CPAN Testers

June proved to be an eventful month for CPAN Testers. We had several updates and passed a couple of significant milestones.

Firstly, the updates. As well as the blog and wiki updates earlier in the month, David Golden also released a new version of CPAN-Reporter. This release is notable for two reasons. It now defaults to using the Metabase, rather than the email transport method of old, and also attempts to automated as much of the Metabase ID creation as possible. In order to encourage new testers, simplifying the setup process is a must. David made significant changes to CPAN.pm regards auto-configuration, and he's incorporated the same ideas into CPAN-Reporter. Ultimately we'd like to see a common client, which uses the APIs of CPAN-Reporter and CPANPLUS-YACSmoke, and can abstract away much of the configuration and processing of smoke testing. The aim being to encourage more casual testers to send us reports from the real world, as distributions are installed or upgraded.

And so to secondly, the milestones. Our first milestone was passing 13 million test reports. I haven't made a big deal of this for reasons I will come to shortly. Our second milestone, which I should have also included in last month's summary, was the number of test reports we've had submitted this month. In May we had 670243 reports submitted, which at the time was the highest number of reports we'd ever had submitted in a single month. That subsequently got blown away for June having seen 744831 reports submitted in a single month. We have seen some new testers join, but it seems in general testers have just been contributing more. All good news.

Having so many reports submitted does mean that the Reports website builder is now running at full capacity. For the past few weeks we've seen request volumes increase, both generated and processed, and while the builder can cope, it is currently running 3-4 days behind for the oldest request. I have a new mechanism that is being worked on to help improve the bottleneck, which I'll hopefully implement sometime this month. It does mean that the plans to release the code for the Report site has been put on hold for the moment until all the improvements have been implemented.

Gabor Szabo became a regular tester last month, and suggested some ideas for ways to promote CPAN Testers. Following a blog post here and on Perl Blogs, we had several suggestions for things we could do. Andy Lester suggested some ideas for articles, and I plan to work on some of these over the coming months. While these summaries are primarily for the CPAN Testers echo chamber, it would be useful to have some articles and blog posts that could reach out further than the CPAN Testers or Perl community. We have lots of good things to promote, we just need to promote the project in the right way. The numbers might be interesting for analysts, but to the regular user it's more important to promote why having CPAN Testers is a great thing for CPAN and Perl.

If you have any stories where testers have helped you, I'd be very interested to hear from you. If you post an entry in your own blog, please let me know and I'll include a link to it here too. As an idea, David Golden posted his experience some time ago. How have CPAN Testers helped you?

Which brings me back to the 13 million reports milestone. Following the discussion surrounding promotion of CPAN Testers, while these sorts of milestones are obviously of interest to CPAN Testers, with the volumes of reports we are now seeing submitted, the million milestones are now too frequent to make an impact, especially if we start seeing them almost on a monthly basis. As such I'll be making more of a splash about the 15 millionth and 20 millionth report submitted. You can still keep an eye on the milestones we do pass on the Interesting Statistics page.

Which brings me to another update. Several suggestions, includng ones from Gabor Szabo and David Golden, prompted me to look at adding them to the Statistics site, as part of a major rework I had already begun. Some of the pages on the Statistics sites are viewed very infrequently, but take a significant amount of time to construct. The site is built roughly every 4 hours, however, the rework now allows these lesser used pages to be generated just once a day. I'll have a more indepth explanation of the changes in another post soon.

And finally...looking forward to next month we have YAPC::Europe, where both Léon and myself will be presenting CPAN Testers talks. While I will look at the bigger picture of why we have CPAN Testers and how to get involved, Léon will be looking at a specific example where CPAN Testers helped him to develop his code.

We've had several suggestions for new sites, pages, articles, images and code recently, so look out for some news and updates very soon. until next time, happy testing :)

Cross-posted from the CPAN Testers Blog.