Until the past year I never truly appreciated how much faster make will build an executable when you are running on some heavy iron and use the -j option. As a p5p committer I have a shell account on dromedary, a server donated by booking.com++ and maintained by Dennis Kaarsemaker and friends (more ++). On that box I set TEST_JOBS=8, which enables me to configure, build and test perl so fast that I have never bothered to time it (probably less than 5 minutes).
Today, however, was the first time that I learned that running make -j${TEST_JOBS} can obscure the results of make.
New York Perlmongers held a successful hackathon this past weekend at an event hosted by Rubenstein Tech in lower Manhattan. By our count, at least 24 people participated, coming from as far away as Harrisburg PA and eastern Connecticut.. Participants ranged from people making their first contributions to Perl-related open source software projects to current and former Perl 5 release managers and pumpkings. Every participant contributed, in part or in whole, to at least one patch to an open source project. When all patches are applied, at least six people will have made their first contributions to the Perl 5 core distribution. Other projects receiving contributions included HTTP::Tiny, Term::Readline::Perl, Module::Metadata and Catalyst, along with more streamlined procedures for creation of .rpm files for CPAN distributions and improvements in PAUSE, the Perl Authors Upload Server.
Special thanks go to Jaron Rubenstein of Rubenstein Tech — they’re hiring! — and the five other Rubenstein staffers who participated in the hackathon.
Like the St Louis Perl Hackathon organized in November of last year, this event was organized around the philosophy of being a distributed hackathon — an event that encouraged the participation of all levels of Perl developers, eschewed elaborate preparations, did not entail a big search for a venue and required minimal out-of-pocket funding. A distributed hackathon is an event which any Perlmongers group anywhere in the world ought to be able to carry off. Any number of participants greater than, say, three counts as a success, particularly if it leads participants to begin making contributions to the Perl ecosphere more consistently.
Where will we hold the next distributed hackathon? Maybe your city! We’ll be happy to help with advice on how to hold one.
A. In the Perl world, it's a meeting, generally of one day's or a weekend's length, where Perl hackers come together in one physical location (primarily) to work collectively on projects which will improve Perl, CPAN and the Perl ecosystem.
Q. What is a "distributed" hackathon? Is it something like a "distributed" source code control system?
A. In a way. A distributed hackathon, like a distributed source code control system, is designed from the outset to…
I blog about Perl.