Recent

Author Topic: Improving Lazarus IDE...  (Read 46279 times)

BigChimp

  • Hero Member
  • *****
  • Posts: 5740
  • Add to the wiki - it's free ;)
    • FPCUp, PaperTiger scanning and other open source projects
Re: Improving Lazarus IDE...
« Reply #60 on: June 05, 2013, 04:30:50 pm »
Sorry BigChimp.

I am just getting unreceptive to people who continue to justify completely outdated technology.
And I'm getting unreceptive to people who do not read, do not think and read intentions in posts that are not there.

In short: good luck with the improvements and hope to see some good patches soon so we can all benefit.
Want quicker answers to your questions? Read http://wiki.lazarus.freepascal.org/Lazarus_Faq#What_is_the_correct_way_to_ask_questions_in_the_forum.3F

Open source including papertiger OCR/PDF scanning:
https://bitbucket.org/reiniero

Lazarus trunk+FPC trunk x86, Windows x64 unless otherwise specified

Chronos

  • Sr. Member
  • ****
  • Posts: 256
    • PascalClassLibrary
Re: Improving Lazarus IDE...
« Reply #61 on: June 05, 2013, 08:57:05 pm »
Heh, this is already the same situation. Somebody come with question or even suggestion what should be improved and responses are all the times similar.

Be able to have multiple installed versions of application which could have separated configuration is completely valid requirement. In fact, every coder who is trying to use IDE for real work will end with such situation. For example be able to support multiple versions of own products or evaluate new alfa versions.

Then if such feature is not implemented in present, there should be raised issue with feature suggestion in bugtracker to be not forgotten and to be able ask similar questions in future. There should be also analysis of possible implementations. For such purpose combination with wiki descriptive page and bug tracker issue and topic in forum could serve well.

This is again about lack of coordination and leadership. But this is for different discussion...

Chronos

  • Sr. Member
  • ****
  • Posts: 256
    • PascalClassLibrary
Re: Imporoving Lazarus IDE...
« Reply #62 on: June 05, 2013, 09:24:51 pm »
Lots of duplicated effort. It is a pity, the energy could be spent better. CodeTyphon itself will suffer most from this decision, that is why I called them non-clever.
Duplicate effort would not be always bad. For example many companies do creating applications which do the same but yet every is bit different. There are plenty of linux distributions but is it really bad? Main reason is that no one is good enough so people are forced to do things in different better way. Each can possibly inspire other and be inspired from others. This is not useless duplicate work. It is called diversity and evolution. Which reminds me genetic algorithms for example to system of nonlinear equations but it is different story. If people would be interested then there should be more open source or commercial implementations of pascal/delphi compilers and IDEs. Also there would be more fpc or lazarus forks or even viable alternative to lazarus which could use FPC as base too. If somebody is only taking code and not contributing it is merely his legitimate choice. There many such projects which just take linux source code. So basically only one who could be displeased are lazarus contributors. I see, you are one of them so you want to defend it and it is understandable.

An honest question: do you see a design flaw in AnchorDocking, so it should be replaced and not improved more?
What docking lib you would prefer?
It would require deep analysis to write where are weak places in anchordocking. But basically anchordocking is equally limited by LCL as other docking solutions for lazarus. So to successfully implement solid docking one have to enhance LCL in many ways. But this is not new either surprising information.
For example it doesn't support tabbed docking and popup forms. Also there are other usability issues. You are happy to implement TControlBar and TCoolBar? Then you could be interested in implementing delphi TDockTabSet which is also component with interesting code.
For now I am not expecting nothing more than RAD Studio-like docking. After this will be finished we can discuss other features.
But Delphi itself haven't components for docking just TDockTabSet and some docking support in general components. So you can examine JEDI VCL components and their docking solution.

I should make wiki page with summary of existed docked solutions and what delphi have and what lazarus have and what should be done to be able to implement fully features docking.

JuhaManninen

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 4675
  • I like bugs.
Re: Imporoving Lazarus IDE...
« Reply #63 on: June 05, 2013, 11:46:38 pm »
So basically only one who could be displeased are lazarus contributors. I see, you are one of them so you want to defend it and it is understandable.

No I am not displeased. Competition is good. If Lazarus was as badly organized as you keep repeating, then it would die, but I don't see it happening now.
Mostly I am surprised because from CodeTyphon's point of view it just makes no sense to fork the Lazarus core IDE.
If I was the person making decisions there, I would try to improve Lazarus through patches and then becoming developer with commit rights, at the same time benefiting from all the other contributions. That is a normal procedure between many other OSS projects and commercial companies.
But no, they did not want to try that.

BTW, for multiple Lazarus installations the --pcp parameter was suggested. What was wrong with it?
How does it prove "lack of coordination and leadership" for you?

In reality Lazarus is near OSS top league by many measures.
Look at the commit history. We are at revison > r41500. The number of commits per day is around 10 average consistently.
Look at the number of posts in mailing lists and forum, and the number of valid answers to questions.
Look at the number of solved bug reports and patches applied. Ok, you had 2 patches that were ignored for a while, but they are balanced with hundreds of applied patches in just few months. The commit message then mentions "issue #xxxxx" and "patch from xxxxxx" or similar.

Many better known projects are not as active. Maybe Ohloh gives statistics of it.
The biggest problems may be in the public relations front, not in coding itself.
Management in the same sense as in a commercial environment is not possible, because the work is volunteer. Donations like your possible bounty can partly change the situation.

I don't know yet if I will "grab the bait" with the docking project, simply because I am not sure if my knowledge and capacity is enough.
« Last Edit: June 06, 2013, 12:55:16 pm by JuhaManninen »
Mostly Lazarus trunk and FPC 3.2 on Manjaro Linux 64-bit.

Pascaluvr

  • Full Member
  • ***
  • Posts: 216
Re: Improving Lazarus IDE...
« Reply #64 on: June 06, 2013, 05:46:26 am »
Quote
BTW, for multiple Lazarus installations the --pcp parameter was suggested. What was wrong with it?

Excellent solution for those that can find it...   i've been here for 2 years and its the first i have heard of it.

A newbie may find this option if he stumbles across this thread.

I will suggest tho that very few newbies use command line parameters ...  they aren't used in the mainstream.

LOOK - I am not criticizing Lazarus - I think its great - I was just stupid enough to place a comment in a thread called 'Improving Lazarus...' - I will never make such a stupid mistake again.  The slightest suggestion for 'improvement' is taken as an attack - FFS people - GROW UP!
Windows 7, Lazarus 1.0.8, FPC 2.6.2, SQLite 3

ludob

  • Hero Member
  • *****
  • Posts: 1173
Re: Improving Lazarus IDE...
« Reply #65 on: June 06, 2013, 08:06:16 am »
Quote
I will suggest tho that very few newbies use command line parameters ...  they aren't used in the mainstream.
:o
Opening or printing a file from explorer is based on the registered program being able to accept command line parameters.  Command line parameters are used almost everywhere. Not always visible, I agree, but that is why desktop shortcuts were invented.

Don't confuse old technology with obsolete technology. The wheels on my car are still round although round wheels were invented a very long time ago;)

Having multiple independent IDE's is never something for newbies or for people that don't want to touch command line parameters. A little tour of the "competition":
Eclipse: http://www.javalobby.org/java/forums/t18678.html
Netbeans: http://ptrthomas.wordpress.com/2008/04/13/never-install-netbeans-again/
Delphi: can't install multiple IDE's of the same revision. There are tricks with "portable" versions which I will ignore here.

Quote
LOOK - I am not criticizing Lazarus - I think its great - I was just stupid enough to place a comment in a thread called 'Improving Lazarus...' - I will never make such a stupid mistake again.  The slightest suggestion for 'improvement' is taken as an attack - FFS people - GROW UP!
Nobody says you are wrong when you say that the installation can be improved. Only the suggestion to make parallel installations easier is clearly not perceived as being high on the list of the other users compared to support for docking or projects or ...   The problem here is that you try to impose this as a high priority problem to solve. That is not how open source projects works. If something is important for you, add it and if it is compatible with the overall project it will be accepted as a contribution.

Chronos

  • Sr. Member
  • ****
  • Posts: 256
    • PascalClassLibrary
Re: Imporoving Lazarus IDE...
« Reply #66 on: June 06, 2013, 08:28:50 am »
No I am not displeased. Competition is good. If Lazarus was as badly organized as you keep repeating, then it would die, but I don't see it happening now.
Mostly I am surprised because from CodeTyphon's point of view it just makes no sense to fork the Lazarus core IDE.
If I was the person making decisions there, I would try to improve Lazarus through patches and then becoming developer with commit rights, at the same time benefiting from all the other contributions. That is a normal procedure between many other OSS projects and commercial companies.
But no, they did not want to try that.
It is simply matter of opinion and chosen strategy. They evaluate situation and they choose own strategy. They know their reasons. You can ask them. Do not contribute back is not necessary evil. Evil could be some kind of attacking developers, servers, trademark, ...

BTW, for multiple Lazarus installations the --pcp parameter was suggested. What was wrong with it?

I used this parameter as well but it is simply not intuitive. It is technical and allow config directory change but you have to learn it by yourself, dig deep in documentation. If someone says that it cannot be done better then I simply not agree. If it would be ok, then there is no need for other starting scripts and managers.
For example why Lazarus have startlazarus additional application? If you need to recompile and restart lazarus you can simply exit application and start it manually again. Yet automatic restart is more convenient.
Another example, If you start lazarus and it doesn't find needed files it show configuration dialog. It is definitely useful and more user friendly than fixing directories in config files directly.
So be able to maintain simply multiple lazarus installations make good sense for me.

How does it prove "lack of coordination and leadership" for you?
You can judge project from many different perspectives. Book linked by you was mentioning some pitfalls which could arise during project existence.
* support for legacy technologies
* support for current technologies
* support for future technologies
* leadership, coordination
* community support
* usability, look and feel
* previous, current and future estimated project activity
* new contributor engagement, promotion, community size change
and more...

You can mark each item from 1 to 10.
Current status is simple. Lazarus and FPC are project of type, where some volunteers who have some interest in project contribute to project especially to part which is interest for them. They don't care about other parts and any promotion. Some people manage some servers so they can give commit access other write access so it can be called as lowest leadership form. But to judge particular project you have to compare it to other existed. You can compare to small projects and to huge projects. To projects which already show that something is possible and can be done. Did I mentioned LibreOffice and OpenDocument foundation? You want to object that it is secured by huge financial capital? And what about charts of ever increasing number of comiters?

So lack of leadership is related mainly that there are not set priority according what is needed in general by users but for what contributors have time and motivation. It is perfectly understandable but if somebody talk about growing of community then there are many reasons that community cannot grow significantly. It is clear that most of people who came here just used Turbo Pascal, Delphi or other pascal compiler.
This is for longer discussion which ends with something like, people doesn't have time and money and nothing can be done about it.

Chronos

  • Sr. Member
  • ****
  • Posts: 256
    • PascalClassLibrary
Re: Improving Lazarus IDE...
« Reply #67 on: June 06, 2013, 08:43:21 am »
The problem here is that you try to impose this as a high priority problem to solve. That is not how open source projects works. If something is important for you, add it and if it is compatible with the overall project it will be accepted as a contribution.

It is not simple as you stated. There are various open source projects which use different models. There is public forum so he can register and have right to demand some features. It is perfectly ok. But now there is second part related to answer to his demand. And there many possibilities what can be done. For example somebody will answer him "you are dumb because this simple feature already exists."(means complicated undocumented feature) . Or just be formal and neutral: "Dear user, you can do it by ... . We don't support better way for now.". Or answer positive "Hello, thanks for you question. Now you can do this by this more complicate way... You can find documentation here ... Your suggested feature is not recorded in bugtracker and will be done then we will have resources for it. Also you can support our project by....".

As Juha mentioned, at least public relation is missing. But even if somebody will record and discuss all new features there is second part. When and if they will be eventually implemented. It is another part of project coordination how obtain money, who pay with them, volunteers or professionals. It can be done in various ways. If most of lazarus team is satisfied with ways how it works now then this discussion is wasting of time.

Pascaluvr

  • Full Member
  • ***
  • Posts: 216
Re: Improving Lazarus IDE...
« Reply #68 on: June 06, 2013, 12:20:04 pm »
Quote
I will suggest tho that very few newbies use command line parameters ...  they aren't used in the mainstream.
:o
Opening or printing a file from explorer is based on the registered program being able to accept command line parameters.  Command line parameters are used almost everywhere. Not always visible, I agree, but that is why desktop shortcuts were invented.

I agree entirely.  Command line parameters are rarely visible these days - thats why i said they aren't mainstream - I wasn't suggesting they weren't often used.

Quote
Don't confuse old technology with obsolete technology. The wheels on my car are still round although round wheels were invented a very long time ago;)

Having multiple independent IDE's is never something for newbies or for people that don't want to touch command line parameters. A little tour of the "competition":
Eclipse: http://www.javalobby.org/java/forums/t18678.html
Netbeans: http://ptrthomas.wordpress.com/2008/04/13/never-install-netbeans-again/
Delphi: can't install multiple IDE's of the same revision. There are tricks with "portable" versions which I will ignore here.

How condescending of you!  Of course there are examples of other apps that fall short - but there are MANY apps that handle this with ease.  On my PC I can have multiple operating systems and I don't need to be a computer geek to achieve this. 

Quote
...   The problem here is that you try to impose this as a high priority problem to solve. That is not how open source projects works. If something is important for you, add it and if it is compatible with the overall project it will be accepted as a contribution.

Where have I suggested this is HIGH priority?   I simply made an addition to a topic called 'Improving Lazarus' for discussion. This forum is not the place to make feature requests - but it could be a place to discuss possible requests - instead - its a forum that attacks anybody that dares suggest that Lazarus is lacking in any way.

As I stated previously, I will NEVER upset members here again with making any suggestion ever again - it seems too many  members here take such suggestions as a personal insult.
Windows 7, Lazarus 1.0.8, FPC 2.6.2, SQLite 3

ludob

  • Hero Member
  • *****
  • Posts: 1173
Re: Imporoving Lazarus IDE...
« Reply #69 on: June 06, 2013, 03:09:23 pm »
It is simply matter of opinion and chosen strategy. They evaluate situation and they choose own strategy. They know their reasons. You can ask them. Do not contribute back is not necessary evil. Evil could be some kind of attacking developers, servers, trademark, ...
I would add to the evil list releasing software to the general public that is clearly not ready for release, such as fpc or lazarus trunk. That is potentially hurting the user and upstream. We all know who gets the blame when a bug pops up  ;)

DelFly

  • New Member
  • *
  • Posts: 38
Re: Improving Lazarus IDE...
« Reply #70 on: July 14, 2013, 11:40:48 am »
Im going to try lazarus again. used to different flavors of delphi.
But one of the most annyoing things is the not single window  mode.  :(
like you have with delphi 2005+ and gimp 2.8 Visual studio.
Lazarus is a window mess on my screen and every time you start lazarus ,
he windows are again not saved at the right place. The way lazarus works (delphi 5 look)
its a bit disorientating. because windows pop under / over  and wrong sizes. And add ons i dont try, i think in the past i tried it and did not do a good job or correct.

Single window makes it also less messy, because it cover the whole screen so you dont see other application below it and need to search for the right [X]
There more good reasons to go single window then stay the old way.

Only wish the made a single window lazarus.
Cant touch anything on computer without hitting some bug.

DelFly

  • New Member
  • *
  • Posts: 38
Re: Improving Lazarus IDE...
« Reply #71 on: July 14, 2013, 11:53:15 am »
Ok. before somneone screams  :P Im going to try KZdesktop.
Cant touch anything on computer without hitting some bug.

eny

  • Hero Member
  • *****
  • Posts: 1653
Re: Improving Lazarus IDE...
« Reply #72 on: July 14, 2013, 11:53:35 am »
Im going to try lazarus again. used to different flavors of delphi.
But one of the most annyoing things is the not single window  mode.  :(
If you want single window mode stick to something that suits your needs

Quote
like you have with delphi 2005+ and gimp 2.8 Visual studio.
Lazarus is a window mess on my screen and every time you start lazarus ,
he windows are again not saved at the right place.
After repositioning them after a new install they are saved at the right place.

Quote
The way lazarus works (delphi 5 look) its a bit disorientating. because windows pop under / over
Then do not open other programs when you want to work with Lazarus. Simple.
Or add a shortcut to your quick launch bar that clears the desktop for you. Simple.

Quote
Single window makes it also less messy, because it cover the whole screen so you dont see other application below it and need to search for the right [X]
You do know that you can can choose File/Quit from the menu don't you?...

Quote
There more good reasons to go single window then stay the old way.
There are not...

Quote
Only wish the made a single window lazarus.
I wish another 500000 on my bank account.
All posts based on: Win10 (Win64); Lazarus 3_4  (x64) 25-05-2024 (unless specified otherwise...)

DelFly

  • New Member
  • *
  • Posts: 38
Re: Improving Lazarus IDE...
« Reply #73 on: July 14, 2013, 12:03:06 pm »
Quote
"After repositioning them after a new install they are saved at the right place."
Well that never worked here. only when you save the  project.
but the never get saved without project.

Quote
You do know that you can can choose File/Quit from the menu don't you?.
Never use that in applications or very rare, because
its to many mouse movements. something i need to avoid.

Quote
I wish another 500000 on my bank account.
Well, here i can agree. with that amount on the bank
account you can buy some programming language  %)


Installed KZDesktop. and thats preciese what lazarus where missing. WOW  :)
Better contrasts, better organized. 1001 reasons to have a single window IDE.
cant find any for the old one.

No time wasting hide and seek anymore with lazarus  8-)

« Last Edit: July 14, 2013, 12:09:14 pm by DelFly »
Cant touch anything on computer without hitting some bug.

DelFly

  • New Member
  • *
  • Posts: 38
Re: Improving Lazarus IDE...
« Reply #74 on: July 14, 2013, 12:11:27 pm »
Well, the warned for it.  :( KZDesktop still unstable. one form resize in window8 desktop and errors fly around in lazarus
Cant touch anything on computer without hitting some bug.

 

TinyPortal © 2005-2018