"J Thomas" <jtd514@ameritech.net> wrote in message news:dg13nb$2bff$2@digitaldaemon.com...
> If you want to write an open source compiler,nobodys stopping you; go for
> it. I dont see the point in trying to reinvent Walters
> work when theres so much cool stuff to write in D.
>
> Personally, Im not concerned with the compiler. I think Walter is doing a very good job, considering. Im more concerned with the class libraries and tools. As the tools evolve so will the compiler.
>
> What I think we need to do is stop talking about walter and actually _organize_ an open source community for the serious developers; develop the systems and tools that D needs. develop a community roadmap for the class libraries and tools. I dont expect walter to organize the disorganized and write a great compiler. Anyway, i think you get the idea. also a real forum to discuss this stuff would be nice, which is another reason we should organize outside of the digital mars newsgroup.....

You might want to check out dsource http://www.dsource.org and the ares project in particular, which aims to replace phobos. Personally I believe phobos is fixable and a new external standard library isn't needed. I don't want Walter to spend time fixing phobos, though (or the doc). Since Walter is the code maintainer I don't feel comfortable discussing, designing, coding and sending him "serious" patches only to get an email at the end with "sorry I don't want to do that" or something - which is how the process works today. For example there were lots of ideas in http://www.digitalmars.com/d/archives/digitalmars/D/20858.html and subsequent threads about phobos changes that never got "the green light" so they never even really went to the design/discussion phase. So I've only been sending small changes to individual modules that clean up small issues - but the big issues still remain.

Ben Hinkle wrote:
> "J Thomas" <jtd514@ameritech.net> wrote in message news:dg13nb$2bff$2@digitaldaemon.com...
> 
>>If you want to write an open source compiler,nobodys stopping you; go for it. I dont see the point in trying to reinvent Walters
>>work when theres so much cool stuff to write in D.
>>
>>Personally, Im not concerned with the compiler. I think Walter is doing
>>a very good job, considering. Im more concerned with the class libraries
>>and tools. As the tools evolve so will the compiler.
>>
>>What I think we need to do is stop talking about walter and actually
>>_organize_ an open source community for the serious developers; develop
>>the systems and tools that D needs. develop a community roadmap for the
>>class libraries and tools. I dont expect walter to organize the disorganized and write a great compiler. Anyway, i think you get the idea. also a real forum to discuss this stuff would be nice, which is another reason we should organize outside of the digital mars newsgroup.....
> 
> 
> You might want to check out dsource http://www.dsource.org and the ares project in particular, which aims to replace phobos. Personally I believe phobos is fixable and a new external standard library isn't needed.
> <snip>

Personally, I think dsource is seriously disorganize.
Most projects are being worked on by people who "have lives too"; development can be pretty slow.
Another thing is, there are several "library" projects .. these need to be orgenized into a standard library or "API" so to speak.
I'd love to see a huge and collective library like the Java API and/or the .NET framework.
I personally am not much familiar with .NET, but I am familiar with Java, and the Java API is one of the (if not the only) powerful features about java that pushes me to use the language. I regret to say this, but I use java alot more than I use D, and the only reason is the extensive API.

I think What we need is very few people (possibly one person?) with leadership qualities who have/has a vision and can organize and lead an open library/API project in which the whole community can participate in and/or contribute to.

In article <dg17cq$2e66$1@digitaldaemon.com>, Ben Hinkle says...
>
>
>"J Thomas" <jtd514@ameritech.net> wrote in message news:dg13nb$2bff$2@digitaldaemon.com...
>>
>> What I think we need to do is stop talking about walter and actually _organize_ an open source community for the serious developers; develop the systems and tools that D needs. develop a community roadmap for the class libraries and tools. I dont expect walter to organize the disorganized and write a great compiler. Anyway, i think you get the idea. also a real forum to discuss this stuff would be nice, which is another reason we should organize outside of the digital mars newsgroup.....
>
>You might want to check out dsource http://www.dsource.org and the ares project in particular, which aims to replace phobos.

For what it's worth, I think one aspect of the Ares project could be useful whether it succeeds or not.  That is, since Ares does aim to be a full-featured standard library, it needs to be well documented and it has to run on DMD/GDC pretty much seamlessly.  Thus, one of the first goals was to clean up, separate, and formalize/document the interface between the standard library, the garbage collector, and the compiler runtime (which are currently somewhat interdependent in Phobos)--which basically means that I've been cleaning up and formalizing the interactions between different portions of Phobos.  If you really wanted to, you could take Phobos.std, drop it on the Ares core, and it would pretty much Just Work.

Sean

In article <dg068a$1m7h$1@digitaldaemon.com>, Kyle Furlong says...
>
>Ben Hinkle wrote:
>> "Kyle Furlong" <kylefurlong@gmail.com> wrote in message news:dfvo0p$12fo$1@digitaldaemon.com...
>> 
>>>Walter Bright wrote:
>>>
>>>>"Kyle Furlong" <kylefurlong@gmail.com> wrote in message news:dfs229$1u34$1@digitaldaemon.com...
>>>>
>>>>
>>>>>All that said, he *is* just one man, with the time considerations that a man has. How much better if, under the guidance of Walter and with his approval, knowledgeable and experienced members of the community were allowed to help work out bugs and push dmd towards a 1.0 that has been coming for 6 years, and has never arrived.
>>>>
>>>>D is quite usable as it is, right now. There's no need to wait for a 1.0.
>>>>D
>>>>now is far better than most languages are at 1.0, 2.0, or 3.0.
>>>>
>>>
>>>Ant wrote:
>>>
>>>>The main problem is that I'm desapointed with DMD state, I'm always designing around DMD bugs.
>>>
>>>Ant wrote that in another thread. This is what I am talking about.
>> 
>> 
>> Which bugs are causing problems? The only bugs that come to mind that have
>> caused me to rework code are ones about forward declarations and I haven't
>> hit one of those in a while. That and private access can be overly private
>> (eg - private inside a template or parametrized class is private to the
>> template not the module).
>> Also, not that I'm trying to be mean, but I can't find any posts to D.bugs
>> by Kyle Furlong. If there are bugs that are getting in your way that aren't
>> already known then definitely post about it. Even if it is known Walter
>> might not realize the frequency of the bug and so it would still be useful
>> to post.
>> 
>> 
>
>I'm not speaking as one of the frequent users of D, I am an observer of the newsgroup and I hoped to incite discussion about some of the things that I have noticed.
>
>I guess what I am trying to say is D is capable of taking the world by storm, so why hasnt it?


I agree with you here Kyle. It seems like D is the perfect language, what is holding it back?

In article <dg4ucj$2htk$1@digitaldaemon.com>, Rain Dog says...
>
>In article <dg068a$1m7h$1@digitaldaemon.com>, Kyle Furlong says...
>>
>>I guess what I am trying to say is D is capable of taking the world by storm, so why hasnt it?
>
>I agree with you here Kyle. It seems like D is the perfect language, what is holding it back?

D doesn't fill a popular niche that's current empty, and it's also pre-1.0. That said, I think D has an attractive feature set for certain kinds of programming, but it's also stuff that will not be too terribly common for A While Yet (lock-free multiprogramming being one example).


Sean