Possible Solution to threads like the one that was started.

For other topics related to the FreeBASIC project or its community.
Z!re

Postby Z!re » May 22, 2008 5:06

[Content removed at author's request]
cha0s
Site Admin
Posts: 5317
Joined: May 27, 2005 6:42
Location: Illinois
Contact:

Postby cha0s » May 22, 2008 7:46

Z!re wrote:Added some commands like min(), max() and fixed one or two minor bugs that were posted about in the forums.
Nothing major or special, I just didn't want to convert all my code from .16 to .17.


I have min and max too... they're one-line macros.

At least there'd be a clear goal and purpose.
Now FB is this convoluted mess of cross dialect ambiguity that makes no sense.
People have to specify if it's -lang randomstuffhere when posting code in the forums?
That is not a good thing.
If it's FB code it is FB code.


You have a point, although I think the point of having multiple dialects is very clear. (OPINION) If you aren't using -lang FB, you're on the backburner. Really is it so different than when QB was the best BASIC implementation we found? We were always one step behind the mainstream, and now there's some resentment when that's the case... Just saying. What changed between when there was QB and now? You can believe that some evil satan fairies came and added all these new features to FB, but what's the diff? If you want QB then you wouldn't have any of it anyways....

As it stands now we have some weird hacked together OOP.
That will most likely be thrown out the window when GCC is implemented, which may or may not happen.

[...]

This is not a matter of taste, it is simply fact.


Where are the facts? You're just spewing opinion, which of itself is fine, except when you try to charade it as fact.

Do you honestly think any serious project will be started long-term with those promises of stability?


Yeah, I do. I also think that in the cases of borderline large projects, the decision may be made to code parts in an established language like C++. Good thing our 'hacked together OO' supports the G++ ABI completely, as far as we have implemented. You're entitled to your own opinion, for sure! This is not fact!

Without stability in the product there can be no stable community.


This is not logical, nor a factual conclusion. It is again opinion. Which you're entitled to. And entitled to grow up and deal when people think that spewing opinion and calling it fact is bull $%#@.

You're part of the FB community, I don't know who else I'd want to discuss it with? Perhaps a C developer? ;)


I think you have to realize something. That for all the talk that comes out of the grapevine, very few people step forward and create solutions. That's simply fact and you're blind if you don't acknowledge it.

Now, I can't speak for counting_pine, but maybe he doesn't want the responsibility of creating your "reality", when in actual reality, he's not getting paid a cent to do it.

I think I speak for all FB devs when I say I am very satisfied with what has been accomplished so far. We are not without goals, we are not without ability. On the majority, we are without time and manpower. So you might understand how discouraging it is for you to come in and act high and mighty about the direction of the language, when in reality you have never submitted a single compiler or rtlib patch (that I'm aware of.)

Call it elitism, the fact that we might give more weight to people who have contributed. If that's elitism, then show me the dotted line. I'm signing up.
marcov
Posts: 2801
Joined: Jun 16, 2005 9:45
Location: Eindhoven, NL
Contact:

Postby marcov » May 22, 2008 8:11

cha0s wrote:Call it elitism, the fact that we might give more weight to people who have contributed. If that's elitism, then show me the dotted line. I'm signing up.


Well said. You forgot the standard call for patches though, for little loved functionality :-)

P.s. and update that about page to reflect the opinion of the devels. Even "elitism" ("laborat decidis" sounds nicer, but I don't know if I got my endings right) doesn't lift the burden of communicating that.

Otherwise you are deceiving potential future committers.
cha0s
Site Admin
Posts: 5317
Joined: May 27, 2005 6:42
Location: Illinois
Contact:

Postby cha0s » May 22, 2008 8:29

marcov wrote:P.s. and update that about page to reflect the opinion of the devels.


Not sure what you mean, exactly...

I guess I should specify before all my posts is that I generally speak for myself. Maybe the other devs would want to distance themselves with 'elitism', but that's just how I feel... Feel free to take it out of context people. Tabloids are a multi-million dollar enterprise.
marcov
Posts: 2801
Joined: Jun 16, 2005 9:45
Location: Eindhoven, NL
Contact:

Postby marcov » May 22, 2008 8:35

cha0s wrote:
marcov wrote:P.s. and update that about page to reflect the opinion of the devels.


Not sure what you mean, exactly...

I guess I should specify before all my posts is that I generally speak for myself. Maybe the other devs would want to distance themselves with 'elitism', but that's just how I feel...


I just refered to the project description "about" page (see earlier post) that is a bit QB centric "the most compatible".

You already said it is your opinion, but what is the official stand of FB?
marcov
Posts: 2801
Joined: Jun 16, 2005 9:45
Location: Eindhoven, NL
Contact:

Postby marcov » May 22, 2008 8:58

Btw, while in nagging mode, I had a look at:

http://www.freebasic.net/index.php/abou ... n=features

And then specially the "Unlimited number of symbols" part.

It doesn't say "global", so it is only a matter of time till you get a generated source submitted with one procedure that is several megabytes of source (and e.g. with nesting IFs) that kills your register allocator :-)
cha0s
Site Admin
Posts: 5317
Joined: May 27, 2005 6:42
Location: Illinois
Contact:

Postby cha0s » May 22, 2008 9:27

OT... I dunno I've done some pretty outrageous stress tests on FreeBASIC. I don't get why it would kill the register allocator, if regs are spilled. I'd think the issue would be running out of stack space/ram.
marcov
Posts: 2801
Joined: Jun 16, 2005 9:45
Location: Eindhoven, NL
Contact:

Postby marcov » May 22, 2008 10:24

cha0s wrote:OT... I dunno I've done some pretty outrageous stress tests on FreeBASIC. I don't get why it would kill the register allocator, if regs are spilled. I'd think the issue would be running out of stack space/ram.


Usually there is a fixed number of virtual regs. But I know only a handful of compilers internally, maybe you solved it differently.
Mentat
Posts: 332
Joined: Oct 27, 2007 15:23
Location: NC, US
Contact:

Postby Mentat » May 22, 2008 11:30

marcov wrote:
cha0s wrote:Call it elitism, the fact that we might give more weight to people who have contributed. If that's elitism, then show me the dotted line. I'm signing up.


Well said. You forgot the standard call for patches though, for little loved functionality :-)

P.s. and update that about page to reflect the opinion of the devels. Even "elitism" ("laborat decidis" sounds nicer, but I don't know if I got my endings right) doesn't lift the burden of communicating that.

Otherwise you are deceiving potential future committers.


You're hiring?
counting_pine
Site Admin
Posts: 6174
Joined: Jul 05, 2005 17:32
Location: Manchester, Lancs

Postby counting_pine » May 22, 2008 14:54

Z!re wrote:
I wouldn't mind considering it and discussing it constructively - though again, I'm not sure I'm the best person to do that with.
You're part of the FB community, I don't know who else I'd want to discuss it with? Perhaps a C developer? ;)

OK, maybe I am the best person to discuss it with - for better or worse, I may be the only person on the development team who's still prepared to listen to what you're saying. But I don't know the ins and outs of the compiler. If you ask me, for example: how much cleaner would the compiler be if we stripped out all the stuff that's unrelated to lang fb? My answer would have to be that I don't know. But I'd say the other dialects are generally not a hindrance when we're doing lang fb-related work.


Added some commands like min(), max() and fixed one or two minor bugs that were posted about in the forums.
Nothing major or special, I just didn't want to convert all my code from .16 to .17.
Fair enough. If it was the dialect changes that bothered you though, it shouldn't have been too hard to upgrade to lang deprecated. Assuming you didn't use Gosub...

As it stands now we have some weird hacked together OOP.
That will most likely be thrown out the window when GCC is implemented, which may or may not happen.
The GCC backend is still being worked on. And all the changes in the compiler and the language are being made with GCC compatibility in mind. We have no intentions of throwing anything out of the window.

As I said, it probably would be worth deciding on a final specification for the language at some point - I don't think the question is "should it be done?", but "should it be done now?" and, if yes, then "how can that be done?"


@marcov, I agree that it's probably worth changing the About page to say "similar" instead of "most compatible possible". I don't know who has the rights to change that though...

@mentat: You're applying?
cha0s
Site Admin
Posts: 5317
Joined: May 27, 2005 6:42
Location: Illinois
Contact:

Postby cha0s » May 22, 2008 15:10

We need to make a list for instance pages like this need to be changed if we're gonna do that http://www.freebasic.net/index.php/abou ... res#item_1

Everyone close your IDE's, we have more important work to do here....
Mentat
Posts: 332
Joined: Oct 27, 2007 15:23
Location: NC, US
Contact:

Postby Mentat » May 22, 2008 15:29

@mentat: You're applying?

If there is need for more manpower, I would be glad to help.

Back to the matters of the topic, I wouldn't cry if the -lang qb was done away with, but it is convenient when dealing with qb code. I have .bas files default to FBide, so it's easy just to open up a qb file then run it. I don't see it as a necessity, but as an add-on tool. It's worked for me, so I don't see what's wrong with using the most stable version and just throwing it on before future releases, unless it has to be changed to due to changes in the FBC itself.
Z!re

Postby Z!re » May 22, 2008 18:30

[Content removed at author's request]
Skyler
Posts: 242
Joined: Sep 26, 2006 16:30

Postby Skyler » May 22, 2008 18:44

It makes sense that if they want to compile QB programs, they could just download QB. I assume they can still do that?
Lachie Dazdarian
Posts: 2338
Joined: May 31, 2005 9:59
Location: Croatia
Contact:

Postby Lachie Dazdarian » May 22, 2008 21:40

Well said.

Return to “Community Discussion”

Who is online

Users browsing this forum: No registered users and 1 guest