Hi,
I would be happy to be a part of the LuaJIT team, moving forward.
I have a large and ambitious project of my own built on the LuaJIT engine
and philosophy. I've been fairly shy about it, for my own reasons, but this
seems like a propitious time to start discussing that work with a larger
audience.
I grew up on Pascal and Lua is my favorite flavor of Algol at the moment.
I'm invested in Lua playing a larger role in software, perhaps even
supplanting the dreaded Javascript in some future, less
financially-constrained model of the Web/Internet.
Personal matters (mostly family related) have kept me out of the work
environment for most of the summer and halted work on my own projects. I'm
an unknown in this community and frankly I can be temperamental and hard to
work with, my own fervent desires that it be otherwise being insufficient
and irrelevant.
Rather than explain myself further, I'll just toss my hat in the ring and
see what I can do to bring the community up to speed on my own work. Please
keep me in mind as a resource going forward.
Mike, I'd also like to thank you for singlehandedly rescuing the tracing
JIT approach from languishing, along with Smalltalk, in obscurity. With
some help, we'll manage to pull the state of the art into the 1980s, one of
these days.
cheers,
-Sam Putman.
mnemnion.github.io
https://gitlab.com/u/atman
github.com/mnemnion
On Fri, Jul 31, 2015 at 11:16 AM, Tim Caswell <tim@xxxxxxxxxxxxxx> wrote:
Mike, thanks for working on this for so long! I've used it in a few
projects and have always been fascinated by the implementation of luajit.
I've been studying and experimenting with VM implementations for the last
several years and would love to help where I can. I know the bytecode
format pretty well and use luajit in my luvit.io project.
I would also like to see collaboration on Github. It works great when you
have a distributed team. I've used the github + mailing list + irc model
on several project for several years with great success.
Good luck on whatever you plan on working on next.
Something that would help a lot in the transition is if you wrote more
docs about the implementation and in particular the reasons for various
technical decisions and what you learned. This knowledge can't be learned
from reading the code alone and go a long ways to prevent the brain drain
caused by you moving on to other projects.
-Tim
On Fri, Jul 31, 2015 at 9:43 AM, William Adams <
william_a_adams@xxxxxxxxxxx> wrote:
Well, ten years is a long time to do anything in tech. Thanks tons for
your contributions.
I think I might like to be a contributor, although I'm not quite sure
what I would contribute.
If the thing gets released on GitHub that might make it a no brainer to
gain community participation. Already has issue tracking, and whatnot, and
fairly easy to do change requests, complete with commenting system.
It's funny, just the other day I had this discussion with someone about
the fact that LuaJIT is the work of one guy, and what that might mean for
its longevity. And here we are.
-- William
===============================
- Shaping clay is easier than digging it out of the ground.
Date: Fri, 31 Jul 2015 15:49:33 +0200
From: mike-1507@xxxxxxxxxx
To: luajit@xxxxxxxxxxxxx
Subject: [ANN] Looking for new LuaJIT maintainers
It has been almost 10 years since the first public release of LuaJIT.
It's time for me to move on.
I'm looking for new maintainers to take over the LuaJIT project.
I would like to see LuaJIT to be well maintained and wisely evolved
into the future. I want to give it in trusted hands. A group of people
with different responsibilities would be best.
For this, the project needs to be set up differently. And, at least in
the beginning, it needs some manpower to organize people, to set up
repos, bug trackers, accounts, etc.
I'm happy to announce that CloudFlare has offered to take over
stewardship for the LuaJIT project. This will certainly aid in making
it a smooth transition.
CloudFlare has a vested interest in the future of LuaJIT: they're one
of the biggest users of LuaJIT and they've sponsored valuable LuaJIT
features in the past. I fully trust CloudFlare to fulfill this
stewardship while preserving the goals and the spirit of LuaJIT.
Please note that stewardship does not mean ownership. The maintainers
and contributors decide. A stewardship is mainly a commitment to help
organize a group of developers. And to solve some issues that may be
difficult for individual contributors.
If you would like to help shape the future of LuaJIT as part of a team
of maintainers, please apply here on the mailing list. Thank you!
All of that said: don't worry, LuaJIT is in good shape and it's NOT
going away! It's open source, it has a great community. It just needs
a new management.
Also, I'm still around, I'll answer technical questions, and I'll
promise to take a look at the really tricky bug reports. But I want to
reduce my engagement over time.
I think it should be possible for the transition phase to happen
during the next 1 or 2 months.
That's also the time frame when I'll move LuaJIT 2.1 (as-is) into beta
and cut the 2.1.0 release. Maybe a few things are missing, but 2.1 has
valuable new features and it's stable. There will be a 2.2, 2.3, etc.
for new features.
I'd be really delighted if the discussion here on the mailing list
would focus on the future of LuaJIT. Let's leave the drama and
speculation to the social networks -- they'll have a busy weekend.
--Mike
