gevent-socketio Sprint on Wednesday Sept 18th

[Alexandre Bourget]

Hi folks,

I'd like for the community to gather on the #gevent-socketio IRC channel on Freenode.  I'll be there all day and the goal would be close up any remaining issues and release by the end of the day.

There are a couple of stalled issues in the gevent-socketio, most notably some about gunicorn and its different versions and conflicting PRs attempting to solve that.

Anyone who has a fork or has implemented things that they needed for themselves, please be there so we can pick and choose what goes in the next release, and what makes sense for the community at large.

Thanks!

Alexandre

[Lukas Hetzenecker]

I couldn't attend the meeting - is there any log available?

[Alexandre Bourget]

Weeeeelll, it was a one man meeting :)  I guess I posted to close to the date.  So logs would be someone like:

08:00> hello?

22:30> oh well

:)

Alexandre

On Wed, Sep 25, 2013 at 8:42 AM, Lukas Hetzenecker <lukas.he...@gmail.com> wrote:

I couldn't attend the meeting - is there any log available?

--
You received this message because you are subscribed to the Google Groups "gevent-socketio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gevent-socket...@googlegroups.com.
To post to this group, send email to gevent-...@googlegroups.com.
Visit this group at http://groups.google.com/group/gevent-socketio.
For more options, visit https://groups.google.com/groups/opt_out.

[Elvio Rogelio Toccalino]

I am very sorry to hear that nobody else showed up, Alexandre. I know I couldn't make it, but tried ushering people who used gevent-socketio there... Guess it didn't help.

Please don't take this the wrong way, but in my case it wasn't only that I saw the notice came too close to the date (and I did), but also that it wasn't clear what the focus would be on. You mentioned in your email some stale issues, but without detail. Just looking at the issues in github's tracker I see several issues years old, those might be stale or just plain dead.

I think it would be good to have a roadmap, even if there are no deadlines associated. If it helps, this is the list of things that I see as important and that I'd be willing to help with:

- issues TODOs
  - categorize: errors, improvements
  - prioritize: most are high priority, there's no medium :S
  - kill issues older than a year or two?
  - issues not yet tracked?
  - test deployments to hunt down new issues

- the PIP package
  - is it up to date? (it was uploaded 14 months ago)
  - what about the package description?

- core functionality
  - memory leaks
  - websocket related errors
  - core issues (disconnect failing, release threads, etc.)

- tests and docs
  - any tests failing?
  - travis failing?
  - should we add selenium tests?

- integration with other technologies
  - breaks in some cases with gunicorn, right?
  - which other servers well supported? instructions?
  - are the docs good? are there examples for full deployment?
  - how about those issues regarding python 3?

- socket.io integration
  - which versions of socket.io can be used?
  - what will happen for socket.io 1.0 (if it ever comes around)?
  - minimum examples compatible with those in socket.io/#how-to-use

- community reach
  - tutorials
  - easily accessible examples
  - demo site!

- the chat example
  - is it working? can we fix/improve it?
  - does it need to be upgraded?
  - is it too hard? can we simplified?

Discussing a list like the one above could help make it clear which way the community is pushing. Count me in for extra help.

Cheers,

Elvio

2013/9/25 Alexandre Bourget <al...@bourget.cc>

[Alexandre Bourget]

Wow, thanks for this exhaustive list.  This could be the starting point for a meeting in the coming weeks.  Maybe I could post again with that list, and we'll see which pieces people want to tackle ?

Thanks for the feedback.. I rarely organised things online and I agree it did lack in structure :)

My wife is going to give birth in a week or so, so my time is short until then..  After that, I'll be on parental leave so I'll be available for online meetings more easily.

Alexandre

[Elvio Rogelio Toccalino]

That is excellent timing, then. And the good news is that soon you'll have an extra pair of hands to help :P

If you post your list, we can surely discuss which items are more important, who can work on what, and what is best to leave for the following meeting. I think it's also a good opportunity to do spring cleaning in the tracker, and try to bring fresh, more pressing issues in.

2013/9/25 Alexandre Bourget <al...@bourget.cc>

[Alexandre Bourget]

Right, I'll put that on track.  I'd really like to be able to release to PyPI.. the release up there is getting quite old.

thanks for pushing it :)

Alexandre