University API Un-conference Debrief

I recently attended a new conference (well new to me anyway) called University API Un-conference. This is the 3rd time it’s run at BYU and is sponsored by the CIO / BYU Enterprise Architects group. They didn’t ask me to write this but in favor of transparency they sponsored me coming out to talk to expose attendees to the ideas driving ELMS, especially in areas of xAPI.

But anyway, I’ve been to plenty of “unconference”-ish events given that Drupal is usually pretty informal, hosting Birds of a Feather (BoF) rooms where anyone can write down a topic and meet informally. Room number, whiteboard, and go. I was pleasantly surprised by the quality of a facilitator lead unconference!

We started the day writing a brief topic descriptions on big sheets of paper, then we sat in a circle and gave like a 20 second pitch of what our topic idea was. Then we put them up on a board based on room and timeslot we wanted. It was an interesting collaboration and meet and greet in one. There were also ground rules associated with no worrying about who was in your session or how many and also not getting offended if people moved around.

Topics were all over the place. From university API (they have an API-frist architecture that all systems integrate with, it’s pretty incredible in and of itself), “help me learn about API” style working sessions, pet projects, LRS development, documentation talks, Event management APIs, ELMS / OER family of topics; really anything edtech builders it was there. As I expected, there were several past Apereo attendees / member organizations including Duke represented. I figured as much because I felt incredibly comfortable with the audience, people knew how to build things instead of just buy them, and the open culture really creates a vibe not present at some other edu events.

My big take aways:
– Swagger is something I really need to learn more about. BYU has a system similar to the elmsln read the docs integration where documentation can be ingested and replicated in a drupal site (funny, same workflow, same starting point, same endpoint, different bridge code to do it). It would really help us with our Polymer elements long term I think since it’s for building and documenting APIs clearly.
– Tone is incredibly important. During an OER discussion, I mentioned the notion of overthrowing past governence structures (said much more flippently then this even) and someone challenged me as to what I meant by it. She then described what she thought I was saying, which was the same thing basically but sounded way less threatening. This stuff only happen in Open and when all parties are willing to question and be questioned as equals.
– API centered design, several years after being put into practice, is something incredible to behold. Day 1, it can be painful and those painpoints were expressed, but now a few years out it’s rather amazing the kinds of integrations / solutions BYU can craft as a result of this approach.
– The idea of an “event” / “activity” API endpoint is something I’ve never thought of before and something they have working very well (even if it’s early on) at BYU. The idea is that any action that you want to track / report of any kind gets sent to an events endpoint. Anyone can create a type of event / action and supply the data to back it up
– Sovrin – I’ve never heard of it before but Sovrin the technology for global distributed identity (think blockchain but for identifying individuals securely across any solution out there) is a pretty awesome concept that’s gaining some traction. There were several discussions around the idea of Sovrin the technology and Sovereignty as individual student ownership of data and empowering students to be able to take their credentials / skills / badges with them. They are calling this Learner Profile and while it’s still in the early dreaming phases, I understand why we see eye to eye on so many things. Once you start thinking in terms of redecentralization, you can’t unthink it.
– I love hallway / lunch conversations. I met someone who had a PhD in the History of Technology; which is an STS topic that’s right up my alley from my Information Altruism work.
– Get off the island is more then just code – I was able to provide a unique perspective to attendees (and them to me) as a result of my bulk of experiences. Several times I referenced the Writing conference I attended (cause no one there did), Apereo / Sakai work (which I’ve been to and others have / use Sakai so shared experience from multiple perspectives), and being a hard-core Drupal developer when many people were sysops; which gives me a unique perspective from the “inside” of a very different open source community from the open culture of the event. Overall, this really validated that I need attend less Drupal events with people I’m immediately at ease with and instead try and search out these hidden gem events.

I don’t require it, but wow does it feel good sometimes. It was very nice to be some place with people I didn’t know, who didn’t know me or my work, and have them see the totality of the work our team has been doing (both PSU college centric and ELMS:LN developer centric) and be floored by it. It’s funny, I kept saying how blown away I was with the level of open culture exhibited and the API infrastructure they all were apart of, while at the same time people kept telling me my work was the first sign of a way forward beyond the thinking of the LMS without breaking the student / faculty cognitive load issue of multiple solutions. I had a couple people tell me this was the first tangible example of a NGDLE they had ever seen and I had many long conversations
Other things:
– People are looking for what OERSchema provides but not realizing it. In describing the problem it solves and demo’ing it to a small group they were very impressed with the idea. Always fun to present someone else’s work and have it validated.
– We’re on the right track with Polymer, BYU’s been doing it with different components in the form of straight up webcomponents for some time and have been happy with the results.
– ELMS:LN is maturing to a point that I can present the idea at a high level and delve into the trenches and not have to hide anything along the way

Going forward / Next steps
– Hopefully we can get another voice from Duke to vouch for us to the Open Apereo foundation board of organizations. We need 1 more voice to say ELMSLN should be accepted into their incubation program. This would help the project gain visibility but more importantly provide our community mentorship in what practices we could be doing a better job of with getting people on-boarded and sustain efforts.
– We need to do a better job justifying xAPI in the age of Caliper among some people, but because of our worldview with regards to data (anything is a course, anything an action, anything learning, not just what happens in higher education LMSs) xAPI will always be more flexible. It might be messier, but it’s going to always allow for more flexibility in design over Caliper. This isn’t to say we shouldn’t be open to supporting Caliper, but it might be more of a filter that converts xAPI statements into Caliper ones based on skimming off the ELMSLN context properties we’ll mixing into xAPI to extend it.
– Standards all the things; We need to work more on the external tool integration aspect as well as supporting LMS APIs natively. This would be the next step in suggesting this system is higher then the LMS in what it can do and what it’s meant to provide as an ecosystem driven experience.
– Libraries need an OER hub / routing solution: Members of Duke & BYU had an interesting perspective on the OERSchema / Git ingestion workflow for markdown files. What if a library was the curation engine for OER to vet it for quality, and then much like an old library, provided an open bookshelf of sorts that people could checkout the books from. In this connotation, checking out a book would be cloning the repo so now you have a copy. This would allow distribution and help ensure knowledge never dissipates (a critique Michael Collins has of Database driven / single point blogging platforms / silos). OERSchema becomes icing on the cake as it would incentivize faculty to not just tell people to draw from an online text but allow them to legally and easily pull the book apart and remix it into new forms

Overall this was an awesome event and everyone was very approachable. Thank you all so much more having me and I can’t recommend this event enough and hope it comes out East Coast (Duke tossed out the idea of hosting toward the end of the year which I’d be up for coming down to).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s