Caveat Lector

The human body is a marvelously weird thing.

I cracked my toe (if that’s what I did, and I think it may be) a week ago. On Wednesday, it started turning pretty colors. Today it’s a marvelous melange of violet, scarlet, and chartreuse. As far as I can tell, this impressive polychromatism has nothing to do with me taking a friend’s advice and taping it to its neighbor for the past couple of days.

The digit does, however, have more range of motion than it did, and it hurts less; I’m going to try shoes again on Monday. I’d really like to be able to cope with being shod by the 23rd, because the notion of braving a two-day job interview in socks-and-sandals is just not on.

And that said, I’m going to go work on some PowerPoint for the interview presentation…

I’m not quite sure what it means when the person in charge of the search committee that just invited me to a face-to-face interview seems vastly more nervous about the prospect than I am.

I mean, they’ve never seen me, so it can’t be my spurious powers of intimidation at work. It could be the nature of the position; the office-politics vibes I’m getting spell “sea-change,” and that’s always nervous-making. It could well be pressure from above; the other office-politics vibes I’m getting (and that worry me, frankly) spell “unfunded/unrealistic mandate from Powers That Be.”

But I don’t know. Could be almost anything. I guess I’ll have to wait and see. And be reassuring, in the meantime. As I in fact have been.

Searching a library OPAC should be more like doing taxes.

No, I can’t believe I just said that either. But I mean it. You see, this year I succumbed to the blandishments of free e-filing, and I tell you what—I am never going back to pencil, paper, calculator, and cursing. What sent me to e-filing in the first place was the hassle of figuring out whether Form 8863 or the education deduction made us come out ahead. Gah, to hell with it, I said; let the software worry.

(The IRS and the Wisconsin Department of Revenue should be happy, too. In all the years I’ve lived in Madison, I’ve never managed to turn in a state tax form without at least one egregious error. Usually to my disfavor, too, and I thank the WIDeptRev for fixing them.)

What made e-taxes so lovely compared to the paper-and-pencil version was the quick adaptation of the software to my particular tax situation. Huge swathes of forms and questions I simply didn’t see because they don’t apply to me. I got no kids; what do I need an earned-income credit form for? Whereas with paper, all the stuff I don’t need to see, along with all the instructions for it, is still there, getting in the way.

Likewise, I was led right down the garden path to forms I did need. Now, I’ve been running TAG a few years, so I know from small-business taxes. (Well, enough to do my own, anyway.) The software was smart enough to figure out I needed Schedules C and SE from one simple question about whether I run my own business. A paper form can’t do that; all it can do is spray me with every form imaginable and make me sort through it all to find what I need.

So with OPACs. These things are just not designed for getting anything done. They’re designed to spray people with options whether they need them at the moment or not. They don’t bother to find out what the patron’s task is in order to conform to it pleasingly. They don’t even bother to explain just what tasks they can help the patron with (which is far from “all of them.”)

This business of “basic” and “advanced” and “expert” search—it’s a stupid, stupid design decision. It’s meaningless to a patron, who can be forgiven for not understanding whether the different search screens are tailored to different search types (which they’re not) or different search functions.

Naturally the full array of search options on an OPAC is overwhelming. Of course the patron shouldn’t have to confront them all at once. The patron should only see the options relevant to the current information need. Distinguish search screens by type of search task, not by some random collection of present or absent features!

(Someone in the peanut gallery is protesting that much research has gone into those screens, based on what OPAC functions patrons actually use. Know what? I don’t care. Patrons use what they understand, and the present systems are so abysmally designed that it’s a wonder they understand as much as they do.)

The first thing a library catalog ought to present the user with is a list of common tasks. Find an item by author, composer, or other creator. Find an item by title. Find a specific piece of information (which link shouldn’t take patrons to the OPAC at all—should take them to online ready-reference, or virtual-reference services). Find materials on a particular subject or topic. Et cetera. Each link then takes the patron to a tailored search screen. A really smart OPAC system remembers what the patron decided the search need was, and further tailors result screens to assist the patron in determining whether the need has been fulfilled.

That’s only the start, really. Given what we know about search behavior, we could make our systems smart enough to figure out their own precision- or recall-enhancing strategies in response to the results they pull or the patron input they get. Tax software can’t fill out a return based on the answer to one question; why should we expect OPACs to handle a complex information need based on a single line of input? We can get more input out of our patrons, if our systems are gracious and adaptable enough to convince them that every bit of effort they put in brings them closer to their goal.

As usual, though, we won’t get anywhere with this until we first give up on the ridiculous notion that The System Is $DEITY and the patrons should conform their behavior to the Almighty System in all its Holy Crotchets. (Left-anchored title searches. I ran into that a whole semester ago, and it still burns my britches. Stupid. Stupid!) Unfortunately, pace the most excellent Andrew Pace in this month’s American Libraries, I don’t see much evidence that librarians are even beginning to question the efficacy of their systems.

Oh, and Form 8863 netted us more. I have no idea why, but I’m happy to believe that the software knew, and did the right thing.

I subbed tonight at the Literacy Council’s drop-in program, and had a lovely time doing it. For confidentiality reasons I can’t talk about the learner or the specifics of what we did—but I will say that the learner made things very easy on this first-time tutor.

But. I am. Exhausted. After class, four hours of work, a LITA meeting, a hurried dinner, and tutoring. Brain fried now; a little salt and pepper and that’s good eatin’.

I got the topic for the 20-minute presentation I have to do for Ruritania U. They couldn’t have pitched a softer ball, I must say. This’ll actually be fun.

On the not-so-fun side, I just found out that the burg Ruritania U. is in has no public transportation, nil, nada, zilch. Um. This could be a problem. I’m reserving judgment until I get a sense of the size of the place and the near-campus housing situation.

I put on shoes last night to walk up to the library and back.

This turns out to have been a very bad idea. The shoes, not the library.

It’s back to sandals, probably for the rest of the week. Ow. And we have to walk to a little campus library during class today…

Well, then. I guess I really didn’t bomb that phone interview.

On-campus interview at Ruritania U. Either February 24–25 or March 3–4. (Will have to miss a How To Search Good lab session. Will live through it.) A day and a half of scoping each other out, including one 20-minute presentation from me, topic TBD. (No big. I can do 20 minutes in my sleep. Three hours, now, that takes work.)

Wow. Am not placing high on the articulacy scale at the moment. Just—wow. Honestly didn’t see myself getting quite this far this fast.

One hesitates to interpose oneself when giants get to rasslin’. This time, though, I’m going to risk getting squashed, because my usual ox has been gored and I want to talk about it.

Another tidbit in the last Cites and Insights chastised a few librarians for being happy about RSS feeds of catalogue data, since patrons who want online access to new-title lists or whatever don’t know diddly about news aggregators. (This, of course, begs the question of how long that ignorance will continue—but that isn’t where Walt went, and it’s not where his responders went, and so I won’t go there either.)

To which Jenny points out that an RSS feed doesn’t have to go directly to a patron’s news aggregator to be useful. It’s not hard to integrate an RSS feed—say, of newly-purchased titles—directly into a library website, and doing so might save website-updating time for librarians, who apparently put these data on the web by hand at present.

Whoa. Hang on. Hold the phone. By hand? If I may be permitted an emotisound, *facepalm*.

This is a hackable problem. One doesn’t need RSS to solve it, though vendor-provided RSS might well make life easier. Essentially, it boils down to three tasks:

1. telling the catalogue what data we want to display;
2. figuring out the format the catalogue returns the data in and munging it as necessary; and
3. integrating that into the web page.

Well, and maybe a fourth—setting up a cron job or daemon or whatever to poll the catalogue and update the web page automatically.

If the library has already got email alerts for new titles going out automatically, well, even easier—hack those. I would, in a hot minute, because I’m better at text-munging than I am at messing with databases.

It honestly shouldn’t take a whole lot of programming chops to set this up, assuming a minimal amount of transparency in our ILS’s underlying database. (I’m guessing a Z39.50 gateway would be enough and more than enough, and I hope it wouldn’t even get that complicated. Don’t these suckers allow SQL queries?) What it takes is recognizing that it’s a solvable problem and being willing to bash the system with a large rock, cursing the whole time, until it does what you’re asking it to. This is what I personally do to systems. I am not a trained programmer; I just beat things with rocks until they work. It’s not expertise, just bloodyminded determination to build something useful and time-saving.

I believe I have mentioned before how disappointed I am in librarians’ general unwillingness to beat things with rocks until they work? “I’m not a programmer!” doesn’t cut it, sorry. I could solve the problem just outlined with the half-semester’s worth of SQL I have and a wee bit of help with PHP. It’s just not that hard, and an ugly kludgy inexpertly-hacked solution (such as I might come up with) will suffice. No librarian, anywhere, EVER, should be putting up new-title lists by hand. The very idea is an abomination.

Just hack it, folks. Just HACK IT. If I can, you can.

Now, that said, I can think of fascinating wizardly things to do with catalogue and vendor RSS feeds. Imagine a new-titles list with reviews from a reputable review source (as opposed to some random A-a-o-.c-m customer). Imagine such lists by genre, by author, by media type (new DVDs!) or by subject. RSS is a convenient data format to mess around and play with, and if enough entities start putting out their data in it, network effects start coming into play.

So, you know, maybe Walt’s right and it’s not so big a deal that the vendors aren’t RSSing, because we’ve got other ways into the data if we’re only willing to work at it. And maybe Jenny’s right that a little RSS sugar goes a long way.

I still want to know why the hell we’re waiting for the vendors on problems like this. If the problem is data transparency (and it may well be), lean on the vendors for that rather than specific gizmos like RSS feeds, because it’s more versatile in the long run.

More things to just hack. More things to beat with rocks.

Five-minute phone call. I seem to be past the HR elimination round; it’s just a question of something popping up that would be a good fit—and I am given to understand that their managers are gearing up shortly to cherry-pick new library-school grads.

Moreover, they’ll have people at ACRL and I can meet them there and perhaps do a little scouting.

Oh, and the position that I applied for that was filled? Is un-filled. The person they offered it to didn’t take the offer. So I’m back in the running. Cool.

I have got to get off this job-hunting adrenaline rollercoaster I’m on. It’s going to wear me out.

The good news is that I’ve got another nibble, from someplace I’m very interested in working (and some CavLec readers’ ears should be burning right now). Joy and happiness from that floated me right through networking class—I climbed five or six flights of stairs to inspect the wiring in server closets without so much as noticing my sore toe.

The bad news is that they filled the position I actually applied for (with a “more experienced candidate;” oo, that smarts!) and there isn’t a damn thing in their currently-advertised openings that I want. Which isn’t necessarily horrible, because not all openings are advertised, and they must be interested in me qua me or they wouldn’t have bothered to call once they filled that spot. But still. It makes the call tomorrow (yes, there will be a call tomorrow) a wee bit awkward.

And now the neurochemicals are all signaling “crash” and I’m just plain tired. I think I will be very glad when this is all over.

I’m wearing sandals over socks to campus today, not because the weather calls for it (we’re actually having a sullen, nasty winter rain) and not because I have a burning desire to start a new fashion—but because I whacked my left little toe but good against the side-table yesterday and it hurts quite a lot to wear shoes.

I don’t think I broke it, and even if I had there’d be nothing much to do for it. I can still get where I’m going, wincing a little, and as long as I don’t whack it again or get it stepped on, I think it’ll be fine.

Nonetheless, I feel silly wearing sandals in February in the Frozen North.