C++ Standards Committee membership for Boost?
The relationship between Boost and the C++ Standards Committee has been strictly informal. That's been fine, particularly given that most Boost moderators were members of the standards committee and regularly attended meetings. As the standards committee gears up to accept library proposals, the relationship is getting even closer. I believe it would be advantageous for Boost to join the C++ committee as a voting member. Procedurally, that would mean joining the INCITS (www.incits.org) PL22.16, which is the US national body for C++ standardization. Boost is consider US domiciled in ISO-speak. The advantages of joining include: * We get to participate in formal votes. Right now, we only participate in straw votes. * Boosters would be able to attend committee meetings by right rather than as a courtesy. * Boosters would be able to participate in committee mailing lists with less hoops to jump through. * Psychologically, it may give us a bit more weight in committee matters. Comments? Opinions? --Beman
Hi, I don't think I'm directly concerned but I just have 2 questions that spawned while reading: 1. Boost would have one unique vote, whatever the count of boosters. Is my understanding correct? 2. What is a "booster"? An accepted library writer? candidate library too? Joel Lamotte
On Thu, May 10, 2012 at 9:05 AM, Klaim - Joël Lamotte <mjklaim@gmail.com> wrote:
Hi,
I don't think I'm directly concerned but I just have 2 questions that spawned while reading:
1. Boost would have one unique vote, whatever the count of boosters. Is my understanding correct?
Correct. That's always true; the giant corporations get one vote, too, same as individuals who are members. Since the committee works primarily by consensus, that isn't an issue. And most of the technical votes are straw polls where anyone in the room can vote.
2. What is a "booster"? An accepted library writer? candidate library too?
My definition of a "Booster" is anyone who has been participating in Boost, and the admittedly vague definition of "participating" is very broad, and goes far beyond library authors. --Beman
On Thu, May 10, 2012 at 11:49 PM, Beman Dawes <bdawes@acm.org> wrote:
My definition of a "Booster" is anyone who has been participating in Boost, and the admittedly vague definition of "participating" is very broad, and goes far beyond library authors.
Wouldn't it be a problem for the C++ commitee? My understanding it that the mailing list is voluntarily kept closed to avoid a lot of problem like trolling or out-of-scope subjects that occur naturally in open communities (even boost). I might be wrong, I didn't read any official info about this. But allowing this large definition of boosters to be able to mail the C++ committee mailing list would then open a breach for them, wouldn't it? Joël Lamotte
On Thu, May 10, 2012 at 11:11 AM, Klaim - Joël Lamotte <mjklaim@gmail.com> wrote:
On Thu, May 10, 2012 at 11:49 PM, Beman Dawes <bdawes@acm.org> wrote:
My definition of a "Booster" is anyone who has been participating in Boost, and the admittedly vague definition of "participating" is very broad, and goes far beyond library authors.
Wouldn't it be a problem for the C++ commitee? My understanding it that the mailing list is voluntarily kept closed to avoid a lot of problem like trolling or out-of-scope subjects that occur naturally in open communities (even boost). I might be wrong, I didn't read any official info about this.
But allowing this large definition of boosters to be able to mail the C++ committee mailing list would then open a breach for them, wouldn't it?
Don't worry. Anyone who is disruptive will get kicked off the committee reflectors. There is a very close, personal, relationship between the committee's officers and Boost's key people that goes back many years. We know how to make it work:-) --Beman
On 05/10/2012 05:11 PM, Klaim - Joël Lamotte wrote:
But allowing this large definition of boosters to be able to mail the C++ committee mailing list would then open a breach for them, wouldn't it?
It would probably be any Booster who has a compelling reason to get on board, such as a candidate proposal.
From: mjklaim@gmail.com
On Thu, May 10, 2012 at 11:49 PM, Beman Dawes <bdawes@acm.org> wrote:
My definition of a "Booster" is anyone who has been participating in Boost, and the admittedly vague definition of "participating" is very broad, and goes far beyond library authors.
Wouldn't it be a problem for the C++ commitee? My understanding it that the mailing list is voluntarily kept closed to avoid a lot of problem like trolling or out-of-scope subjects that occur naturally in open communities (even boost). I might be wrong, I didn't read any official info about this.
But allowing this large definition of boosters to be able to mail the C++ committee mailing list would then open a breach for them, wouldn't it?
Not everyone agrees with the closed nature of the committee mailing lists. I for one think that since C++ is an open standard, the standard's development should happen completely in the open, too - that is, everyone should have read access to the committee mailing lists. Limiting write access is reasonable to facilitate efficient operation of the Committee, as you say. Here we could have a scheme where a potential "Booster" requests write access on boost-dev, and an SC commmittee member (or a library author, if that's too much work for the SC committee) approves the request at their discretion. Library authors would get write access up-front. Regards, Nate
Nathan Ridge wrote:
Not everyone agrees with the closed nature of the committee mailing lists.
I for one think that since C++ is an open standard, the standard's development should happen completely in the open, too - that is, everyone should have read access to the committee mailing lists.
+1 Allowing read access for everyone is a good idea! Regards, Michel
On a related side note, I just got notified on a french development website ( http://cpp.developpez.com/actu/44204/Nouveaux-forums-publics-pour-le-commite...) that public discusssion groups have been openned by the ISO commitee... is it really the commitee decision? https://groups.google.com/a/isocpp.org/group/lang-discussion/about https://groups.google.com/a/isocpp.org/group/lang-proposals/about https://groups.google.com/a/isocpp.org/group/lib-discussion/about https://groups.google.com/a/isocpp.org/group/lib-proposals/about Joel Lamotte
On 16/05/2012 07:43, Klaim - Joël Lamotte wrote:
On a related side note, I just got notified on a french development website ( http://cpp.developpez.com/actu/44204/Nouveaux-forums-publics-pour-le-commite...) that public discusssion groups have been openned by the ISO commitee... is it really the commitee decision?
It's an initiative led by Herb Sutter. He proposes that the committee moves to publically readable (but not writable) discussion groups for the reflector and also that we replace comp.std.c++ by new public read/write groups (which are the ones you linked).
on Thu May 10 2012, Nathan Ridge <zeratul976-AT-hotmail.com> wrote:
From: mjklaim@gmail.com
On Thu, May 10, 2012 at 11:49 PM, Beman Dawes <bdawes@acm.org> wrote:
My definition of a "Booster" is anyone who has been participating in Boost, and the admittedly vague definition of "participating" is very
broad, and goes far beyond library authors.
Wouldn't it be a problem for the C++ commitee? My understanding it that the mailing list is voluntarily kept closed to avoid a lot of problem like trolling or out-of-scope subjects that occur naturally in open communities (even boost). I might be wrong, I didn't read any official info about this.
But allowing this large definition of boosters to be able to mail the C++ committee mailing list would then open a breach for them, wouldn't it?
Not everyone agrees with the closed nature of the committee mailing lists.
And there's some indication that said nature may be changing. Stay tuned... -- Dave Abrahams BoostPro Computing http://www.boostpro.com
On 10.05.2012 14:55, Beman Dawes wrote:
The advantages of joining include:
* We get to participate in formal votes. Right now, we only participate in straw votes. * Boosters would be able to attend committee meetings by right rather than as a courtesy. * Boosters would be able to participate in committee mailing lists with less hoops to jump through. * Psychologically, it may give us a bit more weight in committee matters.
Comments? Opinions?
Who would count as a "Booster" in this scheme? Sebastian
On 10/05/2012 14:55, Beman Dawes wrote:
Procedurally, that would mean joining the INCITS (www.incits.org) PL22.16, which is the US national body for C++ standardization. Boost is consider US domiciled in ISO-speak.
Who would pay the fees for this? I don't think Boost has its own funding, does it?
On Thu, May 10, 2012 at 9:16 AM, Mathias Gaunard <mathias.gaunard@ens-lyon.org> wrote:
On 10/05/2012 14:55, Beman Dawes wrote:
Procedurally, that would mean joining the INCITS (www.incits.org) PL22.16, which is the US national body for C++ standardization. Boost is consider US domiciled in ISO-speak.
Who would pay the fees for this? I don't think Boost has its own funding, does it?
Yes, Boost does have its own funding, administered by the Software Freedom Conservancy. Authorization for would come from the steering committee. BoostCon/C++Now and Google Summer of Code involve money and/or legal contracts, and the SFC provides the infrastructure that allows us to handle all that with non-profit legal status. Thanks to Bradley Kuhn for administering that at the SFC! --Beman
Beman Dawes wrote:
The relationship between Boost and the C++ Standards Committee has been strictly informal. That's been fine, particularly given that most Boost moderators were members of the standards committee and regularly attended meetings. As the standards committee gears up to accept library proposals, the relationship is getting even closer.
I believe it would be advantageous for Boost to join the C++ committee as a voting member.
The advantages of joining include:
* We get to participate in formal votes. Right now, we only participate in straw votes. * Psychologically, it may give us a bit more weight in committee matters.
Comments? Opinions?
Hmmm - wouldn't this mean that we'd have to form some consensus amongst ourselves as to how to vote? And don't we already have enough to argue about given, build systems, revision control systems, modularization, reviews, etc. etc. Don't we have all the influence we really need already? Most (if not all) of libraries added to the standard came from boost and use the boost version as a reference implementation. And I would hope that the commitee wouldn't add a new library to the standard without have a reference implementation in hand. I have to confess I don't understand the motivation behind a lot of the commitee does, so getting people like me sucked into the discussion (even indirectly) might turn out to be regretable. Robert Ramey
--Beman
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
participants (8)
-
Beman Dawes -
Dave Abrahams -
Klaim - Joël Lamotte -
Mathias Gaunard -
Michel Morin -
Nathan Ridge -
Robert Ramey -
Sebastian Redl