My last post had a fairly depressing tone, as I struggled to figure out how to take my spreadsheet data and create a framework to allow people to access it. I’d had my eye on the data organization structure of Open Context since I first heard about it at the 2015 SAAs but for some reason had never just talked to Eric about hosting the data there. I guess for some reason I thought I had to do it all myself? I finally “hit bottom” after that last blog post, in terms of feeling like a reasonably capable human being, and I reached out to Eric to discuss what I was trying to do. Turns out he can totally host my data on Open Context and then I can use their API with my website so that people can pull records. So, here’s one really good lesson I’ve learned in this process: You don’t have to reinvent the wheel. Has someone already developed a system that will let you do what you need to do? Then partner with them.

Virtual Valdivia is up in Open Context

Virtual Valdivia is up in Open Context.

I did a few data conversion things (transforming number codes to text descriptors, thanks OpenRefine) and sent Eric the database. He got it up pretty quickly, and then I learned my second lesson: No matter how clean you think your data is, there are always things to tweak. We’ve gone back and forth a bit and things are getting there, but I think I still need to add one or two hierarchies so that the data sorts out in a logical way.

The next, and most important, step, was preparing a document for all the metadata and object descriptors, so people browsing the site will know what things mean. That document will be about 20 pages, when it’s all said and done. That step was a nice reminder for me, that digital projects still involve a lot of the traditional activities that we’re accustomed to (like writing). I still need to add a few more sections of information for that, but it’s getting added to OC bit by bit.

My next steps are:

  1. Finalize all data descriptors
  2. Translate the whole thing into Spanish (oh yes, it’ll be bilingual)
  3. Produce images for each of the sherds in the collection (I thought I was nearly done with this but then discovered that only about 50 of my 300 completed files had saved properly, so I’m basically having to start over. There’s my fourth lesson: Don’t assume technology is working just because it says it is).
  4. Start to play with the API to bridge the gap between OC and the VV site
  5. Add more (bilingual) background information to the Virtual Valdivia website

Before my next blog update I will definitely have the first and third tasks completed, and I’m hoping to have at least a reasonable start on numbers two and four. I’ll beef up the Virtual Valdivia site as I’m able, and should have a tidy little project to launch by the time we hit the Institute.