Re: Build Boost Back Better

From: Niall Douglas (s_sourceforge_at_[hidden])
Date: 2022-04-04 19:27:39

On 04/04/2022 18:43, Vinnie Falco via Boost wrote:
> Well, now that I have everyone's attention just before the release,
> I'd like to draw attention to the elephant in the room. That is, the
> declining level of activity on the Boost mailing lists and the
> declining level of participation in the Boost formal review process.
> Both in terms of the number of reviewers, and in terms of the
> difficulty in finding a review manager.

The last time there was a period of declining activity both on the 
mailing list and the lack of new libraries I decided to do something 
about it, and based on results achieved I'd say it got turned around.

However unless there is somebody actively stoking the input fires, the 
pipeline doesn't pump itself, and things stop moving along.

> Clearly, a website update is necessary and along with it some type of
> campaign to highlight once again the important role that Boost serves
> in the C++ community for both professionals and amateurs alike. But
> what more can we do?
> 
> This came up before but it is worth mentioning again; in addition to a
> website update to make boost.org modern and relevant, how do we feel
> about a transition to forum-based discussion instead of the mailing
> list? I realize this will ruffle some feathers but surely the
> alternative, a descent into irrelevance due to steady declining
> activity is worse?

Similar ideas were floated last time round.

What brings life to Boost is people seeing it as relevant and current. 
If they think it is relevant and current, they participate. If they 
don't, they stay silent. I think that's the case no matter the website 
nor the discussion forum. Note how Outcome's review drew hundreds of 
responses on this mailing list. As did Beast's. As did Hana's. Those 
libraries engaged the wider community.

Get reviews pumping, then people will participate. Without changes worth 
commenting upon, people won't engage.

To get reviews pumping, you need a combination of new libraries which 
are attractive enough, and willing review managers. I solved that last 
time by drawing in young engineers who had the fresh ideas to make 
interesting libraries, and the free time to get them past review. I also 
spent a lot of time on the conference circuit cajoling people to act as 
review managers.

I stopped attending conferences four years ago, around the same time I 
stopped being Boost's Google Summer of Code admin.

Niall