Library-tool typography
I haven’t had time to comment on all the excellent writing lately about library-resource usability. I had some thoughts I wanted to kick back to Lorcan Dempsey (mostly about how having data locked up by for-profit vendors pretty much guarantees no APIs into it, so how can libraries fix the usability problems?), but they’re nothing other people can’t say better.
One question did bubble up to the top, though, and I’m as qualified as anyone else to ask it: Why is the typography and layout of library-search results so wretched?
Type a search into your favorite OPAC. What you get back is more than likely to be an HTML table, gray borders and all, with every piece of information in the same typeface, with the same font weight, at the same size, in the same color. The labels are probably boldfaced. That’s all. (Endeavor Voyager, I am so staring right at you. I was looking at a Voyager results page while I wrote this.)
My local library’s Dynix installation appears to have followed A-a-z-n.c-m in its general approach to layout and typography of individual book-metadata pages: cover thumbnail, title larger than author, which is larger than the rest of the text, and so on. It’s a start, but frankly, A-a-z-n.c-m hasn’t done much pioneering in this area either. I’m betting that if we set our minds to it, we could do better.
Labeled tables are how computers manage metadata: label, content, label, content, label, content. Let’s face it, this is not how human beings best absorb information. Human beings use good old-fashioned typography and layout. Positioning. Font choices. Color, sometimes. Pick up the five books closest to you. Look at their covers and title pages. Imagine plunking that information into boring old HTML tables. See what I’m saying? It takes longer to absorb the very same information in tabular format, because we humans pick up typography cues a lot faster than we associate labels with data.
To be fair, layout and typography aren’t all we use. The closest that human beings get to a computer-like method of metadata control is with bibliographic citations, which are ordered and rule-driven within an inch of their lives. Even so, how stupid is the idea of a Works Cited section laid out as a table, with every part of every citation labeled? What a waste of space! What a waste of the reader’s patience! Yet that’s not far from what some OPACs and databases do, and it’s exactly what every DSpace installation in creation does.
So, no, OPACs aren’t the only villains here, not by a long shot. It’s not really surprising that libraries haven’t incorporated human-oriented typography and layout into their search tools, either; ISBD came about precisely because one couldn’t professionally typeset one’s catalogue cards! Nor could one do anything much with the typography of a Telnet-based command-line tool (ask any DIALOG user).
The Web is different. Typography is all but free, and so is layout; design it right once, and every single search result benefits.
Certainly if I get my hands on a DSpace installation (which I may, one of these fine days), this will be something I look into. Save the time of the reader, said Ranganathan—which means designing our information around how that reader best absorbs it, which in turn means layout and typography.
The whole idea behind SGML and XML markup is that humans are smart about typography, whereas computers are so dumb about it that they need explicit labeling. The converse is true about tables and suchlike labeled data, which computers love and humans hate.
It’s past time we stopped designing our search-results pages for computers.