This is the thirteenth in a series covering the library system migration of the Bartlett Library at the Museum of International Folk Art and our partner library at the Laboratory of Anthropology, aka “Museum Hill Libraries.”
At last our data has been pushed and shoved into the proper formats and we can see it in the test system for the first time!
It is very easy to be distracted – as soon as I see the public catalog for the first time all I want to do is tweak the format and colors to make it prettier – but that’s not the stage we’re at right now. Really we are here to look at the data and see how it loaded.
At first glance, most of our fields seem to have gone into place very cleanly (insert giant sigh of relief). There are some tweaks we need, and surely we’ll find more over the next few weeks. There is something else to think about, though. It’s time to think like a library patron (or user, or customer, or whatever you prefer to call those people who keep us in work).
You know what your patrons need. You probably have a good idea how most of them use your catalog to search for what they want. Look at the catalog. Look at how your data appears. Is your data playing nicely with the catalog software?
Here’s an example of what I mean: our new catalog, like most, has a series of “panes” on one side of the window. If we do a search and see the results screen, one pane will say “refine your search” (in some systems it’s “narrow your search” or something like that). There are many choices for ways to refine: by author, by series, and so on. The choices – often called “facets” – our system gives are great for most libraries, but not quite right for ours.
Two fields we would really like to use to refine searches are geographic area and ethnic groups. We catalog those terms very carefully, and in the data migration we made sure to map them to indexed fields (fields that will be searched in a keyword search). But it turns out the places where we put the data aren’t supported as facets to refine a search. Whoops! Now we’re spiraling back to look at our mapping again. Is there a better place to map those fields? One that’s faceted as well as keyword-searchable?
So this is how migration goes. You get through step one, move to step two, see something funny that makes you re-do step one, move on to step three… Migration is never a straight line. Expect to circle back often. All the choices made during this transition work together. Pull one thread, change one decision, and you need to step back and watch for cascading effects everywhere else.
So here we are, eager to go on to fixing up the catalog, but needing to go back to look at the data again.