Perl 6 Performance and Reliability Engineering - Grant Extension Request

Category: Grants

Comments (16)


I'm all for the extension. A lot of my previously-unusable (due to segfaults) Perl 6 code now works correctly thanks to jnthn++'s fixes. More would be awesome!


I think it's a no-brainer, the improvements have been impressive thus far.

Also lots of little things I'd still like :)


Jonathan's contributions under this and previous grants have been foundational, involving fixes and improvements that few if any other members of the community could have done. The Rakudo and MoarVM of today are vastly more stable and trustworthy than before his work, and frankly we're getting a gigantic bargain.

I wholeheartedly support extending his grant; it's probably the single most effective place to put money intended to benefit Perl 6.


+1
Jonathan++ did great work on it, so I hope he will continue on it, thanks.


This is vital work. Improving the speed and reliability of Perl 6 will remove barriers to broader adoption. Jonathan has deep knowledge of all layers of the Perl 6 stack and is best placed to lead the optimisation and reliability efforts.


Jonathan's work has improved the stability of Rakudo Perl 6 and will help increase adoption and functionality.

Additionally, my work project has directly benefited from Jonathan's work on the core!


Jonathan++

It has been nice watching the barriers for adoption by devs come down. Ecosystem will also need attention to lower ops barriers.


Definitely, his work has been exemplary, both at fixing bugs and improving performance.


I also fully support the extension of the grant. It is clear to see from the grant reports he writes that he's very good at finding out even the trickiest bugs, for example concurrency-related ones.


I've already expressed my feelings about this in a comment section of the previous grant, but I'll write another comment anyway.

Please drop the performance part, work on reliability.

Performance is nice, sure, but a 10x speedup gives no benefit at all if your code segfaults after a minute.

“A lot of my previously-unusable (due to segfaults) Perl 6 code now works correctly thanks”

That's very nice to hear, really. But I've done a lot of stuff in perl 6 that is still waiting for a day when it will stop crashing. Every month or so I give this stuff another look to make sure that segfaults are not caused by stupid me accidentally doing concurrent access to something, but usually that's not the case.

Moreover, there are not many people who actually have the guts to work on that stuff. And these problems are rarely golfable to one or two lines, so it would be nice to see somebody skillful to resolve these problems (jnthn looks like a great candidate!


Awesome work!


jnthn++

i support this 100%, and am willing to contribute to the fund to keep jnthn working on perl6.


Jonathan does absolute fantastic work on MoarVM, nqp and Rakudo, which moves Perl 6 forward. He is also very kind in his interactions with the community.

Please fund this.


Much is left to do, and so much have benn done by Jonathan so far.

Keep on good work !

So, with no doubt : jnthn ++


Jonathan has done great work in improving the practical reliability, performance and maintainability of MoarVM/NQP/Rakudo. Extending the grant is probably one of the best ways in which perl6 / rakudo can be improved.


With Perl 6 finally release, the work by Jonathan and colleagues to make it stable and fast has been impressive. Jonathan knows what Rakudo/Perl6 needs to get 'there', so he's the right person for the job.


Sign in to add comment