QA Hackathon 2014 in Lyon

It was awesome. Thanks so much to all who made this possible: Laurent Boivin, Book, Wendy, all the sponsors ( :, Splio, Perl6, ELIGO, Grant Street Group, Dyn, EVOZON, Campus Explorer, elasticsearch, Mongueurs de Perl), and all participants: you were great, you are great.

As has been noted, the value of the get together often is the groundwork for lifting new ideas off the ground. And so it was for me, too:

Talk with Christian Walde, Slaven Rezic and David Golden about a test module to be written that would become a generally applicable reporting module on all relevant modules for a given distribution. It would read all declared dependencies, recommendations and suggestions of said module from the MYMETA.yaml, load them all into one process and then dump the currently installed versions of everything found in %INC.

Talk with Christian Walde and Barbie that we should add some useful information to the mails that are sent out by cpantesters on failing tests. The information would have to be provided by an API at analysis and give hints about which correlations have been found by so far

Talk with Christian Walde about variables analysed at analysis. So he believes we should consider the combination of OS and perl version as a single input variable [talk about multivariate statistics]

Talk with Christian Walde about his wish to have a button on analysis to influence the next calculations

Talk with David Golden, Karen Etheridge, Ricardo Signes, Graham Knop, Kenichi Ishigaki about open pause issues and then the evaluate/accept/deploy/debug cycle on a power series of pull requests for PAUSE from Ricardo and Kenichi

Talk with Philippe Bruhat about the hack that he has put into his copy of because of the lack of proper hooks; I promised to work on hooks RSN ({pre,post}-{get,prepare,make,test,install})

Talk with Elizabeth Mattijsen and Tobias Leich about how we shall go about storing perl6 modules on the cpan; consequently evaluate/accept/deploy/debug a couple of pull requests for PAUSE from Tobias

Talk with Laurent Boivin and Salve Nilsen about how to organize a Hackathon

Talk with Jens Rehsack, Tim Bunce, Peter Rabbitson, Salve Nilsen, Slaven Rezic about how to spec external dependencies for cpan modules

Talk with David Golden about the next release of and the bugs I had encountered that were not easily translatable into tests yet and then accept pull requests from David that fixed up all the loose ends

Talk with Barbie and David Golden about the reported dates in cpan testers reports, pointing out that the date and time is based on the clock of the reporter, not on the clock of the receiver

Talk with Leon Timmermans and Slaven Rezic about a bug in Test::Harness with parallel testing that can cause deadlock

Salve Nilsen interviewed many of us during the hackathon. This was an interesting experience, I had to reflect about things we usually take a given. I'm not sure my words made much sense but it was a great and unusual challenge to speak out loud what makes our small community such a long lasting overwhelming cultural success.

Talk with Steffen Schwigon and Slaven Rezic about what Tapper is and what it provides and how to get a handle on it. The answer in short: it's a general smoker for all kind of flavours you ever would like to smoke. It's very mighty and has a steep learning curve. It may need another hackathon before I'm ready to use it productively.

And hugs, and hugs, and hugs. And then a hug from Karen (awww!)

Looking forward to see you all next year and do some Perlin' in Berlin!

Leave a comment

About Andreas Koenig

user-pic Perlin' for Berlin