I’ve spent the week thus far in bed, for the most part. Not the good kind of “in bed”, alas, but the kind that involves lots of achiness, hacking coughs, and occasional delirium.

I’m still not up to snuff, but at least now my temp is consistently under 100F and my skin doesn’t hurt. I probably should have gone to the doctor when my fever hit a hair over 102 Monday, but I just couldn’t muster the energy needed.

Still have a miserable cough and tons of congestion, but I’m sitting up straight. One takes progress where one finds it.

On the way plus side, Lost returns tomorrow. Yay!

While I’m here….

We had another fight at work yesterday, and again I was the odd one out (which I guess technically also means I was the troublemaker).

We’re trying to get the interface for our federated search engine set up.

I’m going to assume, rightly or wrongly, that most of you don’t know what a “federated search engine” is. You know how libraries subscribe to all sorts of specialty databases that index the scholarly literature and sometimes provide the full text of articles? A federated search engine is a front-end that will search several of those proprietary databases at once.

The federated-search product we’ve bought allows two different types of point-of-entry. The first type, which we’re calling Basic Search and which we’re probably making the default screen, allows the librarians to predefine sets of resources from which the user can choose. These Quick Sets have the advantage of making sure that the user searches at least a few (probably) relevant resources, but they are by their nature limited. There’s an additional problem in that the Basic Search screen will only support about fifteen Quick Sets without forcing the user to scroll down. This means we’re having to lump all of the engineering resources into a single Quick Set, which really isn’t great.

We’re calling the second type of point-of-entry the Advanced Search. This screen allows us a lot more flexibility, as we can define broad subject areas and then break them down into more specialized sets of resources. We can, therefore, offer an Engineering subject area and then allow the user to select a Mechanical, Civil, Environmental, or what-have-you subset. Once a subject and subset are chosen, the user is then presented with a list of appropriate databases which can be checked for inclusion into the search.

Now, I’ve been wise enough not to say this to my colleagues, but the Advanced Search is a UI nightmare. There are (at least) two search boxes at the top of the screen, and each search box has a drop-down menu allowing the user to specify which field (Author, Title, Subject, plain keyword, etc.) is being searched for that box’s search string. This may sound complicated when described, but it’s really a pretty standard interface; the search boxes/drop-downs aren’t the problem.

Beneath those search boxes are three boxes from hell. Two smaller boxes are stacked in a column occupying about 1/6 of the screen’s width, and the third very large box matches the stacked boxes in height and extends across the rest of the horizontal screenspace. In the left-hand column, the top box contains the subject areas; once a subject is chosen, the related sub-topics appear in the bottom box. Once a subtopic is chosen, the related databases appear in the big box. Once those appear, the user must choose which ones he or she wants to search, as none are selected by default. In all of these boxes, the number of options exceeds the allotted screen space, so scrolling must be performed within each box.

In using the Advanced Search, then, the user must make make selections in three different areas of the screen before being able to execute a search. This is an ugly, confusing, frustrating interface, and it cannot be simplified further without sacrificing functionality.

Now, I haven’t told my colleagues that this is an ugly, confusing, frustrating interface. I honestly believe they wouldn’t see what is ugly, confusing, and frustrating about it. One of them is in fact actively campaigning for this to be the default interface, and the third is on the fence.

This wasn’t what the argument yesterday was about, though, at least not directly. This argument was a bit more philosophical.

One colleague, the one who wants the Advanced Search interface to be the default, thinks the Quick Set interface is not only insufficient but actively dangerous. She thinks the lazy searchers will come to rely on the predefined, inherently limited Quick Sets and so will miss out on material that might reside in a more specialized database that is excluded from the subject-generalized Quick Sets.

My position, as articulated, is that the people who will be using the federated search will be either advanced searchers who will click through to the Advanced Search option if their needs aren’t met by the Basic Search or are people who would otherwise be happy with whatever they found on Google Scholar (or, more likely, plain ol’ Google). Even if the Quick Set results aren’t the best results possible, they are better than nothing.

Left unspoken was the fact, or at least position strongly suggested by the literature on information-seeking behavior, that even the most advanced users are rarely interested in finding the best possible information but instead are looking for the first pieces of information that suit their immediate needs. I think I might have ended up with a black eye if I’d interjected that into the discussion.

So here are the two positions, really: She feels that we must not inculcate habits of laziness in people who will at some point need to perform at a level beyond what the Quick Sets can support. I feel that if we deliver a product we must design something that will be usable by a typical patron. She thinks I’m pandering to the greatest common factor, I think she’s letting the best be the enemy of the good. At base, our contentions are on orthogonal planes of consideration, her being concerned with quality and me being concerned with usefulness.

So, what do you guys think? Keep in mind that I tried to represent her position fairly, but I still probably presented mine more sympathetically.

So my wireless router…no, let me back up.

Back in November, or maybe even October, the wireless card in my still-new laptop stopped working. I tried every obvious trick, then went to the manufacturers site and found that this is a known problem with the latest models in two of HP’s product lines. They’ve extended the warranty for six months on these models specifically for this problem and are even providing free shipping both ways for repairs. Nice, but I haven’t screwed up the necessary courage to go two weeks without a laptop. I’ve instead been relying on the USB wireless adapter I scavenged from my neglected desktop.

So this situation had been serving me well, if clunkily, until about a week and a half ago. In the middle of an evening, my wireless adapter and router ceased communications. This was not terribly unusual, and typically could be fixed by power cycling the modem and router. After three cycles, I decided this wasn’t working.

I plugged the ethernet cable directly into my laptop and let the router chill overnight, a tactic that had been known to work in the past. The next evening I let the router out of time-out, cycled everything back up, and was able to get the router and adapter to recognize one another. They had only entered a state of detente, though, as the router would only allow the adapter access to the (nonexistant, for practical purposes) local network. I reset the adapter, but it did no good.

Today I began an aggressive program of disconnections, power cycles, reseatings, and anything else I could think of. Finally, I got all three bits working together, however queasily. I have no idea which bit has been responsible for the breakdown in communications, but I hope the uneasy peace will last for a while.

I find myself increasingly sympathizing with Sam McGee. Right now we’re at the point where I’ve been putting off a needed trip to the grocery store because I don’t want to deal with the frigid air in my car. I am now completely out of rice, though, so something must be done.

(I got a rice cooker recently, and I think I’ve used it for at least 3/4 of the meals I’ve prepared since. It’s amazing how versatile that stuff is.)

That is all.

Do you find it delightful or frustrating that people who speak sniffily of “pandering to the lowest common denominator” are demonstrating, by use of that phrase, their severe deficiency in basic mathematics?

So we’re adding all kinds of fancy third-party apps to our web interface, and my department spent most of the day going over the variety of options available to us.

It was mostly pretty easy-going, with the occasional point of contention hashed out in a couple of rounds of “but this is why my way is better” talk, but the one truly heated argument seemed, and seems, to me entirely inexplicable.

OK, let me set up the following paragraph: when I write “cost”, I’m being all-inclusive I’m talking about monetary costs, labor costs, intangible costs associated with status and perception and such, costs of maintenance, cost of installation, the whole shebang.

So. The argument concerned a possible timing-out option. We can set sessions to time out after any given period of inactivity. We could set it to five minutes to kick off people who go to the bathroom, or we could set it to three hours to accommodate people who might break for dinner in the middle of a heavy research session. The cost for either a brief or an indefinitely long option is, as far as I can tell, zero. There is no benefit for going short, and there is no penalty for going long.

I argued for long. Actually, that’s not quite the case; “long” seems such an obvious decision that I couldn’t even muster an argument, just a series of statements of fact. One of my colleagues was strongly in favor of a medium-length period, 15-20 minutes. He could conceive of someone taking a longer break in the middle of a session, but he thought such an occurrence would be so exceedingly rare that we need not accommodate it.

I didn’t, and I don’t, get it. There is no penalty–economic or otherwise–for serving the needs of the most extreme outliers in this case, so why not do it? Why play to the middle when there is absolutely no cost for being all-inclusive?

It got ugly.

So can anybody see where my colleague might have been coming from? My other two colleagues seemed generally in his camp, actually, but not strongly enough to argue for it. This is alien to me, and I’d like to see a good explanation of the mindset.

On the way to work, I walk across a bridge overlooking a highway. Today, I happened to look down at the foot of the hill below and noticed, on the busy highway’s shoulder, a pair of blue jeans, a hoodie, and blue plaid boxers arranged in a little jumble.

They didn’t appear singed, so I assume it wasn’t a case of spontaneous human combustion. That only leaves one likely alternative, no? I reckon they’ll be signing us up for our forehead-or-wrist marks any day now.

So I’m watching this Tom Petty and the Heartbreakers concert recorded for PBS, and Stevie Nicks has come in to duet on “Stop Draggin’ My Heart Around”.

Ms. Nicks sounds just as good as ever, maybe even better. On the chorus, she started really letting go, getting into the performance as physically as she was vocally. You could see the tension in her torso, her arms, around her mouth…but her forehead and upper cheeks remained placidly smooth, taut from some pretty obvious work she’s had done. It’s good work, as far as lifts go, but the way it divorces the top of her face from the rest of her body during performance is disconcerting.

It’s a shame, really. There’s Tom Petty, wrinkles falling into the crags of his face, but just as acceptable as ever to his audience. In order to stay relevant to more or less the same folks, though, Stevie Nicks had to submit to the knife (and, perhaps, the collagen injector). At least she was able to buck the patriarchy enough to maintain a normal body shape.

Just really, really depressed. Not in the “wanting to hide under the covers” way, more the “yearning for something to look forward to” with a heavy dollop of “headpiece filled with straw” kind.

Bleh.