Boost logo

Boost-Maint :

Subject: Re: [Boost-maint] [boost-maint][concept_check] Pull request
From: Marshall Clow (mclow.lists_at_[hidden])
Date: 2014-02-18 23:31:22


On Feb 18, 2014, at 7:22 PM, Ahmed Charles <acharles_at_[hidden]> wrote:

> ----------------------------------------
>> From: mclow.lists_at_[hidden]
>> Date: Tue, 18 Feb 2014 19:02:26 -0800
>> To: boost-maint_at_[hidden]
>> Subject: Re: [Boost-maint] [boost-maint][concept_check] Pull request
>>
>>
>> On Feb 17, 2014, at 9:46 PM, Ahmed Charles <acharles_at_[hidden]> wrote:
>>
>>> ----------------------------------------
>>>> Date: Tue, 11 Feb 2014 07:45:54 +0000
>>>> From: dnljms_at_[hidden]
>>>> To: boost-maint_at_[hidden]
>>>> Subject: Re: [Boost-maint] [boost-maint][concept_check] Pull request
>>>>
>>>> On 9 February 2014 06:21, Ahmed Charles <acharles_at_[hidden]> wrote:
>>>>> https://github.com/boostorg/concept_check/pull/1
>>>>>
>>>>> It's just removing some left over files that were apparently deleted in master but not in develop years ago.
>>>>
>>>> This is probably a conversion error, I had the same problem in another
>>>> module with files that were deleted using CVS. CVS had a weird way of
>>>> deleting files, and the initial conversion to subversion was a bit
>>>> screwy which seems to have confused the conversion to git.
>>>> 'bibliography.html' is not in either branch in subversion:
>>>>
>>>> http://svn.boost.org/svn/boost/trunk/libs/concept_check/
>>>> http://svn.boost.org/svn/boost/branches/release/libs/concept_check/
>>>
>>> How do I get this change through the CMT process? I'm attempting to use it as a sort of litmus test, if I can't get something this simple reviewed and committed, then the process doesn't seem to be working.
>>
>> Part of this is on me, since I was at the C++ standards meetings all last week (which went from 8:30 AM until about 11 at night).
>> Sorry - I’ll get to this tonight.
>
> The reason I mentioned the 'process' is because I don't intend this to be about any individual. If you think it's on you, that simply means that the process isn't where it needs to be yet, which is fine considering how new it is.

Yes. I’m trying to lead by example here, and not doing a good job of it.

> Long term, having you be the only person that can submit changes doesn't scale, no matter how busy or not you end up being that week.
> So, I'll rephrase: Long term, what is the plan for making this process more efficient/have higher throughput or at least not depend on one person?
>
> The CMT wiki mentions 'CMT managers', how many of those are there?

Currently, that is me.
One of my top priorities is to recruit others.

The goal is to have a set of people (size unknown) who are proactive in proposing and reviewing changes, and a second group (also size unknown, but with a large overlap with the first) who are willing and able to commit other people’s changes (after they have been reviewed) and make sure that they are applied correctly, don’t break things, and get into upcoming releases. (No more having patches languish on trunk).

Since you’re unsure, I must not have made that clear enough on the CMT page.
        https://svn.boost.org/trac/boost/wiki/CommunityMaintenance

— Marshall


Boost-Maint list run by bdawes at acm dot org