I actually never participated that much,
so this is just my outside view:
- mailing lists are terrible technology in year 2022(actually in year 2012
they were already bad), beyond simple queries stuff is hard to follow, e.g.
thread hijacking is easily solvable with "new" technologies(available for
15+ years) where you can move that part of conversation easily out from
original discussion. mutability ftw ;)
- C++ standardization has taken the wrong path of making happy trillion
dollar corporations that do not want to spend millions to update their
codebases while migrating to new versions of C++ and that is not really
something that inspires people to do free work.
- Rust - I know some people hate Rust and consider it a fad, but it is
absolutely true in my experience that a lot of my coworkers are much more
excited about Rust than C++ 23/26, although this is a mix of Rust fanboy
stuff and disappointment in C++ standardization speed.
- Titus is a merciful god but you have angered him by rejecting his ABI
breakage proposal ;) Since Google gave up on C++ it is kind of hard to get
excited about future of C++.
One important thing about standardization: I think primary problem is lack
of resources(e.g. Sankel talk about Rust/C++ mentions the speed of adoption
of features, only in C++ land 3 year period is considered a small cleanup
release), and I am grateful to everybody who donates their free time and
expertise to standardization, but but but it is also true that
standardization choose terrible path of stable ABI, co_ nonsense...
C++ choose slow certain death over highly likely survival in case of going
for breaking changes.
And just to be clear: I am not talking about next 2-3 years, more about
next decades so I am certain there is still a lot of value to provide for
people who work on standardization and/or in Boost.
P.S. Regarding resources: I know money is considered dirty in most open
source communities(I never understood that), but if Boost would set up some
way to support people doing work and resources were spend transparently I
think that would be beneficial to the community and Boost itself. Again I
might be too optimistic, mold linker author recently mentioned how he is
unable to support mold development from donations, but you can count on
mine 5$/month :)
On Sat, Dec 3, 2022 at 3:03 AM David Sankel via Boost
Hi all,
At C++Now this year I gave a talk on the status of Boost and presented this graph which illustrated the trend of Boost mailing list participation.
https://docs.google.com/presentation/d/1ocX8Dh4B98gxfXJTWOriaEU1tmJN61M4RfjU...
I'm interested to hear your opinions as to what happened over that time period to cause such a trend. There's truth to the claim that this question is being asked of the wrong people; those who left, stopped participating, or would otherwise start probably have good insight. Nevertheless, I think it would be interesting for us to brainstorm on this.
Thanks in advance for your thoughts.
-- David
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost