jamesgpearce

This presentation at OSCON 2014 runs through some of the things we've tightened up: https://www.youtube.com/watch?v=fzL6Zoy_ndk

Full blog post announcing the framework is here: https://code.facebook.com/posts/721586784561674/introducing-...

it's now launched; thanks for your patience :)

PS: we're going to use this text. Thanks!

Yes. Maybe this is why we are not in marketing :)

I'm involved & left some thoughts in a neighbouring comment. Also: https://github.com/walmartlabs :)

Certainly as someone in a company's open source program, we should talk :) I am too.

Sometimes even company projects reach a level of maturity at which they can feasibly graduate to community or foundational ownership.

At Facebook, both React and HHVM have a stated interest in pursuing this outcome for example - but I'm pretty sure this does not make my role redundant any more than a teacher becomes redundant once their oldest pupils leave school each year :)

For open source in general, we absolutely agree... but company-run projects aren't always as successful - and yet we feel they can be a valuable part of the overall open source landscape.

I shared a few more (personal) thoughts about what the group can achieve above in a neighbouring comment.

No, very much inclusive. We'll be getting back to all the companies interested in the very short term.

Hey all, I've been part of pulling this initiative together and thought I'd just put together some of our early thoughts.

The primary purpose is to get the open source program managers from companies together. It turns out there are a bunch of challenges for companies who want to run open source projects effectively, and we discovered we had only rarely even shared our own experiences of doing so - let alone figured out how to improve things.

A roadmap for the group's work is still to be defined... at this point you should consider this an invite to get involved. But we imagine some of the outputs might include things like:

- shared best practices for running large open source programs well

- certain criteria or qualities that 'good' projects should aspire to (responsiveness to community, maintenance, test-coverage, clear licensing, etc)

- tooling & instrumentation that makes it easier for companies to reach and maintain those standards

- a directory of projects from members that meet some or all of those criteria

I don't want to imply that we're committing to do all of these, but perhaps these non-normative examples give a bit more colour to the sort of things we'll tackle.

If you're at all interested, please sign up. We hope to have a face-to-face workshop very soon to kick off the work and understand what members want to prioritize.

Our engineers talk about lots of things (including how the architecture scales) on https://code.facebook.com/posts/

Facebook also provides a (relatively recent) dedicated library of Go libraries: https://github.com/facebookgo

Actually our recent shortcuts-for-framer project happens to use CoffeeScript, but this is very much an outlier. https://github.com/facebook/shortcuts-for-framer

Yeah, we should have embedded it to skip me, for sure :)

Also there's a blog post with more details of the library at https://code.facebook.com/posts/234067533455773/introducing-...

NB: this is a branch, rather than a fork.

The presence of the grant does not mean there is a patent covering this project. It just means that if there is, you're protected as a developer by a license to use it. (This also what the Apache 2.0 license does)

It's in the blog post that we are active contributors to the project... :)

It takes all sorts! and the best way to prove your chops is to just jump. You can fork HHVM and start by picking up a few of the easier issues and working through them. Our engineers on that project are really friendly and helpful. What can I say... dive in! What is the worst that can happen?

Yes indeed - and we use it extensively at Facebook. It just wasn't new this year.

Actually, we have a few examples of Cassandra being used here too: Buck uses it (http://facebook.github.io/buck/concept/buckconfig.html) and so do Instagram (http://planetcassandra.org/blog/post/instagram-making-the-sw...) ... so as with all of these things, we evaluate options on a case-by-case basis.

Thanks :)

I help with open source @ Facebook and wrote this article. AMA.

fixed! - thanks

Working on improving these things! Indeed we hope this deal demonstrates that.

Why? No plans to let it rest.

This is a big & exciting step forwards for Facebook and what we can offer developers.

(And we're working hard on all your concerns too...)

Working on it. There are some specific challenges around the way the Facebook SDK for JavaScript works in this context, but we'll get there.

Enjoyed. Though I disagree with 'time is running out'.

Doing these kinds of trips and long-term sabbaticals with kids is the best education you can give them.

(Stay wired a little, and apparently they barely dent career prospects too.)