be some improvement there (and there have been in the past, e.g. the userchanges pages). New features should at least be written down anywhere, with some pointers for documenters.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on August 27, 2020, 03:26:01 pm
Quote
Quote from:marcov on August 22, 2020, 07:45:16 pm
 
Quote
Quote from: Fred vS on August 22, 2020, 07:45:16 pm
    But it would be great if somebody (you) take care (administrator) of the fpc GitHub and GitLab mirrored site.
    For example, if somebody does a pull request, you transform this into a svn patch and sent it to fpc-svn.

This won't work. A very large of the patches are tossed back to the users for rewrite/cleanup. This is much harder with push/pull requests and at that point the history is toast anyway.

I was talking about the end-users.
I can understand that people likes the Github or GitLab web sites, very intuitive, with easy "Download as Zip" button, etc...

There is also the great "Issues" feature.

So the solution to have mirrored sites of  svn on Github or GitLab, regularly synchronized, is perfect.

And for all the "Pull-Request" from Github or GitLab, a clear message saying that those sites are only mirrors, that you better sent a patch using the svn server, that those pull-request will be ignored ( but not by asdf1337 who will gently make patches for it and sent it to svn-fpc  ;) ).

Staying with svn mirrored on Github and GitLab simplify all and makes majority of people happy.

Fre;D

Title: Re: The Ranking of Free Pascal in the Tiobe-index
Post by: Kays on August 27, 2020, 05:14:12 pm
[…]
Agreed, if you contribute something you should also take care its getting documented properly. Otherwise its useless if nobody knows about it or knows how to use it.

You don't always know the full scope. Major functionality is often created in trajectories of an year and longer, and then some more before it comes into endusers hand. […]
Title: Re: The Ranking of Free Pascal in the Tiobe-index
Post by: marcov on August 27, 2020, 05:34:53 pm
I didn’t write that but asdf1337 did.

I'm sorry then.

Quote
Well, that’s bad practice. There should be a policy in place that doesn’t allow anything to be released unless it’s properly documented. Merely mentioning it, “some pointers for documenters”, is not sufficient.

It is easy to say, but like with the style stuff very hard to enforce in a volunteer driven effort. In a company such things are easy, you just tell them to and hold them to account, not so in a volunteer project.

Release progress is already too slow to pile on more delaying policies.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Thaddy on August 27, 2020, 05:42:56 pm
And FreePascal is a volunteer project with a very, very high standard of technical documentation. I wish I had the same kind of standard available all the time when I was still working professionally. That does not often happen - read: lacks most of the time. (except in banking)
Reason: professional programmers tend not to like documenting and team leaders find it too costly as long as the code works.

In banking there are usually higher standards, because functional design, technical design and code is part of any audit: you have to... Strange but true.

Of course you will find 1% errors and 1% omissions, but the FPC documentation scores 98% which is much higher than most professionals or volunteers do achieve.
Lazarus is a different question...
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on August 27, 2020, 06:21:49 pm
So the solution to have mirrored sites of  svn on Github or GitLab, regularly synchronized, is perfect.

As said there is a majority of core devels in favour of migrating to GIT, with an own gitlab install. Jonas even managed to get the whole history (CVS+SVN) into GIT, though there are still some issues remaining, that is considered doable.

So it is the practical details of maintaining the fixes branch that are the main problem.

Quote
And for all the "Pull-Request" from Github or GitLab, a clear message saying that those sites are only mirrors, that you better sent a patch using the svn server, that those pull-request will be ignored ( but not by asdf1337 who will gently make patches for it and sent it to svn-fpc  ;) ).

Some devels get starry eyed with pull requests too, so I expect some experimenting. But they will be on our own gitlab instance, not the externally hosted mirrors. Forwarding would be hard since authentication systems would be different.

It is not critical path, so a decision on it is not really a blocker for GIT migration (though having many unusable pull requests pending looks bad).  I'm bringing it up because the constant assumptions that this will make contributing magnitudes easier for non regular contributors now. I doubt that very, very strongly for reasons already stated.   


Quote
Staying with svn mirrored on Github and GitLab simplify all and make majority of people happy.

Or the GIT mirrored on there. But nothing more.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: twolivesbehindschedule on October 02, 2020, 06:37:35 pm
People likes polished stuff, and Pascal was never polished even back then, I gave up on it once due to suggestions by people that now I know that they don't understand the strengths and the weaknesses of any language. I hope this language to survive and get developed and remain unpolished at the same time


EDIT

With the intent to give a tiny bit of help to this community
I will lay here my experience from microsoft support forum for VS where programers get paid well to try to answer any BS of a question I made there... this is also the case for independent forums where I don't know if they get paid or not. Knowing that you people not get paid for answering questions, and solving other people's problems I will not blame you

but I suggest before change the name of any application start by changing your moto

Quote
When you ask a question that is actually answered in the documentation, you are either lazy or a moron.

Of course I will not read every documentation in existence and of course I will ask for convenience for anyone having the time and the mood to answer me, learning on the way...

How many new users have come here asking questions that have been answered somewhere and be considered morons?
Attitude maybe the problem
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Thaddy on October 02, 2020, 08:16:17 pm
Well, if you do not read documentation..... >:D Seriously, what an idiot.
FACTS are the documentation. Opinion is a different matter. Again: what an idiot.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Martin_fr on October 02, 2020, 09:12:56 pm
I will lay here my experience from microsoft support forum for VS where programers get paid well to try to answer any BS of a question I made there... this is also the case for independent forums where I don't know if they get paid or not. Knowing that you people not get paid for answering questions, and solving other people's problems I will not blame you

but I suggest before change the name of any application start by changing your moto

Quote
When you ask a question that is actually answered in the documentation, you are either lazy or a moron.

Of course I will not read every documentation in existence and of course I will ask for convenience for anyone having the time and the mood to answer me, learning on the way...

Well, if you do not read documentation..... >:D Seriously, what an idiot.
FACTS are the documentation. Opinion is a different matter. Again: what an idiot.

Well, there are people here who are not paid (i.e., not paid for anything they do on the forum), and who a nice fellows and will do their best to help anyone with any questions. Including stuff that is documented. At least point you do the specific docs.
The term "help" can have different meanings though. I.e., sometimes its a final solution, other times it is a how to find the answer ("not giving out the fish, but teach how to fish").

And well, there are people here who are not paid and make it their business to throw out insults or insult-like comments. (which some would argue makes them the opposite of the bodily opening generally used for food intake.... ;) ).

You just have to ask, ignore those who are less friendly and use the answers of others.


Quote
but I suggest before change the name of any application start by changing your moto
The forum is a place for user to user conversation.
While you will find some of the developers/team members here, and may get occasional answers from them, the main focus of the forum is that users can converse with other users.
Under that consideration, the forum is not bound to any "moto" of the project.
It is a collection of the combined moto of all our users.

Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: 440bx on October 02, 2020, 10:36:49 pm
Well, if you do not read documentation..... >:D Seriously, what an idiot.
FACTS are the documentation. Opinion is a different matter. Again: what an idiot.
I doubt that gratuitously insulting people leads to desirable results. 

For the record, if you had read the documentation you would have known that boolean operators are not evaluated left to right as you claimed.  Didn't you read the documentation ?
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: trev on October 03, 2020, 05:22:06 am
Well, if you do not read documentation..... >:D Seriously, what an idiot.
FACTS are the documentation. Opinion is a different matter. Again: what an idiot.

You are free to not help, but abusing other forum members is simply going out of your way to be unhelpful which detracts from an otherwise very friendly and helpful community.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: devEric69 on October 03, 2020, 09:37:21 am
If a person claims not to want to read the documentation, so I'm one of the people he sees as "his" documentation.
To put it another way, he takes me - in a certain way - for an idiot, an owed servant.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: dbannon on October 03, 2020, 11:06:42 am
And FreePascal is a volunteer project with a very, very high standard of technical documentation.

From what we keep hearing from Thaddy, he must have a lot of experience in this game. So, somewhat surprising that he thinks FPC and Lazarus have very, very high standards of technical documentation. Its scattered over Wiki, LHelp, PDFs and HTML and, indeed, the forum. Many, many methods are not documented anywhere. We all try to help but its a big and diffuse job.

Thaddy, if paid developers are not documenting their work, its not their fault. Its their supervisors who are negligent. But in a volunteer project, and one that does not attract big business contributions, we cannot force that documenting to happen. Dishing out personal insults to someone who does not understand that will only do harm.

Do you honestly think that OP will go off and say "Wow, Thaddy said I am an idiot, I had better improve my self" ?  Instead, he will go off and tell anyone who asks that the FPC people are rude, abusive and refuse to help anyone. Thanks !

In fact, this forum is usually helpful, polite and accurate.  Most new people's experience is good unless they are unlucky enough to attract the forum troll.

Davo
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: howardpc on October 03, 2020, 12:44:09 pm
Before this thread is derailed entirely, can we hold back from further Thaddy-bashing, and perhaps be a bit forgiving.
With over 10000 posts no one can say Thaddy is not an enthusiastic forum contributor. Yes, he has succeeded in annoying many people, and his current signature characterising most of us as lazy or morons does not help.
Nevertheless, there is goodness and helpfulness among all those many contributions; though when he has a bad day, we all know it, and realise he has issues.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: Fred vS on October 03, 2020, 01:01:51 pm
Quote
can we hold back from further Thaddy-bashing, and perhaps be a bit forgiving.

Yes, of course forgiving is a must.
But to be true, forgiveness requires some rules.

Thaddy knows perfectly Pascal, he is a reference for many things and his skill is highest.

But sometime his comments are very hurtful and not always true.
And in these cases, admitting the mistake and apologizing for it helps a lot to forgive.

All that said, I like you Thaddy (even if sometime I want to bite you) and please continue to contribute (respectfully).

Fre;D




Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: twolivesbehindschedule on October 03, 2020, 05:04:23 pm
If a person claims not to want to read the documentation, so I'm one of the people he sees as "his" documentation.
To put it another way, he takes me - in a certain way - for an idiot, an owed servant.

For my part, I do give info when I have it and see someone else seeking for it, especially when that info have been given to me freely by someone else before... unless someone asking you to write for him a whole program. I suggest not came hard with common knowledge that someone else lack

Learn to separate common knowledge from your own intellectual property and give the former freely fast and without second thought. You help your own self that way too

EDIT

Dishing out personal insults to someone who does not understand that will only do harm.

Do you honestly think that OP will go off and say "Wow, Thaddy said I am an idiot, I had better improve my self" ?  Instead, he will go off and tell anyone who asks that the FPC people are rude, abusive and refuse to help anyone. Thanks !

I would't do that, thick skin is hard to feel abused.
My comment is more helpful than harmful,  I only want to point what you are competing with, not to send people there.

And perhaps I must mention that microsoft picks up all your code from VS for research and debugging purposes, and the only assurance that the user have is that the top on their field and the most well paid programmers in the world will not lay their eyes on someone's code... For some this is good enough for others it's not.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: devEric69 on October 03, 2020, 08:06:04 pm
Quote
Before this thread is derailed entirely [snip]

Indeed: seriously, has anyone heard from the person who opened this thread: he is still not satisfied with these answers ^^ ?


Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: simone on November 05, 2020, 08:09:05 pm
While I've never trusted Tiobe, I'm glad to see that Delphi/Object Pascal is again in the top 20 this month.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: lucamar on November 05, 2020, 08:54:22 pm
While I've never trusted Tiobe, I'm glad to see that Delphi/Object Pascal is again in the top 20 this month.

That's not surprising; if you look at the historic (https://www.tiobe.com/tiobe-index/delphi-object-pascal/) you'll see that it varies somewhat wildly, even from month to month. TIOBE's measuring (at least on some languages) doesn't really reflect the real world. You have but to see the swings that SQL traverse: nothing to do with how widely SQL is really used :-\

One has to take TIOBE with a pinch of salt and use it only to check tendencies, if anything at all.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: simone on November 05, 2020, 10:21:16 pm
I agree with you. I pointed out to the CEO that this instability of the indices demonstrates the unreliability of the methodology, as in the real world the existence of large code bases makes changes very slow.

Moreover in the last few months I have seen hilarious results, eg Logo in the top 20. Frankly unrealistic (although I love this language, which I used when I was a child).
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: 440bx on November 05, 2020, 11:28:37 pm
...this instability of the indices demonstrates the unreliability of the methodology, as in the real world the existence of large code bases makes changes very slow.
That is likely the best indicator that their index isn't resting on a methodology with a solid scientific foundation.  It's not even clear what it is they are measuring (or attempting to measure).

If they wanted to measure the use of the Pascal language, they should take into account all the Pascal dialects, then possibly as additional information, show a breakdown for each dialect.  They should do that not just with Pascal but, for every programming language.


Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: dtoffe on November 06, 2020, 02:01:35 am
That is likely the best indicator that their index isn't resting on a methodology with a solid scientific foundation.  It's not even clear what it is they are measuring (or attempting to measure).

From the TIOBE Index Wikipedia page at https://en.wikipedia.org/wiki/TIOBE_index (https://en.wikipedia.org/wiki/TIOBE_index)

Quote
The index covers searches in Google, Google Blogs, MSN, Yahoo!, Baidu, Wikipedia and YouTube. The index is updated once a month. The current information is free, but the long-term statistical data is for sale. The index authors have stated that it may be valuable when making various strategic decisions.

It doesn't need to be particularly scientific, it just need to be convincing enough to business people. If you are a coder or a mid-level tech manager, it will probably make your life miserable.

Daniel
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: dbannon on November 06, 2020, 06:23:57 am
And, as I have said before, the real problem is this Forum, its so good that we don't need to hit Google ....

Maybe we should set the Forum Search engine to use Google ?

Davo
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on November 06, 2020, 09:24:22 am
Maybe we should set the Forum Search engine to use Google ?

That wouldn't help, because Tiobe looks for "pascal xxx" or "object pascal xxx" and things like that and if you have a specific forum search through Google you wouldn't need to add those terms when searching the forum.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: simone on November 06, 2020, 12:39:54 pm
One of Tiobe's problems is that it relies on the wikipedia definitions. We all know that technically Delphi is the IDE and Object Pascal is the programming language, but it is also true that the term Delphi is also often used for the language. Delphi's wikipedia page is changed often. When wikipedia defines Delphi as an IDE, Tiobe no longer considers the term Delphi and the popularity of the language is consequently underestimated, as the name Object Pascal is less well known. In IT, I have many friends and colleagues who have heard of Delphi, but hardly anyone knows what Object Pascal is. Even on the web, many sites dedicated to Delphi or Lazarus do not use the term Object Pascal. Let me say that popularity of this programming language is also (but not limited to) a communication problem ...
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: asdf1337 on February 13, 2021, 01:00:28 pm
So the solution to have mirrored sites of  svn on Github or GitLab, regularly synchronized, is perfect.

As said there is a majority of core devels in favour of migrating to GIT, with an own gitlab install. Jonas even managed to get the whole history (CVS+SVN) into GIT, though there are still some issues remaining, that is considered doable.

Maybe consider using https://codeberg.org/ instead of hosting something yourself? Located in EU and reduces the amount of work with keeping all the software up to date to avoid security holes and exposing of private user data through that... %)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: egsuh on February 13, 2021, 02:49:30 pm
 From the perspective of brand strategy consultant, Delphi/.... Pascal/Lazarus + similarities should name themselves as "Pascalite", pursuants of core programming philosophy.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: CoderSI on February 13, 2021, 03:21:35 pm
Why is the keyword FreePascal not present in the Tiobe index (in the definitions):
https://www.tiobe.com/tiobe-index/programming-languages-definition/
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on February 13, 2021, 09:58:35 pm
So the solution to have mirrored sites of  svn on Github or GitLab, regularly synchronized, is perfect.

As said there is a majority of core devels in favour of migrating to GIT, with an own gitlab install. Jonas even managed to get the whole history (CVS+SVN) into GIT, though there are still some issues remaining, that is considered doable.

Maybe consider using https://codeberg.org/ instead of hosting something yourself? Located in EU and reduces the amount of work with keeping all the software up to date to avoid security holes and exposing of private user data through that... %)

Not. Going. To. Happen. We're continuing to do private hosting and that is not negotiatable.

Why is the keyword FreePascal not present in the Tiobe index (in the definitions):
https://www.tiobe.com/tiobe-index/programming-languages-definition/

Because the Tiobe people have no clue.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: marcov on February 13, 2021, 10:15:47 pm
From the perspective of brand strategy consultant, Delphi/.... Pascal/Lazarus + similarities should name themselves as "Pascalite", pursuants of core programming philosophy.

Bob Swart named his daughter Delphine, does that count ?
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: trev on February 13, 2021, 10:51:18 pm
Why is the keyword FreePascal not present in the Tiobe index (in the definitions):
https://www.tiobe.com/tiobe-index/programming-languages-definition/

Surely it's Free Pascal and not FreePascal? Yep, from the freepascal.org website: "Free Pascal is a mature...".
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: korba812 on February 13, 2021, 11:18:09 pm
Bob Swart named his daughter Delphine
Wow. Just like a celebrity.
Because the Tiobe people have no clue.
...or they don't care. Their business is selling own products, not promoting FreePascal.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: egsuh on February 14, 2021, 03:57:55 am
Quote
Bob Swart named his daughter Delphine, does that count ?

Hope she will marry someone with family name of Pascal.
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: VTwin on April 06, 2021, 01:51:00 am
If you care, Delphi/Object Pascal moved up on Tiobe from 22 to 12 this past year in April  :).
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: PascalDragon on April 06, 2021, 09:20:04 am
Which means they probably adjusted their search algorithm yet again... ::)
Title: Re: The Ranking of Free Pascal in the Tiobe-indexl
Post by: simone on April 06, 2021, 11:21:13 am
Object Pascal has been rising for a few months in Tiobe. The amplitude of the oscillations shows that the methodology is completely flawed. Seeing Logo in the top twenty a few months ago was hilarious. However, given the widespread overestimation of Tiobe, I am pleased to see Object Pascal in a good position again. It improves the reputation of the language and, perhaps, raises some doubts to those who consider it dead for decades.
TinyPortal © 2005-2018