[Members] 2011-08-17 Board meeting minutes

jehan at zemarmot.net jehan at zemarmot.net
Wed Aug 17 19:59:30 UTC 2011


On Wed, 17 Aug 2011 14:08:03 -0400, bear <bear42 at gmail.com> wrote:
> Minutes, XSF Board meeting
> 2011-08-17
> xsf at muc.xmpp.org
> http://xmpp.org:5290/muc_log/muc.xmpp.org/xsf/110817

Damn! I knew I would miss it. I took a _small_ nap, thinking I would
wake up on time for meeting, and… I wake up just after. I so wanted to
be here! I had things I wanted to say on several points. That sucks that
Asia is the unloved child of XSF meeting times.

Late, I will still add notes below under some items.

> Attendees:
> - Mike Taylor (Board chair)
> - Peter Saint-Andre (executive director)
> - Will Sheward (Board member)
> - Florian Jensen (Board member)
> Absent:
> - Jack Moffitt (Board member)
> - Nicolas Verite (Board member) (partial attendance)
> Agenda:
>  CommTeam members?
>  - Team Lead has been missing in action
>  - No objections to adding Ludovic and giving him permissions to write
> blog posts etc.

I am not in comm-team, but yes I'd say we would definitely need to have
more "comm" activity! Like someone said about the GsoC, and which is
true in general: XMPP has not enough visibility and we are definitely
not transparent about our internal activity.
So it would be nice to encourage participation from anyone who proposes
oneself (like Ludovic apparently)!

>  - Best practice is to write a post and ask for reviews on the CommTeam list
>  - Nÿco joined the meeting and i've asked Ludovic to talk to him
> directly about joining the CommTeam
>  Discuss Jingle Interop event
>  - there is a request to hold an interop

As it was from me, more details below about what I would have told if I
were at the meeting.

>  - when would the next good time be?
>  - virtual interop in October or November, physical interop at next Summit

Yes I think we don't necessarily need it to be "physical" at first
(though it can be later as proposed). I think October is a bit early.
Maybe November or December is indeed better, in my opinion because there
is much to do if we don't want to do a useless event.

>  - have neustradamus email the members list to start working on an
> online interop agenda
>  - Florian to help coordinate

As I said in my original emails on members@, I am also willing to help
organize, but this time it should not be like the previous interop (or
at least what I read from it on the wiki) which really was kind of *sad*
and totally pointless.
I have already prepared some notes (as I was suggested the idea of the
event 2 months ago, I quite a lot thought about it already and made some
TODO list lying in my hard drive). I will put them down in an email and
send my proposition separately, probably alongside a wiki page. But
tomorrow. Now is like 5h AM. And I have a work meeting to attend at 9h30

>  XMPP Authentication plugin for WordPress on the website
>  - does the site even need this to be added, consensus is no on adding it
>  - Peter will be testing it on his blog

About this plugin. Ludovic, who is very enthousiastic about it and I
thank him, added the item to the meeting. For my own, I was planning to
tell during this meeting that I wanted to propose to install the plugin
*later*, probably when it reaches version 0.4 (currently 0.2) as I have
several cleaning/rewrite to do (the current version is more a "working"

So now why do I think this plugin would be great for xmpp.org:

1/ I did this plugin originally *for* the XSF and xmpp.org in
particular (and my own website as secondary target). I give this
precision because I want to make clear I am not trying to use xmpp.org
for publicity (not that I would think it would be bad if someone did
this, mind you). It was simply the target of my development from the
start in my mind.
I am even willing to give my copyright to the XSF if needed.

2/ I think XMPP is definitely under-exploited. We have a great platform
but most people (most of us included for what I see) use it for its most
basic use only, which is IM (and derived: MUC, audio/video chat…), even
though we have so many other nice XEPs lying around. I think it's sad.
Hence if there is at least one place to make live demonstration of the
interest and power of XMPP on other features, that's xmpp.org itself. If
the XSF is not even willing to use its own technologies on its own run
services, how do we expect to have any adoption from others?

3/ This extension in particular (XEP-0070) is very nice in my own
opinion. There exists openID, which kinda failed because it was using
HTTP urls as identity. Then our current Internet is filled with
proprietary systems (like Facebook and all) that are trying to imprison
users into making advertisement from us. Recently Mozilla started
drafting its own browserID, using again HTTP, but also email (which is
unsecure and not instant by nature) as the identity, and again an
"identity provider" concept as a third party. I have not finished
reading the spec of browserID, but it does not look like the best
choices from what I read until now.

Meanwhile in XSF, we have XEP-0070, a draft status in our Standards
track, waiting since 2005 for being used. Searching the web, I have
found a few, more or less complete, implementations from clients, and
only a single server-side implementation (and that was to connect to an
openID server! That's definitely a step to pass over). I think the XEP
is quite secure (considering the whole very secure XMPP background),
does not need any third party provider that we need to trust/subscribe
on, and a quite nice use flow too (don't hesitate to try it on the demo
website I made especially for you to try!). Peter tried, I think he
liked the flow using Psi (it's nice with Gajim too, though Psi+ has a
nicer pop-up). Maybe you can ask him. Or make an opinion for yourself on
my demo!
If we want to demonstrate XMPP, let's start with xmpp.org!

4/ Also we currently do not allow comments on xmpp.org (which I
understand is for spam. And I understand that none wants to really
bother cleaning the mess here). This plugin would be quite a nice filter
for spam and would allow to finally comment on xmpp.org which I find
would be great.

5/ Finally for people who don't trust my code enough (it's Free though,
review at will) and as it messes with authentication, know that features
can be separately enabled. You can enable only the commenter auth part,
and disable the user auth part at first, if you wish so (note that the
version-to-come even gives possibility to each user to disable in one's
profile the XMPP auth, or at the opposite disable password profile).

> Jappix on xmpp.org or jabber.org
> - Jappix is a social network with an XMPP component
> - Ludovic to start a discussion on the mailing list (it is possible to
> have the name of the ML?)

I have not tried Jappix yet, but I think that would be a great idea to
finally have a good web client (from the few I saw, Jappix seemed to
give me good hope. But I saw there are quite a few other emerging too.
Some comparison tests might be worth it before choosing) on jabber.org.
For years the only usable web client was jwchat, which has a terrible UI
and poor user experience. It is time we go for better.
As I said for the Auth plugin, the XSF should lead innovation and
demonstrate the various XMPP usages for adoption. Plus, that's just a
web client. We can change it later if we want to give equal chances to
all dev teams, like we do for our servers.

> GSoC update
> - Mike completely blew the task of sending books to the students
> - Kev and Remko want to go to the Mentor summit and are asking if the
> XSF to help with travel costs up front. Consensus is yes, Peter to
> coordinate with Kev

And that's it!
Sorry for the long email, but I really wish I did not sleep through the


More information about the Members mailing list