Subject: Re: [boost] Attn: New Boost library policy text ready for approval
From: Niall Douglas (s_sourceforge_at_[hidden])
Date: 2017-03-20 17:23:52
On 20/03/2017 15:38, Peter Dimov via Boost wrote:
> Niall Douglas wrote:
>> try https://boost-website.nedprod.com/development/submissions.html again.
> This looks good to me. The new procedure makes sense. I only have one
> small nit to pick:
> "Be prepared to pivot your design"
> Can we please not pivot? Pivoting is awful. :-)
I am afraid I do not understand as pivoting is exactly the right term
for what was meant. Can you suggest something else?
> This is off-topic, but I'd really appreciate if the page linked in
> "Some best practices ideas with samples of script and code and links
> into source code in existing Boost libraries can be found on the Boost
> be reworked, made up to date, with the controversial/niche
> recommendations removed. It'd be nice if it reflected practices that are
> unequivocally endorsed by Boost.
Lots of people have said how useful they have found that guide as a
source of ideas and considerations. I don't think anybody actually
implemented any of those practices using the suggested code or formulation.
In that sense the handbook has been highly successful, in that it has
failed in a inspirationally positive way. Updating it, without investing
very considerable additional effort, might therefore be
counterproductive because by removing the controversial stuff, you
remove the food for thought (and remember nobody is implementing the
controversial stuff anyway, but they are finding it inspirational when
deciding to reject it).
> Back on topic, I think that the current process of getting a library
> into the review queue is a bit outdated. I suggest we make use of
> existing infrastructure and make a Github repository "review" owned by
> the Review Wizard in which submissions occur by way of the endorsing
> Boost member creating an issue with the description of the library.
The problem with this is that we don't want one member endorsing a
library for review. We want *lots*.
Almost without doubt when potential review managers scan the list of
review pending libraries, they will prioritise those libraries with the
most public endorsements.
I think that is one of the most valuable gains in this change, and I
would not want to lose it.
> Discussion about the library, as it pertains to the review process, can
> then happen inside this issue; review managers, when found, and
> scheduled review dates can also be posted there, so as the progress of a
> library towards a review can be conveniently tracked by people with an
> interest in the matter.
Github allows people to star projects, but not star issues.
Can you reformulate your idea to be something starrable on github? Or
perhaps not on github? Remember
-- ned Productions Limited Consulting http://www.nedproductions.biz/ http://ie.linkedin.com/in/nialldouglas/
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk