This is my second Academic Publishing in Europe meeting. Last year, I wrote on the Scholarly Kitchen about my first. I'm happy to be back again, as the meeting possesses a scale and energy I find appealing and helpful for discussions.

Tuesday's sessions focused on a number of interesting topics, including data publishing, scholarly publishing infrastructure, and the financing of science and scientific publishing. Sprinkled within these and other broad topics were strong presentations on publishing technologies, a couple of futuristic visions for academic publishing, and some updates on policy discussions in Germany and Europe. There were good questions from the approximately 200 attendees.

Of course, the best part of a meeting is in the networking, and the APE 2016 meeting has this in spades, with a relaxed atmosphere, a nice dinner, and a good conference space.

I found the most compelling session Tuesday to be about scholarly publishing infrastructure, which we often take for granted (that's what you do with infrastructure that's working -- the lights come on, the faucet works). Geoffrey Bilder revisited a great reminder of some of the infrastructure we don't see, putting up a silhouette of a scholarly article, with the text blacked out, which illustrated that even the scholarly format itself is an infrastructure element -- after all, everyone could see this was a scholarly article, and easily identify the zones for title, authors, abstract, full-text, citation information, and so forth, just from the structure of the page. This may help to explain why the PDF is so persistently useful -- it captures useful infrastructure scholars rely upon.

Publishers have been helping to build effective new technology infrastructure for years now, from CrossRef to CHORUS. The way these and other similar infrastructure elements have stabilized and enhanced scholarly communication bears contemplation. From archiving to linking to public access, we are now working on a fundamentally improved infrastructure. What's going to come next? And are we sure what we have will stay and remain viable? Most importantly, will the scientific community own it? Or will commercial entities, entering through public-private partnerships, end up capturing the scholarly infrastructure? This is an issue, as academia tends to fund work in fits and starts, while commercial entities with clear goals can execute consistent strategies for years.

It's clear that we're still in the midst of a series of changes, which may never stop, but which instead speed up and slow down depending on how various trends overlap and push one another forward. It seems wise to have a solid, reliable infrastructure that we can take for granted while we deal with other changes.