The Perl Toolchain Summit 2026

By Paul Johnson on May 9, 2026 9:36 PM

I was once again privileged to be able to attend this year's Perl Toolchain Summit. This is the 13th year (in a row if you discount the Covid years) that I have been able to attend and it is the technical highlight of my year.

This year the event was held in Vienna and, for the first time, my wife accompanied me. We took a direct train from Zürich to Vienna and had a wonderful trip through the glorious Swiss and Austrian countrysides.

We arrived fairly late on Wednesday evening so didn't meet up with anyone then, but we saw a few of the other attendees at breakfast the next morning, and then I set off for the venue where I met up with everyone else, heard BooK's opening speech, took part in the introductions and then split off into a room with the MetaCPAN group with whom I spent about half of my time. Meanwhile, my wife set off to explore Vienna.

I came with a set of work needing to be completed and things I wanted to discuss with people there. The discussions were the most important part and are the raison d'être of the event, and over the four days I had plenty of useful discussions, both planned and non-planned. The first one started whilst walking to the venue on the first day with Paul Evans. We discussed the possibility of separating the behaviour of $^P and PL_perldb - perhaps by using another bit. Devel::Cover uses $^P but doesn't want the behaviour of PL_perldb.

I also discussed further integrating the cpancover infrastructure with that of MetaCPAN. Getting cpancover to run entirely within containers was one of my goals from last year's summit. It wasn't finished during the summit though the problems were solved and all that remained was finishing the implementation, which happened during the year. So now the discussion was more about how to use that. I had a good chat with Leo on that topic and we also sorted out the cpancover backups to MetaCPAN.

I didn't just speak to the English folk though. I spent a fair amount of time talking with Salve on topics primarily related to CPAN security and adjacent subjects. And I chatted with Merijn about life, the universe and everything. He also tested Devel::Cover 2.00 for me and gave me good feedback. We held our second annual CLI throwdown. Many of us live in the terminal so we have a (now somewhat formalised) tradition of sharing our favourite tools.

I had wondered whether I might be in a position to release Devel::Cover 2.00 during the summit but, whilst I was able to do a fair amount of work on it, I realised that there were more things to look at. Hopefully it won't be too much longer before I can put out a test release. This will be the first major (breaking) release since Devel::Cover became stable and it should bring speed improvements, a new criterion and a new report style.

There were plenty of talks and group discussions which I enjoyed, and the four days seemed to race by. It's an intense time which genuinely takes its toll. You can tell that by Sunday many folk are feeling the effects as things start to wind down. Nevertheless it's also extremely satisfying and productive. As usual I left with far more to do than when I arrived - plenty to keep me going until next year.

Beyond work and technical discussions it was lovely to meet up with old friends, to travel around the beautiful city of Vienna and sample some of its lovely cafés and restaurants.

The venue was excellent, die Hauswirtschaft. Our part was small enough that we weren't too far away from each other yet large enough that we weren't on top of each other and with enough space to hold group discussions without disturbing those not involved.

Huge thanks to the organisers. From my point of view, everything ran without a hitch though I'm sure an awful lot of effort went into making it seem that way. And thanks also to our generous sponsors without whom the event could never have happened: