From: Jeff F (TriumphSprint2000_at_[hidden])
Date: 2006-10-10 09:01:31
"Steve Hutton" <shutton_at_[hidden]> wrote in message
> On 2006-10-09, Robert Ramey <ramey_at_[hidden]> wrote:
>> I would make one small change to the above statement - removing the
>> word "standard" so tha tit reads
>> The deciding factor often seems to be the availability of a [standard]
>> (Java, .Net).
>> I see very successful libraries - MFC - totally non portable that are
>> successful because they provide the right toolkit for the job. The
>> fact that they are not standard hasn't been relevant.
> Yes, the majority of the market place doesn't care about the difference
> between standard and de-facto standard. But achieving true de-facto
> standard status means your library must ship with the development
> environments used by the majority of users. In practice the only way
> to achieve de-facto standard level market penetration is to in fact
> to become standardized [*]
> Just look a the level of STL adoption before it was *both* standardized
> and shipped by Microsoft.
This begs the question of how can we get compiler suppliers to include "a
copy" of boost with their compilers? That would certainly remove the single
largest barrier to adoption, would it not? It certainly would in the company
that I just joined.
Are there any on this list that work for MS that would be able to comment on
this possibility? Herb Sutter? What about those actively involved in gcc
development/support, what would it take to make this happen? What other
compiler mfr's are out there?
Then we'd need to define just what is "a copy" of boost. Would it be a
particular release? Some accomodation of compiler release schedules and
processes would be required no doubt.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk