Re: [Asset Stewardship] About the Beman Project

From: Andrzej Krzemienski (akrzemi1_at_[hidden])
Date: 2024-09-11 12:27:54

wt., 10 wrz 2024 o 23:34 David Sankel <camior_at_[hidden]> napisał(a):

> On Tue, Sep 10, 2024 at 2:49 AM Andrzej Krzemienski via Boost <
> boost_at_[hidden]> wrote:
>
>> Let me ask again, mostly the people that promote the Beman project, why do
>> you think Boost is unfit for the purpose of incubating libraries that
>> target the Standard Library?
>>
>
> Well, let's start with the fact that, for a long time now, Boost is not
> part of the incubation process for most libraries going into the standard.
>
> As to why this is the case, I don't think anyone knows for sure. There's
> the advent of GitHub as a C++ library distribution method, the emergence of
> CMake as the industry standard build system, the aging (and loss of
> relevance) of existing Boost libraries, the decreased tolerance of uncivil
> online behavior, the loss of key people in leadership, an increased
> prevalence of high-quality libraries being developed outside of Boost, etc.
>
> I don't want to count chickens before they hatch, but one might question
> why is the Beman project seeing success in this space in such a short
> period? It could be because it's "new and shiny", maybe it's because it's
> designed with today's environment in mind, maybe because the people on that
> project are so great to work with (that certainly keeps me engaged), or
> maybe it's because authors like getting the feedback but without the
> high-stakes Boost formal review process (they get something like that from
> the committee anyway). Who knows? We've got plenty of haters already
> predicting the demise of the Beman project.
>
> Why didn't the people working with Beman try to make changes to Boost that
> would make it more relevant for committee work? Well, several people did
> over a period of many years. Consider the Steering Committee CMake decision
> and the testing of a discourse server. There just isn't enough interest in
> the Boost community for making these changes, which is totally fine! Boost
> still serves an important purpose, but I don't know practically how to make
> it an incubator for most standard libraries while respecting its consensus
> process.
>

Thanks David!
This is a valuable and informative input. I am more comfortable talking
about technical stuff, but the social part of the things is probably as
important, if not more. If I were to summarize what you said with my words,
that would be that one can observe positive energy flowing off one project
and into another. One doesn't necessarily need to know or explain the
reasons, but it is reasonable to be where the positive energy is.

I am considering proposing my small library into Boost, and I also feel a
bit uncomfortable. First, it is my understanding that I need to come
"prepared". That is, have done the research, have a repo, docs, tests,
answers to many questions. That is ok, I guess. The thing I fear most is
that I would have to integrate my library with the Boost infrastructure. It
may be just perception, but the subprojects, b2 stuff is above my level. I
know that I would receive help. But still, the amount of the perceived
complexity puts me off.

> It's sad there's so much negative energy in the Boost project towards the
> people volunteering their free time to support it in the Boost Foundation,
> even recognizing much of it is based on misunderstandings. No one should be
> surprised why most of the board has steered clear from these discussions.
>

It is likely that I myself might also have exposed this ungrateful
attitude. If I have anything in my defense it is that I often cannot see a
good initiative if it is not sufficiently advertised or visible. When a new
library is added, I can see that. When infrastructure is improved, I am
likely to miss it.

Thanks,
&rzej;

>
> -- David
>