Hello,
Jenkins crashed again today. The first time at 6am UTC, I got it fixed.
And again between 9pm and 10pm UTC.
This has been a recurring event since we have upgraded our installation
and the bug is:
https://bugzilla.wikimedia.org/show_bug.cgi?id=48025
Tonight I got Jenkins access log enabled and made Zuul query jenkins
directly instead of passing via SSL + an Apache frontend proxy. That
will help a little bit.
The root cause is some weird issue in Jenkins where one of its thread
will use 100% CPU. I have yet to determine what that thread is doing
though nor what trigger the exact issue. Whenever I get some useful
informations I will fill a bug upstream and make sure it get attention.
Next secret plan: get rid of Jenkins..
--
Antoine "hashar" Musso
A friend of mine is co-organizing a conference about APIs. She asked me
if there is someone from Wikimedia or the community who would be
interested in participating. It will take place in Parc 55 Hotel in San
Francisco on October 23, 24, 25. A bit of information about the conference:
> APIStrat is a vendor neutral conference, and we are committed to
> promote API usage/knowledge, and specifically with the event, to build
> a great program with interesting content and actually create something
> that is useful for the community. People behind it are Kin Lane
> (@apievangelist) and 3Scale. Here you can see speakers from the
> previous event: http://apistrategyconference.com/2013NYC/speakers.php,
> where they brought together over 350 people. Videos of the previous
> event can be found here: http://www.infoq.com/api-strategy-practice/
>
> This time we are expecting between 500-600 attendees, and it will be
> both service providers and API consumers (developers). The previous
> edition was mostly addressed to providers, but we will have an
> exclusive track addressed to developers in this edition.
They'd like us to:
- talk about how much traffic we deal with, how do we do it, who uses
our API (or something similar that we think would be of general interest),
- participate in the "APIs in Government - Towards a Data Commons" panel
giving the perspective of a non-profit (they aim to have federal, state,
education, international, and non-profit representation in this panel).
Seems interesting. Is there anyone who'd be interested in giving a talk
or participating in a panel there?
--
Juliusz
https://lwn.net/Articles/548910/
A summary of talks at a recent conference on test automation, with a
bunch of links for people who want to follow up and watch videos.
> A major theme was WebDriver, which is an API for automating web browsers. Mozilla presented its work on WebDriver support in Gecko and extensions to WebDriver to allow automated testing of FirefoxOS beyond just the Gecko-powered content layer. Google talked about WebDriver support in Chrome/Chromium, including Chrome on Android. Others demonstrated FOSS software that re-purposes the WebDriver API for testing native mobile applications on Android and iOS.
We use Selenium, which makes use of WebDriver.
https://www.mediawiki.org/wiki/QA/Roadmap has more. I believe we are
not currently using any of the technologies mentioned to specifically
perform automated testing on mobile apps or mobile browsers, although I
could be wrong.
Chris McMahon was at this conference and may have more specific "we
should do foo" recommendations. :)
--
Sumana Harihareswara
Engineering Community Manager
Wikimedia Foundation
Hi,
There's currently a proposal on the English Wikipedia about creating a
"Developer's noticeboard" so that people can stay informed about
technical announcements outside of the noisy Village Pump/Technical:
https://en.wikipedia.org/wiki/Wikipedia:Village_pump_%28proposals%29#Develo…
The main argument (as I understand it) is that existing venues (like
the wikitech-ambassadors mailing list, currently used for this
purpose) don't meet the needs of users who want to be notified on
their wiki, in their watchlist.
Rather than creating yet another (enwiki-only) venue for technical
announcements, I'd prefer to find a way to use the current venues, and
augment them to mitigate their limitations.
What I currently have in mind is a bot subscribed to the
wikitech-ambassadors list, that would post every first message of a
thread to the talk page of people who have signed up (like
https://meta.wikimedia.org/wiki/Global_message_delivery/Targets/Tech_ambass…
) or to a noticeboard (that would then be mostly automated).
The bot would create a new section, using the email's subject line as
the section title, and its body as the message. It would ideally link
back to the gmane archive for that thread, so that users can read
follow-up messages there. Messages could be trimmed if they're too
long. The system could possibly also be used for other mailing lists
later.
My intuition is that this wouldn't be exceptionally hard to implement,
but I'd like a more informed opinion. Does anyone have experience with
a similar tool? Would someone be interested in giving it a try?
In the longer term, the Notifications system will hopefully end up
solving the issue in a better way, but if a quick bot hack can be a
good interim alternative, I think it's worth a shot.
--
Guillaume Paumier
Technical Communications Manager — Wikimedia Foundation
https://donate.wikimedia.org
For all Amsterdam hackathon attendees
We've made a first draft of the programme for the Amsterdam Hackathon. We
have room for 6 workshops which will be run twice over the course of the
Hackathon.
Which topics should we cover in them? Based on your motivations, we've made
a list of the top 8 topics that seem to be of interest. Please take a look
at the page and let us know which topics you'd be most interested in:
https://www.mediawiki.org/wiki/Amsterdam_Hackathon_2013/Topics
For some of these workshops we're still looking for presenters. Are you
interested? Let us know! (hackathon(a)wikimedia.nl)
Think you could fill a 1h (interactive) workshop for one of these topics,
but you're not sure? I'll help you create a workshop and help you practice
beforehand.
Finne (henna)
On behalf of the programme committee
Hello,
I'm a computer science researcher in the university of Avignon, in France. I
recently developed a software that automatically and quickly extract from an
UTF-8 text all the (longest) terms that belongs to a large set of terms.
The term extractor works as a server and I tested it successfully with a
thesaurus made of the page's titles of fr.wikipedia.org, en.wikipedia.org
and es.wikipedia.org, i.e. 9,387,079 distinct terms composed from 4,496,195
distinct words.
You are invited to test my demonstration at :
http://dev.termwatch.es/~jourlin/demo.php
The source code can be found at Github (condition of use, redistribution,
modification under the terms of the GNU Public License V3):
https://github.com/jourlin/FELTS
I roughly guessed that it could be of some interest for the development of
Mediawiki but I would very much appreciate any feedback before I look
further into that question.
Best regards,
Pierre Jourlin.