Re: What is this "Beman Project Development"?

From: Niall Douglas (s_sourceforge_at_[hidden])
Date: 2024-05-09 23:11:06

On 09/05/2024 21:56, Ville Voutilainen via Boost wrote:
> On Tue, 7 May 2024 at 20:16, Vinnie Falco via Boost
> <boost_at_[hidden]> wrote:
> 
>>> I don't think the problem is getting a library into a fit state for
>>> standardisation,
>> but rather how library standardisation works at WG21 is not fit for purpose
>>> in
>> my opinion. In other words, the problem isn't a technical one, it's a
>>> _process_ and _political_ problem, in my opinion.
>>
>>
>> I agree with Niall here. The structure of WG21 creates perverse incentives,
>> producing outcomes which are not aligned with the needs of the wider C++
>> community. For example "the standard library can't connect to the internet."
> 
> I have never heard of such an incentive being expressed in WG21, and I
> have attended
> rather more of its meetings than the two people who make just slightly
> questionable claims
> about what WG21's library standardization process is fit for,
> considering how much experience
> they (don't) have about it. :)

Firstly, Vinnie's added comment I wouldn't agree with having seen the 
sausage been made in five years of attending meetings.

However, seeing as you're publicly calling me out on my lack of total 
meetings attended compared to you, have you considered that perhaps you 
have attended so many meetings you have become blind to how different 
things could be? To put bluntly: you are no longer able to see clearly 
due to institutionalization?

Back when I proposed a whole bunch of ways library standardisation could 
be done differently than at present - having seen for myself how the 
sausage is made over multiple years - if I remember rightly you were not 
keen on them. I have memories of you stating words to the effect of 
"that's not how things are done round these parts". To which I counter - 
and will continue to counter - why not? WG21 already does lots of things 
in ways which aren't strictly within ISO's rules. It's the last 
remaining modern programming language under ISO. I think they'll yield 
because they want to keep us. And if they don't, there's always the 
POSIX option. Leaving ISO has had no negative effect I've noticed on 
their standardisation work, if anything the opposite.

Maybe you do genuinely believe the current process is optimal and cannot 
be improved upon. Fair enough. I think C++ can do a lot better if there 
was more will to stop trying to solve non-technical problems with 
technical solutions, which those in WG21 keep doing.

> The mission statement of that project sounds fine, reference
> implementations for standard library proposals, early reviews. There's
> nothing there not to like. Sounds like a highly valuable service.

I wish its authors the best of luck in their endeavours. However, it's 
another example in my opinion of solving process problems with technical 
solutions. Axe meet rock.

If their project was more about changing and transforming incentives 
across the ecosystem, that would get me a lot more optimistic. Processes 
only change when enough people buy into a new process, and for that the 
incentives need to shift.

Niall