Need reviewers – preparing for the first commitfest for 9.1

Now is the perfect time to get involved in Postgres development!

Starting June 15, we’re going to have a “reviewfest”, as a prelude to the first commitfest in July. We’ve already got 28 patches in the queue, and all need reviewers.

Think you’re not qualified to review patches? Think again!

From the Reviewing a Patch documentation:

If you can apply a patch and you can use the new feature, you’re already qualified to start reviewing it.

We will formally kick off a reviewfest on June 15.

We will assign reviewers and get all the patches that are queued up for 9.1 reviewed while the final touches are being applied to the 9.0 release. Have a look at Review A Patch guidelines for information about how to conduct your review. We also have a mailing list to track and recruit reviewers – pgsql-rrreviewers. (The extra R’s are for ’round-robin’)

Please subscribe to the list, and post if there is a particular patch you are interested in reviewing!

User Group Idea: Patch Review Party

On Tuesday, I invited a group of people from PDXPUG over to my house for chili, beer and patch review. PostgreSQL has what we’re calling a ‘commitfest‘ every two months where we buckle down and try to review and commit (or reject) the patches submitted over the last few weeks. Webb and Gabrielle had the original idea to get everyone together for a review party, and they did a fantastic job recruiting people to join in.

Gabrielle gave the details and lessons learned on our PUG site already, so I won’t repeat that.

One thing that occurred to me as we were doing this work was how affirming and *fun* it is to work on patch review with people in person. Several people commented on how they enjoyed doing this work in the company of others, and how the tedious issues around compiling, applying patches and going through all the questions were made so much more enjoyable with a group of good-natured hackers sitting around answering questions.

The atmosphere wasn’t pressured – I gave a little background about commitfest, how it’s been run in the past and what the development group is trying to change about it (mainly, bring in more people, and make patch review faster for people who submit patches, and smoother for the committers). Then we just got down to work in pairs or groups of three.

Working in pairs is a really good idea for this type of event. I certainly learned a few things from John, and over email and in-person again, we were able to wrap our review up a couple days later after the regular user group meeting. Having another person to bounce questions off of was invaluable for the patch that we reviewed, and it was just fun brainstorming variable names, piecing together a test case and then finding a solution to a problem we found.

Another thing that happened was that I had lots of time to chat with people I hadn’t talked with before about projects they’re working on (a really exciting materialized view implementation, and a massive cleanup of our *.bki infrastructure — two very ambitious projects!). Both people are now signed up to give talks at our local user group about their work.

I’ve talked a little bit about the social benefits of commitfest on various mailing lists, and I think the opportunity for user groups to get together and review patches as a team is a great one. I’ll be gathering up some of my other observations about PostgreSQL community and posting those over the next few weeks.

I’ve got a talk about user groups to prepare for (JPUG’s 10th anniversary in November!), so now is the perfect time for me to be gathering my experiences and thoughts from the last three years.