Sean Kelly wrote:
> BCS wrote:
>> Leandro Lucarella wrote:
>>
>>> I know, but that is not a standard, it's a specification. A standar
>>> is predictable and has an agenda, and meetings, and people discuss
>>> and desitions are taken by more than one only person.
>> Well if that is a standard, then lets NOT standardize D!!
>>
>> The parts of a "standard" I have any use for is the part that lets you
>> use it. I don't give a ___ how it's made (D seems to do well with King
>> Walter) just what it says. IIRC in mechanical systems, standards are
>> things like "threads are this shape" not the people that define them
>> (that is a standard committee or organization).
> 
> To be fair, a standard represents some sort of agreement among people,
> typically about a design.  And the usefulness of a standard is directly
> tied to how many people agree about that design--the most common way of
> reaching a consensus being a committee.  So the D language is 'standard'
> insofar as that we have all agreed to follow the design.  The situation
> is a bit unstable in that it relies on the receptiveness and the talents
> of one individual (ie. Walter), but fortunately he has proven himself up
> to the task so far.
> 
> 
> Sean

Right now, D2 is not standard even in that sense -- Walter hasn't reached a consensus with himself. <g>
Look how 'const' has changed over the past ten releases.
So to answer the original post -- I expect at least another year of releases before Walter will say that the language design is essentially complete.

Don Clugston Wrote:
> Right now, D2 is not standard even in that sense -- Walter hasn't reached a
> consensus with himself. <g>
> Look how 'const' has changed over the past ten releases.
> So to answer the original post -- I expect at least another year of releases
> before Walter will say that the language design is essentially complete.

Maybe then it'll be time for D3 ;)

On Fri, 01 Feb 2008 18:22:30 -0000, Jason House <jason.james.house@gmail.com> wrote:

> Don Clugston Wrote:
>> Right now, D2 is not standard even in that sense -- Walter hasn't reached a
>> consensus with himself. <g>
>> Look how 'const' has changed over the past ten releases.
>> So to answer the original post -- I expect at least another year of releases
>> before Walter will say that the language design is essentially complete.
>
> Maybe then it'll be time for D3 ;)

Design by comittee is reknowned for its speed even when there is concensus.
Look at how long it took to get ISO C++ and they daren't put a definite year
to C++0x. It was probably a little optimistic putting a 0 there.

If I may quote the fictional genius Prof. Adonis Cnut

Dr Hannah Awkward: "The human genome project took over a decade of work by hundreds
of scientists around the world".

Prof Adonis Cnut: "but I was working on my own so it was much quicker"

Bruce Adams wrote:
> On Fri, 01 Feb 2008 18:22:30 -0000, Jason House <jason.james.house@gmail.com> wrote:
> 
>> Don Clugston Wrote:
>>> Right now, D2 is not standard even in that sense -- Walter hasn't
>>> reached a
>>> consensus with himself. <g>
>>> Look how 'const' has changed over the past ten releases.
>>> So to answer the original post -- I expect at least another year of
>>> releases
>>> before Walter will say that the language design is essentially complete.
>>
>> Maybe then it'll be time for D3 ;)
> 
> Design by comittee is reknowned for its speed even when there is concensus.
> Look at how long it took to get ISO C++ and they daren't put a definite
> year
> to C++0x. It was probably a little optimistic putting a 0 there.

I won't say that design by committee is fast, but a considerable amount of the slowness for C++ is related to the process required by the standards body itself.  I think the process was really designed with mechanical and electrical engineering in mind, rather than the quickly changing software field.


Sean

Bruce Adams, el  1 de febrero a las 19:00 me escribiste:
> On Fri, 01 Feb 2008 18:22:30 -0000, Jason House <jason.james.house@gmail.com> wrote:
> 
> >Don Clugston Wrote:
> >>Right now, D2 is not standard even in that sense -- Walter hasn't reached a
> >>consensus with himself. <g>
> >>Look how 'const' has changed over the past ten releases.
> >>So to answer the original post -- I expect at least another year of releases
> >>before Walter will say that the language design is essentially complete.
> >
> >Maybe then it'll be time for D3 ;)
> 
> Design by comittee is reknowned for its speed even when there is concensus. Look at how long it took to get ISO C++ and they daren't put a definite year to C++0x. It was probably a little optimistic putting a 0 there.

You can have consensus in other languages that are not ISO Standard, even when they have their BDFL is there a lot more discussion and a more formal process, which means predictability.

And yes, I'm talking about Python as an example :)

-- 
Leandro Lucarella (luca) | Blog colectivo: http://www.mazziblog.com.ar/blog/
----------------------------------------------------------------------------
GPG Key: 5F5A8D05 (F8CD F9A7 BF00 5431 4145  104C 949E BFB6 5F5A 8D05)
----------------------------------------------------------------------------
<Damian_Des> Me anDa MaL eL CaPSLoCK

Don Clugston wrote:
> Right now, D2 is not standard even in that sense -- Walter hasn't reached a consensus with himself. <g>
> Look how 'const' has changed over the past ten releases.
> So to answer the original post -- I expect at least another year of releases before Walter will say that the language design is essentially complete.

D1 is complete as far as I'm concerned, except for bug fixes and changes that improve the compiler without changing the spec.

The current plan is for D2 to be finalized sometime in the fall.

Sean Kelly wrote:
> I won't say that design by committee is fast, but a considerable amount
> of the slowness for C++ is related to the process required by the
> standards body itself.  I think the process was really designed with
> mechanical and electrical engineering in mind, rather than the quickly
> changing software field.

The slowness is driven by a need to develop consensus among a large group of people, and to leave a heavily documented trail. This process is not inherent to the procedures of the standards body, the process is inherent in design by committee, and is merely reflected in the formal procedures.

Reply to Walter,

> Don Clugston wrote:
> 
>> Right now, D2 is not standard even in that sense -- Walter hasn't
>> reached a consensus with himself. <g>
>> Look how 'const' has changed over the past ten releases.
>> So to answer the original post -- I expect at least another year of
>> releases before Walter will say that the language design is
>> essentially
>> complete.
> D1 is complete as far as I'm concerned, except for bug fixes and
> changes that improve the compiler without changing the spec.
> 
> The current plan is for D2 to be finalized sometime in the fall.
> 

at this reate it should be D-07 and D-08  :) keep up the good work.

"Bruce Adams" <tortoise_74@yeah.who.co.uk> wrote in news:op.t5u0eef7xikks4@starquake.cybernetics:

> Design by comittee is reknowned for its speed even when there is
> concensus. Look at how long it took to get ISO C++ and they daren't
> put a definite  year
> to C++0x. It was probably a little optimistic putting a 0 there.

Maybe they're hoping to finish in time to have a C++0x10?  If so, they may not make it. :-)

JMNorris wrote:
> "Bruce Adams" <tortoise_74@yeah.who.co.uk> wrote in
> news:op.t5u0eef7xikks4@starquake.cybernetics: 
> 
>> Design by comittee is reknowned for its speed even when there is
>> concensus. Look at how long it took to get ISO C++ and they daren't
>> put a definite  year
>> to C++0x. It was probably a little optimistic putting a 0 there.
> 
> Maybe they're hoping to finish in time to have a C++0x10?  If so, they may not make it. :-)

2016...?