Skip to content

Post-hackathon thoughts

About a week ago we had the NYC Enterprise MediaWiki Hackathon, a two-day event that was also the first-ever enterprise MediaWiki hackathon.

What does it mean to be an enterprise MediaWiki hackathon? It means that the focus is on code that’s used by companies and other organizations that have a MediaWiki installation – as opposed to by Wikipedia and the like. In practice, that usually means MediaWiki extensions.

There have certainly been MediaWiki hackathons before – there have been about five every year since 2011 – but the focus in all of them, as far as I know, has been in some way related to Wikipedia, whether it’s the development of core MediaWiki, extensions in use on Wikipedia, tools and gadgets for Wikipedia, the visualization of Wikipedia data, etc. Which is all to the good – but there’s also a need for this other stuff.

We first discussed having an enterprise hackathon at SMWCon in Berlin, last October. There was a good amount of interest expressed at the time; and if we had had the hackathon in Europe, there probably would have been more attendees, just by the nature of things. But an event in the US was easier for me to attend and orgzine, so that’s where we did this first one. I certainly hope we can have one in Europe before too long. (We also talked about naming it a “Create Camp” instead – and there are valid arguments for calling it that – but I stuck with “hackathon” for this one just to keep things simple.)

The event was held at the NYU-Poly incubator, courtesy of Ontodia, a Semantic MediaWiki-using (and -contributing) company based there – big thanks to them.

So, how did it go? Thursday, the first day of the hackathon, coincided with an epic snowstorm that dropped a foot of snow across much of the Northeast. And Friday was Valentine’s Day. And the whole event was pretty last-minute; the dates were only set a month beforehand. So turnout was certainly curtailed; but we managed to get seven people to show up on one or both days, from New York, DC and Boston; which is not bad, I think. Nearly everyone there was a Semantic MediaWiki user, and that was a big focus of the discussion and work.

The single biggest outcome of the hackathon, in my opinion, was a set of changes to Semantic Forms that Cindy Cicalese from MITRE and I worked on, that will allow for easily displaying alias values for dropdowns, radiobuttons and the like. That’s a feature that SF users have been asking about for a long time. We also got a specific Semantic Forms implementation issue resolved; people looked into setting up a semantic triplestore (though the attempt was ultimately unsuccessful); and there were various discussions about large-scale SMW-based data architecture, skinning, and other topics.

What can we learn from all this?

  • A hackathon doesn’t need to be big. More projects are of course generally better, but we managed to get a bunch of stuff done with our small size. Having a small group helped us in getting free space, which kept costs minimal. And the round-table discussions we had at the beginning, introducing ourselves and talking about the projects we wanted to see done, might have taken a lot more time with a large group. (Or simply have been prohibitive to do.)
  • It’s good to have people think about what they want to work on ahead of time, and write their ideas on the wiki page. That helps organizers, and participants, try to plan projects out ahead of time, to maximize productivity.
  • Not all hackathon results are just code – though I’m of two minds about this one. There were some good discussions, and probably necessary ones, about various aspects of organizational MediaWiki usage. In that way, this hackathon resembled the informal part of conferences – the discussions that happen during breaks, over lunch, etc. These are often just as important as the main part of conferences, and at a hackathon you can have those kinds of discussions in a really focused way. (This hackathon was certainly not unique in that respect.) Still, as a developer, I’m focused on creating and improving code, and that to me is the real measurable output of such events. So perhaps it’ll be a while before we know the full outcome of this hackathon. But judging from people’s feedback afterwards, even time spent not writing code was time well spent.

Categories: Uncategorized.

Comment Feed

No Responses (yet)



Some HTML is OK

or, reply to this post via trackback.