CPAN Testers Summary - July 2010 - Earth Inferno
So last month saw the big announcement that we have now enabled CPAN Testers 2.0, and we have started to see a shift of testers over to the new HTTP submission system. However, there are still a number who are submitting high volumes that we shall now start pestering to move over. Although the regular testers are fairly easy to help setup their smokers to use the new system, it's the infrequent testers that are going to a little trickier to catch.
Currently we have a large number of report submissions that appear to be sent when someone installs a distribution they need. As it's fairly easy to add CPAN::Reporter to CPAN.pm repetoire, many have done exactly that. Catching those people, when they perhaps only submit 1 or 2 reports a month is going to be a bit tricky. However, hopefully the press release and these summaries will eventually reach those people, who can then upgrade their smoke clients to the new system. With the plan to switch off the list on 1st September, it's also likely that bounceback emails (assuming the smokers have been set up with valid email addresses) might also provide an additional incentive.
This week I've been in Pisa for YAPC::Europe, and have given a talk about CPAN Testers 2.0, which seems to have gone well. The talk covered some of the thought processes behind the move, the basic elements of the Metabase and takes a look at the new style of Fact based reports. Finally it offers some guidance of how to get involved CT2.0, either as a new tester or an existing tester. The slides are now available online.
A few weeks ago, David Golden presented thoughts behind CPAN Testers, as Free QA, at OSCON. I really recommend reading the slides, as they give a great high level view of CPAN Testers, and the concepts behind its success.
Last month the 6 monthly email requesting help with identifying testers went out. Many thanks to those who responded, and I'm sorry I can't respond to everyone individually to say thank you, but it would take quite a while. If you have received this email, and haven't had the time to respond yet, please take a few minutes to do so, as it really does help us provide better analysis. It's also never too late to respond either. It will possibly be the last such email I will be send, although there might be a last attempt to clean up old emails later in the year. With the ability for testers to register themselves into the Metabase, this email will ultimately unnecessary. On top of the people who have responded, I took some time last month to do some digging and have managed to map over 50 addresses.
This month we're mostly going to concentrate on help testers move to CT2.0, but there are lots of things planned to expand the reports and provide more functionality to the eco-system. Next month we'll hopefully be able to provide a bit more insight of what's coming next. Stay tuned.
This is the new account for incidental and summary updates to what's happening with the CPAN Testers. For all the latest news and views please see our blog.
I'm psyched to see strong progress on this super important project. The number one complaint I heard about Perl now relates to trouble installing stuff from CPAN, and better testing reporting can only help that. Thank you! --John
For me (at least part of me on Linux) old system was much simpler. All you need to enter is email. New system requires uploading file somewhere, and recalling 'transport' configuration string.
@chorny that will get easier.
For the moment once you install Metabase::Fact you can run metabase-profile and it will create the profile for you. From there we should be able to include code that will save it and update the configuration for you. The reason it doesn't at the moment is mostly because we need to do it differently for each type of smoker.
There are lots of things we can start improve now, but after the last 8 months, we need to take a step back and ensure we've got everything working right. Then we'll start moving forward again.