Saturday, April 26, 2008

Week 10-11: Troubleshooting Archon

I've been a bit remiss about keeping up on the weekly postings. This is partly because school has gotten incredibly busy, and partly because progress on the EAD project is rather slow going so I don't have a lot of regular updates to make.

As I mentioned in a previous post, we ran into trouble importing our XML files into Archon -- basically, it seems it can't be done. We were hoping this would be easily fixed, but it has turned into a significant issue (or at least, a time-consuming one). John Osborn has been diligently trying to figure out a way to work around the problem, which so far involves converting the EAD into delimited files, importing those into Archon, then exporting them as EAD (I hope I've got that right -- it's a little confusing). It's a roundabout method to accomplish what should be a straightforward task, which Archon is supposed to be able to handle. There's obviously some glitch happening, but we don't know for sure what it is.

In searching online, I haven't found as much documentation as I'd hoped about how others have implemented Archon. I did read a blog post written by a librarian who had the same problem with importing XML files, although I haven't found any information indicating the cause or solution. The main task for the remainder of the semester may well be trying to figure this out.

We've been having meetings with Jen, John, Sue, Linda, and a new apps person, Raj, to discuss the import issue and the search interface design. The meetings are great experience because they give Joanna and I an inside view of how project management and workflow actually happen on the job. I feel less like a student worker than a professional and team member, which is a good feeling.

In addition, we've also been continuing to encode finding aids for the sample set using our template. After much debate about how to handle the intellectual hierarchy of the container lists, which can be hard to discern from the finding aids, we decided to simplify it as much as possible. This means not specifying levels like series, subseries or folder, and just using numbered container tags to indicate different levels. This makes things easier for now, and the levels can always be added later.

With the semester coming to a close, we've had to revise our goals somewhat. We hope to have a small sample set completed and imported into Archon (somehow), along with a front-end search interface for delivery. But it will likely be more bare-bones than originally planned, given the Archon issues we're having. However, there is talk of Joanna and possibly myself continuing work on the project this summer, so there will be more time to do some fine-tuning.

Monday, April 14, 2008

Week 9: Making a Template

The past couple of weeks, we've been having regular work sessions with Jen. We're each working on encoding separate finding aids, and then comparing our work to catch errors and inconsistencies and answer each other's questions. This is very helpful because when Joanna and I first started doing markup we felt kind of lost and didn't know if we were doing it right. Since Jen has more experience, having her do the coding along with us seems to be speeding up the learning process. I feel much more comfortable now, due to her help as well as the amount of practice I've had since the beginning of the semester.

Joanna and I also finally got a draft done of that template we've been planning for so long. I'm sure it will end up being revised, but it's nice to know we have some form of completed document. Our work in the coming week will likely focus on tweaking the template.

Since my last post, we also had a meeting with DLS folk to go over the search/browse interface for delivering the finding aids on Archon. Sue, Linda, and Bryan will be in charge of making our design a reality. The meeting was surprisingly short; I expected there would be the need for some negotiation of our design, but Sue and Linda didn't see any problem with implementing our original idea. It will be interesting to see what they come up with.

Speaking of Archon, we also tried to import one of our completed EAD XML files into the program. Unfortunately, Archon didn't like that and wouldn't take the file. This is problematic, as the whole point of creating a template is so the resulting customized files can be uploaded directly to Archon. Sue and John Osborn are working on resolving the issue, so we're keeping our fingers crossed. We have a meeting scheduled for next week with John to discuss the problem.