Lazarus

Miscellaneous => Suggestions => Topic started by: seghele0 on March 01, 2021, 10:59:46 am

Title: More componenets.
Post by: seghele0 on March 01, 2021, 10:59:46 am
Why doesn't Lazarus make two installation options, like CodeTyphon, which is a standard installation(as current) and a large installation with all available components?
This option would also enrich the knowledge of not often used components.
 ;)

Title: Re: More componenets.
Post by: Handoko on March 01, 2021, 11:21:06 am
The more components installed on Lazarus the more time and memory is needed to rebuild the IDE. I rebuild the IDE pretty often.

Lazarus already has OPM and I personally dislike bloatware. If you can inspect my computer, you will see there isn't any unused programs here. I also removed all the unnecessary 'things' that come from the default OS installation.  My computer is old and doesn't have the latest technology but it's still slim and sexy, running like a cheetah.
Title: Re: More componenets.
Post by: Martin_fr on March 01, 2021, 01:16:24 pm
Well that is what the online-package-manager is for.

Much more flexible. In the old days there were websites for Delphi, with hundreds of packages. You could chose between a dozen components doing almost the same, differing only in small details.
Leaving out the fact that their are currently not as many people contributing, and not as many packages available... It would not be practical to include several near identical components. Yet with the package-manager that is no problem.
And the package manager makes it easier for contributors to have their packages included.

Of course packages in the OPM are as provided by the contributor. We (the Lazarus team) do simple not have the resources to test every single package for every release we make.
If we included them in the main download, we still would not have that manpower. So we would potentially end up shipping broken packages. And while the OPM allows the contributor to submit a working version asap, if included it would be broken until the next release (and then removed, if no fix was contributed).

Here is the difference between CodeTyphon and Lazarus. We (Lazarus) spent our time on the IDE/LCL itself (and that already requires all of our time and more).
Codetyphon presumingly does not need to spent much time on development the IDE itself.  They use what we deliver. So they have time to spent on bundling packages.



So the question here is, do you simply want an all-inclusive download, or do you want the same amount of packages, but with more maintenance done? (i.e. Regular complete tests of all packages on OPM)?

Because maintenance, testing, making sure all version dependencies are accurate, etc.... All that is not a question of "inclusive download" vs "OPM".
That is a question of having someone (or even several people) who want and can spent the time.

If someone volunteers to do that, then it can all be done for OPM.

Same about having more packages on OPM => contacting authors, helping them.... Question of someone doing that.



As for visibility....

If all components where pre-installed, people would not see the forest for the trees.

Of course OPM itself could be flaunted more prominently (as could many features).
Currently it is
- in the menu
- in the dropdown, at the right side of the component palette.

I am not sure if it should have its own tab in the palette (imho not). But it could/should probably be interacting with the "Components" window (palette, right click "Show all"). It could have an option (checkbox), so that the "filter" would search online too. And a button to open the OPM.

If someone wants to work on that....


Title: Re: More componenets.
Post by: Martin_fr on March 01, 2021, 01:24:32 pm
And if this is about "needing to rebuild the IDE" to install a package....
Well, IMHO not a big deal, but opinions on this may of course differ.

Afaik there is work under way to have "loadable packages" that do not require a rebuild of the IDE.
However, I do not know the current progress, and it will almost definitely still be a good amount of time. (I.e. I guess the very earliest would be with Fpc 3.4.0, but even that is not guaranteed / And 3.4.0 is far away)
Title: Re: More componenets.
Post by: wp on March 01, 2021, 02:45:03 pm
I am against deploying Lazarus with "all available components" because it would put the burdon of maintaining "all components" on the Lazarus team. Having an unmaintained third-party component poses a large risk on Lazarus such that it could not compile any more when the component is broken by some changes in Laz or FPC. Even the current situation of having barely maintained component in the Online-Package-Manager is questionable because it puts an enormous amount of work on its inventor, GetMem. But having such components in the official product is even worse because a single incompatible component can stop Lazarus from working.
Title: Re: More componenets.
Post by: Handoko on March 01, 2021, 03:06:08 pm
Having a big fat installer that includes all components has both advantages and disadvantages. Sure, some will like it and it really helps new users to 'see' all the available components without need to search and install them.

But managing a huge amount of components isn't easy. CodeTyphon makes it easier by taking over the codes, maintaining and developing them themselves. That explains why CT wants to make them their own version of the components.

But 'disconnecting' the codes from their original authors can be bad, unless the they really have a lot of skilled programmers in wide range of fields. Usually programmers are good in graphics programming are not so good in database programming. Programmers are skilled in network and internet may not as good for writing game engine. So the development of components should left for their original authors unless it has been abandoned.

Perhaps the manpower issue can be solved and the needing to rebuild the IDE no longer a problem. But how often should we release the big installer? Once a month, once a week or everyday? Each components are are improving, having bug fixed and new features added. Should we immediate release the big installer once one of the component release a new version? I don't like Firefox rapid releasing strategy (https://en.wikipedia.org/wiki/Firefox_version_history#Rapid_releases) and I hate downloading huge files.

I do not against  big installer concept but I just don't like it.
Title: Re: More componenets.
Post by: alaa123456789 on March 01, 2021, 03:14:28 pm
Afaik there is work under way to have "loadable packages" that do not require a rebuild of the IDE.
However, I do not know the current progress, and it will almost definitely still be a good amount of time. (I.e. I guess the very earliest would be with Fpc 3.4.0, but even that is not guaranteed / And 3.4.0 is far away)
i think OPM is good choice at the moment and it could be better if installing without rebuilding the IDE , which you (Lazarus teams) are working on wish you the best to make it sooner.

thanks
Title: Re: More componenets.
Post by: JuhaManninen on March 01, 2021, 03:16:15 pm
Why doesn't Lazarus make two installation options, like CodeTyphon, which is a standard installation(as current) and a large installation with all available components?
This option would also enrich the knowledge of not often used components.
You can easily install all available components from OPM. It rebuilds Lazarus IDE once and there you go.
Did you try it?

i think OPM is good choice at the moment and it could be better if installing without rebuilding the IDE , which you (Lazarus teams) are working on wish you the best to make it sooner.
Dynamic packages are overrated. Building the IDE is very easy.
Dynamic library interface also adds a layer of complexity.
Title: Re: More componenets.
Post by: Handoko on March 01, 2021, 03:32:15 pm
Building the IDE is very easy.

For someone you don't how to rebuilt the IDE, here is how to do it:
Alt + T > Build Lazarus with ... > Answer: Yes

If for some certain reasons it does not work on your computer, try this:
Lazarus main menu > Build Lazarus with ... > Answer: Yes

And if you install a component using OPM, you will be asked to rebuild the IDE, you just need to answer Yes and it will be performed automatically for you.

Dynamic packages are overrated ...
Dynamic library interface also adds a layer of complexity.

If I remember correctly, I heard dynamic packages feature may slow down the IDE a bit. If that so, I prefer no dynamic packages at all.
Title: Re: More componenets.
Post by: funlw65 on March 01, 2021, 05:58:51 pm
When finally a decision will be made regarding which GUI to be made/used, then there will be rock solid stability. Then, the freepascal compiler should achieve that too - and then we will have long term, fewer releases.

Only then you will have long lasting OPM packages with no compilation problems. As it is now, you might have a complete installation of your desired packages, then after an update, you will have to start again from a clean, standard installation (you'll bless developers for sticking with that standard set of packages - and, if you develop an open source, you have to always consider that standard set of packages) as some of OPM packages might be broken.   
Title: Re: More componenets.
Post by: Martin_fr on March 01, 2021, 06:36:43 pm
When finally a decision will be made regarding which GUI to be made/used, then there will be rock solid stability.
What you mean by "GUI"?
To me that would mean Win-API vs Gtk,Qt,Cocoa? But that should affect only very few of the components?

But never mind, there will always be some amount of maintenance required. Happens with any open source package manager (or rather anything where packages/add-ons are contributed).  Node's npm, Perl's cpan, whatever the other languages call it.  ... Firefox add-ons (probably Chrome too?).
Of course if you look at something like npm, with myriads of contributors, there is either a patch readily available (but you have to apply it), or a fork or an alternative package....

Have you ever tried building gdb on Windows? You need either Cygwin or msys and mingw. And then you can start searching for all the right packages, and the right versions, and maybe patches to gdb sources. And usually the error comes a half an hour into the build. => Install a package in Lazarus, you will know in 2 or 3 minutes if it builds.

And packages for components will need updates, every time any OS/Framework changes its API. Nothing that can be done about it. As soon as any distribution changes the way they build/provide gtk, amendments may be needed.



In the end however the problem of having lots of pre-tested packages available would not be solved by shipping them in a big installer.

The reason that are 3rd party packages is, that the Lazarus team does not have enough people/time to develop/maintain all of them.
If we were putting them into a big installer, they would come with the same problems.

In conclusion, what we need is a volunteer (or several), who agree to do the maintenance work.
That is they would have too:
- regularly test all 3rd party packages against the Lazarus svn trunk => inform developers about needed changes, help them to fix it.
- 2 weeks before a release, update the OPM database. Update the compatibly list. Edit which package version to use with which Lazarus.
- Scour the web for new packages to be include. Contact the authors
- Help any author in adding their packages.
- Any other maintenance.
- Do that for all target os/wigetset: Mac,Linux,WinXP-Win10 / 32/64bit / QT or Win/Gtk/Cocoa / latest fpc, and previous fpc // in all combinations

This person we do not have. So if anyone wants to volunteer....

This person would also be required, if (subjunctive) we were to ship it all as an installer.
Because before building such an all-in-one installer you need a working set of components. And then you need someone to run the build script....
Title: Re: More componenets.
Post by: trev on March 02, 2021, 12:07:31 am
Why doesn't Lazarus make two installation options, like CodeTyphon...

Because it's not CodeTyphon maybe?

I think the FreeBSD Forum moderators put it best:

Quote
As of today, FreeBSD Forums staff will actively close down (and eventually remove) topics that serve no other purpose than to complain that "FreeBSD is not (like) Linux" (or Windows, or macOS, or any other operating system), or that "FreeBSD does not use systemd", or that "FreeBSD has no default GUI", or that "FreeBSD does not encrypt gremlins", etc. This also includes topics that devolve into that kind of debate.

Note that this is a general user and administrator forum, where the community aims to assist those who want to install, run, or upgrade FreeBSD as-is. Discussions about what FreeBSD needs to be, or needs to add, or needs to lose, are pointless on the forums. We do not maintain the operating system here.

Those discussions are best held on the FreeBSD mailing lists, where active development and maintenance of the operating system are discussed and performed in excruciating detail. If you do not know which mailing list to use, start at FreeBSD Questions. Do note that "Why is FreeBSD not more like Linux?" topics that do not go into specifics and serve no other purpose than to complain will not be welcomed there either.
.

I dare say the same could apply here.
Title: Re: More componenets.
Post by: JuhaManninen on March 02, 2021, 12:42:39 am
When finally a decision will be made regarding which GUI to be made/used, ...
What are you talking about?
Quote
... then there will be rock solid stability.
Do you plan to fix all bugs to achieve rock solid stability? Good! Thanks in advance.

Because it's not CodeTyphon maybe?
Exactly.
It is a conscious strategy to keep Lazarus distribution slim and light.
In principle only components needed for Delphi compatibility or the IDE itself are included, although there are some exceptions.
Online package installation is essential in this strategy.
I don't understand why a rarely used component should be included in the distribution when it can easily be installed online. It would be dummy to include it.
Title: Re: More componenets.
Post by: Martin_fr on March 02, 2021, 12:55:08 am
I dare say the same could apply here.

Well, I would not go quite as far. Yes it is true, if you want to discuss a new idea with the actual core team, the mailing list may be the better place. But that is, if you actually seek to contribute yourself. Do some work yourself. And even then, in most cases you find that you will get responses on the forum.

New ideas are fine. Occasionally core devels have picked up one from the forum too (I know I have). But most of the cases, even good ideas will need someone to contribute.
That said, your contribution can be to lobby for some other user to implement your idea and contribute it. I don't recall having seen that successful (unless accompanied by a bounty).
But well yes, lobbying is fine too. (So long as you manage your expectation accordingly).

Anyway, the task has been outlined. Now, if someone shows up to whom the shoe fits...

Title: Re: More componenets.
Post by: Martin_fr on March 02, 2021, 01:04:27 am
Also include-all => still means you have to test each component, and without description may overlook plenty of its features.

On the other hand, if a maintainer (team) for OPM evolves, and works on the presentation...
https://packages.lazarus-ide.org/
- Add images.
- Make metadata like Catecgory, Widgetset, ... avail on the web - like it is in OPM.
...

That could be a much better mean to explore...

And yes, make sure people know about it.
Title: Re: More componenets.
Post by: ASBzone on March 02, 2021, 02:44:18 am
I think the FreeBSD Forum moderators put it best:
;D
Title: Re: More componenets.
Post by: funlw65 on March 02, 2021, 04:41:53 am
...
What are you talking about?
...

On what foundation is CustomDrawn based/built?
Is there a blueprint with the schematic of what is desired? How it should work? What was achieved until now? You know, to encourage contribution...

But first of all, in which percentage is CustomDrawn as universal GUI supported/promoted by today's developers and "important users" (if there are such users)?
Title: Re: More componenets.
Post by: avra on March 02, 2021, 08:42:17 am
Why doesn't Lazarus make two installation options, like CodeTyphon, which is a standard installation(as current) and a large installation with all available components?
I have made ct4laz project which aims at providing CodeTyphon unique components to Lazarus users. You can find those components in OPM and they all have pl_ prefix. Bringing them into default installation would make Lazarus download be almost 1GB, and also make users having less then 2GB RAM unable to compile Lazarus - and no one would like that. So, for several reasons there is no need to bloat Lazarus with something that can be installed with a few clicks.
Title: Re: More componenets.
Post by: JuhaManninen on March 02, 2021, 11:25:03 am
On what foundation is CustomDrawn based/built?
Is there a blueprint with the schematic of what is desired? How it should work? What was achieved until now?
It was started by Felipe years ago. He implemented it quite quickly up to its current state but recently he has not been very active.
I don't think there is a blueprint or a schematic. You must study the sources and ask detailed questions.
It should work well but unfortunately it does not. It is alpha state.
To see what was achieved until now, please test and study it.

Quote
You know, to encourage contribution...
It is funny how new users want to encourage contribution from other people.
"If only you explained things well then contributions would pour in!" The reality is different.

Quote
But first of all, in which percentage is CustomDrawn as universal GUI supported/promoted by today's developers and "important users" (if there are such users)?
"CustomDrawn as universal GUI"?
LCL binds to native widgetsets on different platforms. GTKx and QTx are considered native on many Unix related systems.
CustomDrawn would be nice especially for embedded and low-end systems but currently it does not work well.
The percentage of CustomDrawn promotion is low. Somebody should improve it and then promote.
Title: Re: More componenets.
Post by: seghele0 on March 02, 2021, 11:34:33 am
May I thank everyone for the answers.
I am surprised that there are still people working with a very old PC, a new one costs barely 500 euros. Maybe there are still people who don't know we're already in the year 2021.
A doable scenario is to release a new version of a big Lazarus installer, after an update of the FPC, which is not every year.
In addition to the special release, the current small Lazarus can continue as current if this were still necessary.
If there is an opinion that it is rediculous to make a special Lazarus version with a big installation of components, then this means that CodeTyphon has made an unnecessary move with there  extended version.
Are the makers of CT really stupid?
Conclusion:
Perhaps a collaboration with CT is a solution to make a better and more efficient product for everyone, young and old.
I hope that it will be thoroughly considered before throwing everything in the trash can.

 ;)
Title: Re: More componenets.
Post by: JuhaManninen on March 02, 2021, 11:53:46 am
May I thank everyone for the answers.
I am surprised that there are still people working with a very old PC, a new one costs barely 500 euros. Maybe there are still people who don't know we're already in the year 2021.
Uhhh... Do you mean people should spend 500€ for a new computer only because you want to have bloated SW?

Quote
A doable scenario is to release a new version of a big Lazarus installer, after an update of the FPC, which is not every year.
In addition to the special release, the current small Lazarus can continue as current if this were still necessary.
Somehow you did not understand the answers. Lazarus strategy is to use a slim distribution with easy installation of online packages.
However you or anybody can create a competing installation with all packages included and installed. This is open source with GPL.

Quote
If there is an opinion that it is rediculous to make a special Lazarus version with a big installation of components, then this means that CodeTyphon has made an unnecessary move with there  extended version.
Are the makers of CT really stupid?
CT has a different strategy. It is important to differentiate, otherwise it would be a simple copy.
Clearly CT is the right choice for you. Why do you want to turn Lazarus into CT? Please read the answers you got again.

Quote
Conclusion:
Perhaps a collaboration with CT is a solution to make a better and more efficient product for everyone, young and old.
I hope that it will be thoroughly considered before throwing everything in the trash can.
Huh, you don't know the history.
I am happy CT exists and I am happy they have different strategy.
I just don't understand why you come here to complain that Lazarus is not CT.
Title: Re: More componenets.
Post by: Handoko on March 02, 2021, 12:04:56 pm
Are the makers of CT really stupid?

I really appreciate and understand why you want a big installer. But it seems you didn't 'really' read the answers. It's not about stupidity nor the advantages of a big installer. It is about manpower. Lazarus/FPC is maintained by volunteers all over the world, I'm sorry but I would say not-very organized volunteers. CT is maintained by a team, in a uni or a research center or a building or whatever I don't know. They may have a team leader or clear organization structure. And they might get paid monthly for doing it.

If there is an opinion that it is rediculous to make a special Lazarus version with a big installation of components

No. Your suggestion is a good suggestion as I already mentioned some users will like it. But the benefits clearly do not outweigh the effort, unless there is volunteer for it.

Perhaps a collaboration with CT is a solution to make a better and more efficient product for everyone, young and old.

It is a good suggestion. You can try to contact them if you want. But I'm afraid to tell you, you won't get any good result. If you follow the story long ago, you will know there were some not-good experience between CT and Lazarus. On their forum, if someone 'smells' too-Lazarusish, they will ban the user.
Title: Re: More componenets.
Post by: funlw65 on March 02, 2021, 12:25:45 pm
...
If there is an opinion that it is rediculous to make a special Lazarus version with a big installation of components, then this means that CodeTyphon has made an unnecessary move with there  extended version.
Are the makers of CT really stupid?
Conclusion:
Perhaps a collaboration with CT is a solution to make a better and more efficient product for everyone, young and old.
I hope that it will be thoroughly considered before throwing everything in the trash can. ;)

I would ask you, what is the real reason you can't use CT? Better saying, "avoiding" CT...
Title: Re: More componenets.
Post by: Handoko on March 02, 2021, 12:50:36 pm
I am surprised that there are still people working with a very old PC, a new one costs barely 500 euros. Maybe there are still people who don't know we're already in the year 2021.

You mentioned Euros, I guess you're from a developed country. I'm from Indonesia, a developing country. I'm a computer technician, most of my clients are sohos. I haven't made a survey but I can roughly to tell you:

- 60% CPU I handled recently were Core2 Duo
- 40% CPU I handled recently were Core i3 and above
- 50% computers I handled recently were having 2GB of memory
- 30% computers I handled recently were having 4GB of memory
Title: Re: More componenets.
Post by: seghele0 on March 02, 2021, 02:10:12 pm
Apparently, the spirit of negativism is great about this proposal.
I will continue to use CT and retain Lazarus as a reference point, because of this very well-functioning forum.
I would like to thank everyone for their personal opinions and fully respect them.
About the 500 euro, I had not taken account of countries outside Europe, sorry and sorry.
I would like to thank everyone for their personal opinions and fully respect them.
 ;)
Title: Re: More componenets.
Post by: Martin_fr on March 02, 2021, 03:23:09 pm
Apparently, the spirit of negativism is great about this proposal.
It's based on our experiences. It's not against the idea itself. But about the requirements it would "burden" on some of us.

Let me tell you. I am (among other things) the person who builds the installers for Windows.  I do that in my personal spare time.
I joined the team for other reasons, building installers is not my idea of having a good time. But someone has to do it....

I have no problem if there is an all-include installer (in addition to what we have).
But:
- I have neither time nor interest to build it.
- I have no time, to manage the bug reports about 3rd party components that will then end up on our bug tracker
   (And I can tell you neither has anyone else in the team)

So for that reason I will not build it. And for that reason no one else in the team will....

So we do need someone who does.

But I am willing to make a peace offering:
- I will run the build script (Windows) for such an installer, and upload it (I cant guarantee if it will be the main download site, or an alternative side).
  That way people would have an installer build from a trusted source.

But I will only do that, if there is a person who does all the other work.
- provide the build script (it must run to completion without interaction)
- test it before I run it, to make sure there are no build errors
- test the resulting installer.
- answer any relevant bug in our bugtracker
.....

This is what I personally am willing to do, and the requirements for me doing so. After all, its my personal spare time.

I would still have to run it past the rest of the team. But at least an upload to a separate sourceforge page should be possible.
Title: Re: More componenets.
Post by: lucamar on March 02, 2021, 04:23:17 pm
About the 500 euro, I had not taken account of countries outside Europe, sorry and sorry.

It's not only outside Europe: we (from Spain) have about a 10% customers (including some from the countries around) with Pentiums and lower (down to 8 bits machines), which means we need to have (and use) them too to support them. There are many reasons why people cling to old hardware but want "new" software ... and some of them even make sense ;D

I'm not exactly poor and I'd have to think long and deep on the necessity before expending 500 eur in a new machine: my current "main" one (which should last me another ten years) is a 200 eur. laptop with which I'm quite happy. :)
Title: Re: More componenets.
Post by: ASBzone on March 02, 2021, 05:13:47 pm
Apparently, the spirit of negativism is great about this proposal.

Disagreement is not automatically synonymous with negativity.

Especially when it is well articulated disagreement, based on documented and verifiable experience.

Optimism is nice and desirable, until it insists on ignoring objective information about risks, challenges and resources.   Then it becomes wild-eyed utopianism.

BTW, you aren't offering a proposal, because a proposal does not consist 99%+ of what others must do for your benefit.  What you are doing is making a feature request.
Title: Re: More componenets.
Post by: seghele0 on March 02, 2021, 06:23:10 pm
Martin_fr,
Thank you for the info, the future will tell........
 :)
Title: Re: More componenets.
Post by: GetMem on March 02, 2021, 07:22:55 pm
Realistically speaking, how many third party package does an average developer use? Ten, twenty? I certainly don't need more the five. Even if manpower is not an issue, what is the purpose of a super big IDE? I'm not against it, just trying to understand the reason behind it.

@seghele0
Are you willing to invest some of your free time in creating a big ide? 
Title: Re: More componenets.
Post by: 440bx on March 02, 2021, 08:17:38 pm
Apparently, the spirit of negativism is great about this proposal.
You have to take into consideration that not everyone has a desire or use for "kitchen sink" installations.   

Personally, my ideal piece of software is something that is the absolute minimum and allows me to add to it whatever features/functions I need as the need for them arises.

The following isn't a complain about Lazarus but, for me, it already has several hundred times more than I need.  I use it only as a front end to FPC and GDB and, that is _all_ I want (at least at this time.)    The thought of having a big download just to get those basic features is rather unappealing.

Bottom line : different users have different needs. 

I like software that offers features/functions "a la carte". 

IOW, waitress, I'd like to have 2 eggs over easy, two strips of bacon, hash browns and a large cup of coffee, thank you and, hold off on the 20 pork chops, 7 prime ribs,  11 lobsters, 8 chicken cordon blue, 5 trout amandines and whatever else is on the menu today.

Title: Re: More componenets.
Post by: seghele0 on March 03, 2021, 10:15:43 am
GetMem,
Quote
I'm not against it, just trying to understand the reason behind it.
It might be a good idea to contact CT asking why they're doing it.
They can give you a well-structured answer about the "big" IDE.
 ;)
My personal knowledge in programming is of a 'very' low level.
It only has a private purpose to keep my brain active.
 8)


Title: Re: More componenets.
Post by: Handoko on March 03, 2021, 10:56:50 am
Some new users including me, may have problems installing third party components. Long ago, the first time I tried BGRABitmap, I really had to scratch my head. After many attempts and searching information on the forum, I finally could install it.

So, I can think of one reason: it can prevent baldness.  :D
See the photo beside this posts, I already lost many hairs.
 
But, after the inclusion of Online Package Manager. Installing a third party component is as easy as some mouse clicks. Unfortunately, not many newbies know the existence of OPM.
Title: Re: More componenets.
Post by: GetMem on March 03, 2021, 12:10:27 pm
@seghele0
Quote
It might be a good idea to contact CT asking why they're doing it.
They can give you a well-structured answer about the "big" IDE.
I'm interested more in your opinion. So why a super big IDE is a good idea?

@Handoko
Quote
Some new users including me, may have problems installing third party components. Long ago, the first time I tried BGRABitmap, I really had to scratch my head. After many attempts and searching information on the forum, I finally could install it.
That's a good one! Still the OPM approach seams a better solution to me. You can install individual packages in a reasonable amount of time.
Title: Re: More componenets.
Post by: wp on March 03, 2021, 02:05:59 pm
Unfortunately, not many newbies know the existence of OPM.
We could add a note while the installer runs, like there is one for "Why are the binaries so big?". Of course, nobody reads this, but at least nobody can blame the team for not telling.
Title: Re: More componenets.
Post by: Handoko on March 03, 2021, 02:29:09 pm
Unfortunately, not many newbies know the existence of OPM.
We could add a note while the installer runs ...

+1
Yes, we should provide some FAQ and do some promos on the installation screen.

Also I think it is a good idea to mention OPM on the Lazarus main page, in 'highlights' section:
https://www.lazarus-ide.org/ (https://www.lazarus-ide.org/)

We could also mention: Develop Android apps easily using LAMW.
Yes, it is definitely easier than using C++ and some others.

OPM not only shows a list of available components, it also displays a short description and the license information of the component. This is what I like most. All the components enabled on Lazarus default installation are license in LGPL with linking exception, that is good. But installing third party components without checking their license term is bad, unless the programmer does not distribute the program. But who wants to write program without distributing it?
Title: Re: More componenets.
Post by: Martin_fr on March 03, 2021, 06:36:28 pm
It might be a good idea to contact CT asking why they're doing it.

Feel free to contact them, if they want to contribute to our work.
(But even then, it needs to be adapted so it could be used without imposing any extra work on the existing team. / The existing team does not have the time.)

From what I can tell - and a lot of people here seem to share that observation - they have made it quite clear, that not only do they not want to contribute back, but more so that they are willing to go to any length of effort to distance themself from us.
This is my personal impression. It could be totally wrong. => and if it correct, they have every right to do such things.




Unfortunately, not many newbies know the existence of OPM.
We could add a note while the installer runs ...

+1
Yes, we should provide some FAQ and do some promos on the installation screen.
Patches welcome.

Also, as I already posted: OPM could be more prominent in the component pallete/window.

Quote
Also I think it is a good idea to mention OPM on the Lazarus main page, in 'highlights' section:
https://www.lazarus-ide.org/ (https://www.lazarus-ide.org/)
+1

Quote
All the components enabled on Lazarus default installation are license in LGPL with linking exception, that is good.
Sure about that?
Afaik SynEdit is Gpl/Mpl. Haven't checked all the rest.
Title: Re: More componenets.
Post by: Handoko on March 03, 2021, 06:58:02 pm
Quote
All the components enabled on Lazarus default installation are license in LGPL with linking exception, that is good.
Sure about that?
Afaik SynEdit is Gpl/Mpl. Haven't checked all the rest.
I was wrong, I just read the FAQ again. I misinterpreted the words. Sorry.
Title: Re: More componenets.
Post by: ASBzone on March 03, 2021, 07:40:10 pm
GetMem,
Quote
I'm not against it, just trying to understand the reason behind it.
It might be a good idea to contact CT asking why they're doing it.

But since you are suggesting it, you should have the reason for it.

If you don't have a reason for it, then why bring it up as a suggestion?

And if someone else is already doing exactly what you are willing to suggest (for no reason), then there is no problem -- the solution is already present.  Just use it.

IOW, if you are presenting something as a problem in need of being resolved -- especially when it is going to lead to other people doing a ton of work -- then you ought to be able to articulate why.
Title: Re: More componenets.
Post by: JuhaManninen on March 03, 2021, 08:02:43 pm
@seghele0
I'm interested more in your opinion. So why a super big IDE is a good idea?
... and a related question: What problems you had with OPM? Why does it not work for you?
The only reason I can think of for people not wanting an online package system is a lack of internet connection.
Are you not connected to internet? You can post here somehow.
Even then you need a connection only once to install all packages from OPM. Then they stay installed in your IDE.
BTW, you need internet also to download any program installation unless you order a CD or USB stick or something.
Most computer owners nowadays have internet connection. I wonder what is your ultimate motive for not using OPM.
Title: Re: More componenets.
Post by: alaa123456789 on March 03, 2021, 08:18:05 pm
it is good to have more components , but the problem there is shortage of information how to use them  ,
we thank all developers but need more work on help section of each component.

thanks
Title: Re: More componenets.
Post by: seghele0 on March 04, 2021, 03:07:14 pm
 :(
I didn't want any fierce debates.
This topic was just about a question without pressure.
I’ve all respect for the creators/developpers of Lazarus and for  the enormous time they put into this project with high qualities.
In the current context, I’ll stop to comment further.
The case is clear, Lazarus remains Lazarus and CT remains CT.
Thank you all for your opinion.
The end.
Title: Re: More componenets.
Post by: GetMem on March 04, 2021, 03:19:16 pm
@seghele0

I don't want fierce debates either, I just wish to understand why the CT approach(big IDE) is better for some users then the Lazarus(OPM) approach. If you don't want to talk about it, no problem. :)
Title: Re: More componenets.
Post by: trev on March 05, 2021, 12:44:05 am
To sum up:

Q: Why is an apple not like an orange?

Short A: Because an apple is not an orange.

Longer A: There are no geneticists sufficiently interested in genetically modifying an apple to be like an orange in their spare time because it is easy enough for people to decide which one they like to eat better and simply eat that one.

What did we learn? Don't get sucked in next time.
Title: Re: More componenets.
Post by: ASBzone on March 05, 2021, 03:03:29 am
What did we learn? Don't get sucked in next time.


Save the thread, lock it from comments, and link back to it when the next round of questions come in... :)
Title: Re: More componenets.
Post by: Martin_fr on March 10, 2021, 11:00:03 pm
Also I think it is a good idea to mention OPM on the Lazarus main page, in 'highlights' section:
Added / Update should be rolled out in the next few days
TinyPortal © 2005-2018