On Tue, Sep 28, 2010 at 8:43 PM, MZMcBride <z(a)mzmcbride.com> wrote:
Nobody seems to have replied to this thread in about a
week, which I think
indicates a problem, so I'm gonna poke a bit here.
Rob Lanphier wrote:
As you probably know, we're trying to get
into the habit of providing
a monthly overview of all WMF-sponsored engineering activity. The
September update was posted to the techblog here:
http://techblog.wikimedia.org/2010/09/wmf-engineering/
I think these updates are fantastic. And I think there's been some very real
forward progress over the past month, in a lot of tech-related areas.
Excellent!
(I
don't completely understand why Wikimedia needs two blogs, but that's a
matter for a different day.)
It probably doesn't hurt to have a place for us to nerd out and not
have to worry about writing for a general audience. It seems that
occasionally having some sort of "best of the techblog"-type summary
posting on the main blog would be a good thing to do, but that means
someone would have to decide what "best" is, and then write about it,
so it's probably not something that will happen soon.
For October,
we'd like to draft this in public so as to get the
information out a little sooner, and to give you all the opportunity
to help out. Here's where we're drafting this:
http://www.mediawiki.org/wiki/WMF_Engineering_Overview_October_2010
There's been no activity on that page since September 23 (a few hours after
this thread was started). It's nearly October 1. To me, that indicates a
problem.
Funny story there. Most of us EPMs here have been talking daily about
wanting to make sure we have some prose in place before our drafting
session tomorrow, the same way people talk about losing weight or
cleaning out their garage. That's the bad news. The good news is
that we have scheduled a drafting session tomorrow that we'll be
hammering out a draft.
The drafting session tomorrow isn't going to be public, but maybe
future ones will be. Baby steps.
Here's a
very simple way you can help. If you see something on the
list that you're interested in, but don't see the status for yet, ping
one of us, then be bold and add what you learn to the appropriate wiki
page. If you do know the status, by all means add it.
Another useful thing to do: you'll notice that many of the project
pages that the status post links to are pretty sparse. Same rules
apply there. We'd love to get help keeping this up to date.
The current reality is that the paid developers/sysadmins have been isolated
from the community. You seem to be essentially asking people to guess or
make-up the current status of a lot of these projects, most of which aren't
even known about outside paid developers/sysadmins (the Virginia data
center, improved ops monitoring, etc.).
Right now, there's such a hard line between community projects and WMF
projects that you're right, it's going to be hard for community
members to hit the ground running on a WMF project, be it as a
developer or as someone trying to document progress. I'm hoping that
over time as the newer members of the WMF staff get better at hybrid
projects, it'll also be easier for everyone to participate in this
activity.
The operations section might be a bit of an exception
(obviously a good
portion of the user base is familiar with the Pending Changes trial, for
example), but in a lot of these areas, only a small number of people know
the actual, current status. Everyone else can either guess (and then rely on
Cunningham's Law[1]) or leave the page alone (which seems to be the option
that most people are taking).
I put the invitation out there in hopes that something would happen,
but not counting on it. Ask not, get not. However, it was also an
implicit invitation to tell us why it won't work, so thanks for taking
the bait :)
Wikimedia has been having weekly (or fortnightly)
status meetings about most
of the items you listed on
MediaWiki.org, but the notes are being held on
Wikimedia's installation of EtherPad. Either document this EtherPad
installation (I'm not even sure if the URL is supposed to be public, so I'll
omit it here) or stop using it and post all of the notes directly on
MediaWiki.org. These notes in EtherPad are (by far) the most up-to-date and
helpful pages for tracking the status of projects that I've seen, but I
doubt more than a dozen people outside of Wikimedia Foundation staff have
any idea they exist. Though, perhaps a bit ironically, I haven't seen (m)any
ops-related notes on EtherPad, as far as I remember, so that still might be
an area in which only one or two people can give an accurate update.
I think our dream tool would be EtherPad realtime capabilities built
into MediaWiki.
Again, I think the progress over the past month has
been great, and I mean
that. People are opening up a bit, some secret (or "secret") channels are
being deprecated, code review is starting to pick up, and there are some
great, necessary status updates on this list and foundation-l regarding
projects that users care about.
Thank you for this encouragement! It is a source of quite a bit of
handwringing, and we all know we've still got a lot of work to do.
However, if you want people to get involved in
maintaining these blog posts,
they need to have access to the necessary resources. And even then, you'll
probably still need some sort of incentive for people to work on these. Most
companies and organizations use a paycheck, but perhaps in Wikimedia's case,
something simple like giving the primary contributor(s) the blog post byline
would push people into action.
I have been mulling the incentives. I definitely want to at least
acknowledge and thank that contribute somehow. I'm not sure exactly
how to translate that from edit history to a sensible byline. That
was actually the low point on the last update, since I really didn't
acknowledge Tomasz', Alolita's, and Danese's role in writing it
(though that was implicit in listing the EPM associated with each).
I'm pretty biased toward figuring out intrinsic reward rather than
trying to bolt on external reward. For example, one of the reasons
why I edit Wikipedia is because I'll look up something, *not* find the
answer, find it somewhere else, and then want to make sure I know
where to find the answer again.
I'm hoping we can figure out the various intrinic reasons why people
would want to chip in on this type of activity, and make sure we
reinforce those.
Another idea is to work with the "Signpost";
these people are already
pushing out a tech report weekly and have been doing so for years.
My hope is that a lot of this work is complementary to Signpost and
other sources. Part of making the drafting process public is to give
Signpost the opportunity to scoop us.
Thanks for digging this email out of the archive and responding! Look
for more from us tomorrow.
Rob