Did FreeBasic is stagnating / dying?

General discussion for topics related to the FreeBASIC project or its community.
Lachie Dazdarian
Posts: 2338
Joined: May 31, 2005 9:59
Location: Croatia
Contact:

Post by Lachie Dazdarian »

I'm sort of a reactionist, happy with what FB already provides.

"If it ain't broken, don't fix it."

If some people are so needy of OOP, how about changing the programming language? Seriously, what do you really need from FB on then end if it has to be mangled so much for OOP and GCC front-end to work properly?

But my programming interests are quite narrow so many of the needs people express here are beyond me.
Deleter
Posts: 975
Joined: Jun 22, 2005 22:33

Post by Deleter »

@Lachie

Well, the whole reason its taking so long is that they aren't mangling it. Done right, when fb is done, you shouldn't see any difference when you write your code without oop. Just as you can write pure c code in c++.

Oop is just a newer ice cream flavor with some interesting properties when applied to projects. For one it greatly enhances the re-usability and structure of code. Well, it greatly enhances the possibility of those things and reduces the work required to achieve them, it also adds layers of complication that can be hacked to make even uglier code. The point is that oop will be a feature on top of fb, it shouldn't disturb those coding happily without it.
bfuller
Posts: 362
Joined: Jun 02, 2007 12:35
Location: Sydney, Australia

Post by bfuller »

I think the developers are doing an excellent job----well done, keep up the good work. Even if you stop now, your legacy is an easy to use, friendly compiler, with a healthy/active user group.

We all know that you have real lives, with other priorities---who doesn't? Everyone needs to keep a balance in their lives, you are better off for it in the long run. Do what you enjoy---its your time.

Anyway, well done-----OOP,inheritance,CLASS,c, c++, gcc,pointers,BYREF,-----I don't even know what half that stuff means (but @marcov, I do know what a "naked function" is, and it has nothing to do with programming!)

The FB team are doing a good job and deserve full support from all of us.
egocks
Posts: 20
Joined: Mar 27, 2008 2:02
Location: Philippines

Post by egocks »

@devs: thank you, thank you very much for your great work on FB.
counting_pine wrote:...has mainly been busy with...
Real Life
...recently...
Is that the name of his girlfriend? :P
Just kidding. Ok, I'm outta here...
marcov
Posts: 3462
Joined: Jun 16, 2005 9:45
Location: Netherlands
Contact:

Post by marcov »

bfuller wrote: "naked function" is, and it has nothing to do with programming!)
Well still I don't :-)
ciplogic
Posts: 19
Joined: Jun 05, 2005 9:26
Contact:

Post by ciplogic »

Hi guys,

I say from scratch: I did not wanted to do trolling and to requiest big features. FreeBasic have no target IMO to provide for now OOP and that is right. The bad thing is that the today expectations are around OOP, even they are in most time not to be used. I am one that really think that OOP should be made on a great foundation, so polishing FreeBasic till will have not blocker/critical bugs is much better than Delegates, Reflection and OOP. Eventually adnotations and dependency injection. By far, is not right for now.

Probably FB should try again IMO to try to do a C backend, and so, will create a decent code and will became an optimizing compiler. Second thing is to polish their FBEdit to be much more improved. How much? Don't stop on thinking, take any dialog and make it work right, add Linux and OSX. Probably using a QT backend or GTK (which is closer to C/FB).

I am a big fan of FreePascal (sorry for mentioning other project) and yet they do from scratch a text IDE that is polished over and over again, and today is somehow good to be used. So any user should have from scratch syntax highliting, some code suggestions, help, project management. Making FB a user-friendly project, will attract (let's hope) more users.

Regarding me, I care about OOP as a fact in today programming and yet I consider that Linq (from C#) is a nice thing but is not mandatory for today. But I say about them, because LinQ is based on Generics (templates from C++). So will be at any time one person that will ask for one more feature.

So, for me, for now, is my hope that you to focus at all to polish the IDE/language, making let's say a syntax sugar when you get a place, like if ( x > 2 and x < 5) to be written if (2 < x < 5), but not more than that. Add libraries for everything is under the sun: mostly UI (toolkits), database connectors, graphic APIs (most first wish programming is to make a game, so should be ported both OpenGl and DirectX) and of course add in the package the FBIde that integrates with FBC and at any decent requiest, add it to IDE and any user will enjoy using and taking FBIde as a good option.

Regards,
Ciprian
marcov
Posts: 3462
Joined: Jun 16, 2005 9:45
Location: Netherlands
Contact:

Post by marcov »

ciplogic wrote:
Regarding me, I care about OOP as a fact in today programming and yet I consider that Linq (from C#) is a nice thing but is not mandatory for today. But I say about them, because LinQ is based on Generics (templates from C++). So will be at any time one person that will ask for one more feature.
(Linq is a combination of multiple technologies, and syntax. Anonymous functions, lambda expressions, generics)
ciplogic
Posts: 19
Joined: Jun 05, 2005 9:26
Contact:

Post by ciplogic »

marcov wrote:
ciplogic wrote:
Regarding me, I care about OOP as a fact in today programming and yet I consider that Linq (from C#) is a nice thing but is not mandatory for today. But I say about them, because LinQ is based on Generics (templates from C++). So will be at any time one person that will ask for one more feature.
(Linq is a combination of multiple technologies, and syntax. Anonymous functions, lambda expressions, generics)
You miss the point:
- if there is OOP support, someone will say: why not templates
- if there are templates, why not Lambda functions or delegates
- if will be all that, why not Query language?

This multiparadigm should not be the future of FBC, the future should be quality and user-friendlyness. If someone need SQL, should use SQL, but not making a soup in FB programming, at least for now, for that resources.

The point was, at any level will came someone to ask an extra-feature that is trendy, the bubble burst and not focusing that the base compiler may work bad, the debugger works bad, etc.

Hope to get the idea,
Regards,
Cip
krcko
Posts: 163
Joined: Jul 30, 2006 0:34
Location: Serbia
Contact:

Post by krcko »

if there are lamba functions and delegates why not extension methods? :)
stylin
Posts: 1253
Joined: Nov 06, 2005 5:19

Post by stylin »

ciplogic, I agree with you that if you try and shoehorn features from other languages that you come to the idea of "why not just use that language ?" For example, I'd love a more robust FB runtime library, maybe including GUI or internet APIs (taking full advantage of at least the OOP we have now..). But then you'd probably just be making another VB.NET (minus the multi-language interoperability), and why reinvent the wheel when that language is already established and used worldwide ? It would be counter-productive to a certain extent; by that time the only thing unique to FB would be the current runtime libraries (ie., QB-compatibility), and then they'd say "why not just use VB.NET or QB rather than some frankenstein hybrid ?" I'm not sure what direction I'd like to see the FB devs to take without having these questions, but certainly GCC-like platform compatibilty is a great neutral start (and very nice advantage).

So yeah, I really don't see an opportunity for FB to thrive other than what it is now (especially with the limited amount of human-resources willing to actually work on the project). I also agree that a more robust IDE package would be great for FB, something that is developed alongside it, rather than "after the fact"; possibly reusing the compiler/runtime source to provide features like code completion, real-time debugging or even an interpreter ("official" FB support for the IDE would obviously be a plus, and I think is what many users are missing at the moment).

Off-topic, I'm not sure how if (2 < x < 5) could work -- a binary operator is a binary operator after all -- unless you prohibit implicit casts from boolean to integer/float and/or do some other quirk magiks. I won't pretend to know how complex something like that would be to implement, but it seems like it would be rather difficult from what I understand of the current source, not to mention the [arbitrarily large] existing code that will most certainly break. Then again, I'm not a compiler author, so take that with a pile of salt.
jevans4949
Posts: 1186
Joined: May 08, 2006 21:58
Location: Crewe, England

Post by jevans4949 »

ciplogic wrote:... like if ( x > 2 and x < 5) to be written if (2 < x < 5) ...

Code: Select all

select case x
  case 2 to 5
    onething
  case else
    anotherthing
  end select
Mysoft
Posts: 836
Joined: Jul 28, 2005 13:56
Location: Brazil, Santa Catarina, Indaial (ouch!)
Contact:

Post by Mysoft »

if (2 < x < 5)
i used to had problem in school cuz this sentence makes NO sence to me...
i dont like revert it in the other side... and this method is too much specific for 2 members :P
marcov
Posts: 3462
Joined: Jun 16, 2005 9:45
Location: Netherlands
Contact:

Post by marcov »

ciplogic wrote:
(Linq is a combination of multiple technologies, and syntax. Anonymous functions, lambda expressions, generics)

You miss the point:
- if there is OOP support, someone will say: why not templates
- if there are templates, why not Lambda functions or delegates
- if will be all that, why not Query language?
Well, I can only give the standard answer we give if sb asks a "why not" question: http://www.freepascal.org/faq.var#extensionselect
Post Reply