"Marin David Condic" <nobody@noplace.com> wrote in message
news:3FBCB64F.90703@noplace.com...

> ISO is not the way to go. Any library would have to react too quickly to
>   changes in the world around it and freezing it in a standard would be
> to strangle it. Also, it is difficult to write up verifiable
> requirements such that you could validate against an ISO standard. It
> really isn't the place for it.
>
> If the vendors themselves don't want to distribute it, one could
> conceive of other alternatives for a library, but I *still* think that
> it doesn't gain that "de facto standard" status unless you at least have
> the vendors somehow pointing their customers off to it or have something
> semi-official (like SIGAda) behind it.
>
*** Well pointing to it, without offering support to it per se is a
different avenue than having the vendor supply the library with his compiler
I think.  Might be an easier to reach goal.  I think they would probably all
agree to point to it :-). it would al teast get it know and propagated
(especially if it's a "at least minimal" quality library).
>
> It would be conceivable that one might start being a library vendor -
> there are small companies that sell C libraries for specialized
> purposes. However, you'd need to look at the economics of it: Is there
> enough Ada out there to make an Ada library a self-sustaining business?
> Especially considering you have no support from the vendors or any
> guarantee that your library gets considered to be the de facto standard.
>   I think that writing it, documenting it and maintaining it would be a
> full-time job for probably more than one guy. So either it has to
> generate enough income to provide that one or more guys with their
> weekly beer budget plus rent, or it needs to be done by the
> independently wealthy philanthropic Ada developers out there who have
> nothing better to do with their time and money.
>
*** Right now, economically speaking, with the current "GP" user base, it
might be suicide.  The library would play a secondary role, not a primary
one at spreading the Ada word I think.  like a vicious circle, is it the
library or the language vendors that are going to make it work?  Partly
both, in each their different realms, perhaps.  But I think the ones with
the financial backbone to "advertise" Ada properly are the one sthat should
start the big wheel turning.  Hence the vendors.  We could make a "never
before seen" library that can control the world. But if nobody uses the
language we make it for, no one will even control the world "so to speak
:-).  Granted, a library might e a good selling point, although most don't
seem to see it that way unless ISO tells them to see it that way, but no one
ever needed ISO in any other languages to make their mark  Like you said,
ISO isn't the place for it, at least not without specific reorganization for
Standard library mamagement.

> SIGAda might make a forum for doing it, but I'd think it would need a
> small team that was able to get together frequently to argue it out. It
> also would likely need a little representation from the vendors to gain
> some perspective on what customers might want. That would be a volunteer
> effort so it would take a long time before something got very large, but
> it would at least have some kind of official standing.
>
> MDC

*** I agree with you here, it would make a whole lot of sense.
-- 
St�phane Richard
"Ada World" Webmaster
http://www.adaworld.com