Too many people have been decrying the word folksonomy for me to link to them here. Personally I think folksonomy is quite cute, but it drives me crazy anyway because it implies that a folksonomy is a taxonomy. In fact it’s nothing of the kind: a taxonomy is by definition hierarchical while a folksonomy is all about overlapping categories.

However, in one of the texts we’re reading for a KM course I found an absolutely horrendous word:


Meaning technology as applied to knowledge management. That’s so ugly it’s beautiful.

Personal KM: How do you store journal articles?

[Repurposed from a class blog. Sorry if you see this twice.]

I’m now squirreling away the PDFs of papers that I download for iSchool and once I got into the low double digits I realized that I really need a system. I haven’t found a good one yet. Anybody else?

For starters I’m trying to rename them to something logical (Norman-1993-ThingsThatMakeUsSmart.pdf, Markman-2001-Thinking.pdf).

But beyond that I need to keep them organized (1) in an inbox of assignments to be read and (2) with the courses for which they were assigned. I could also use a way to group them (3) by papers I’m writing. I’ve been bitten by the recent fad for (4) tags, so that would be nice. Since book chapters are often scanned without a title page an ideal system would (5) attach them to whatever I’ve got that resembles a standard bibliographic citation. And (6) any system I come up with should scale up and be portable for longer-term use when the semester or indeed iSchool are things of the past.

I could probably do all of the above except for (5) with aliases: put everything in a single folder and drag aliases into separate folders for each course, paper I’m writing, and topical tag. But that sounds pretty klunky, in part because keeping a lot of Finder windows open is problematic on a 12″ laptop.

So what are other people using?

iSchool day 5, Lackboard and panic at OCLC

Today is my fifth class day at the iSchool. After some rethinking of my class schedule I’ve managed to put together a stimulating and (eek!) challenging set of classes. The hardest part, I think, is going to be time management. As a half-time single parent, the time I have for studying comes in big uneven chunks not of my own choosing. On occasion I’m going to have to get a paper completed a week ahead in order to have a prayer of finishing it at all.

The biggest aggravation to date, upon which the gods of irony are no doubt smiling their wicked grins, has been the “e-learning” part of the work. It takes seven steps to burrow one’s way into the Blackboard discussion session which is required for one of my classes, with of course no e-mail or RSS export to keep tabs on what’s been going on. The electronic reserves and e-journal subscriptions are full of usability booby traps and browser issues. It’s all an excellent object lesson in how we need to improve things for our users, although my TAs swear that’s not intentional.

One of the online readings has been an interesting, if uneven, survey of the state of libraries from 2003: 2003 OCLC Environmental Scan: Pattern Recognition. I don’t know what “environmental scan” or “pattern recognition” are supposed to mean but what the heck. The report is full of statistics divorced from context, a few charts that would make Edward Tufte scream in pain, and the shifts in style that plague anything written by committee, but it makes some points worth reading. What’s interesting is the air of panic that seems to have replaced the techno-utopianism of a few years ago. There seems to be a lot of angst around the fact that people use Google more often than they use libraries (which I don’t think should in itself be alarming unless you believe it’s a zero-sum game) or that young people today do a lot of IMing and gaming. There’s a quaint sidebar about gen-X librarians vs. boomer librarians which implies that they may find common ground in their equal fogeyhood in the eyes of those scary young people.

The piece writhes with envy over the successes of and Barnes & Noble and implies that libraries should do more to emulate them, while notably leaving out the anxiety I hear from other quarters that people are buying more books now but reading them less. I don’t know that the latter claim is true, but if so it suggests that maybe libraries should claim a territory distinct from that of the big retailers.

I’d like to know more about the politics behind the sections on the “open access” movement (mostly academics who want to cut the middleman out of scholarly publishing) and on digital rights management. Last I heard OCLC wasn’t exactly in the business of free information, even if its member institutions are, so I have to wonder about a conflict of interest. There’s some talk about how DRM is extremely difficult but an inevitable development in order to make e-publishing work; the preservation section later in the report says only that DRM is “complex” and in “tension” with the open access community and stops short of saying what it really means for preservation, namely total annihilation. I’m convinced that ubiquitous DRM would in time be worse for preserving our intellectual and cultural heritage than if someone were running around burning down libraries.

Finally, for all of its obsession about Google the report couldn’t of course anticipate the news from the end of 2004 that Google is about to start scanning entire libraries and making their contents available online, beginning with those of Stanford and Michigan. Seems to me that will make bigger changes in the “environment” than everything Google has done to date.

iSchool day 1

Lou Rosenfeld has been begging me to use this blog for an exposé of the seamy underbelly of iSchool, so here you go, Lou, this is for you:

seamy underbelly of iSchool

I don’t know what this piece of equipment does but it’s about ten feet tall and was humming busily to itself under the iSchool offices when I walked by this afternoon. Is that seamy enough for you?

Day 1 of my return to student life was fun, if you can believe that. It was nice to go to a physical place to meet face to face with other people engaged in intellectual activity, something I haven’t had enough of in recent years. The faculty seemed to be enjoying themselves and welcomed participation from the class, although that was something of a challenge in one overcrowded section, and none of the students fell asleep or did any detectable IMing. Both faculty and students seemed a lot more clueful about technology than during my brief experience in the program ten years ago (which I guess is no surprise — perhaps there’s a School of Luddite Studies somewhere for which that’s not a true statement). If they were any more up to date they’d be in danger of accusations of trendiness; one course will be using a collective blog, the other a wiki, and one prof even used the bookstore that recently arranged its stock by color to illustrate a point. No explicit mention of or Flickr yet but I think they’re coming.

My schedule may be ambitious: a one-hour pass/fail general intro class, an intro to organization of information, and courses on knowledge management, information retrieval and DBMSs. No GIS course, a subject where I’d hoped to assemble a minor concentration to complement the information studies program, but that may be hard as other departments’ schedules for GIS courses conflict with the iSchool’s. And by starting out with three electives I may be in danger of a crunch when I try to squeeze in all my required courses later.

I wish I weren’t getting so many concerned looks when I tell people about my total of 13 hours, but since 9 hours is considered a normal load it doesn’t seem that far out of line to me. I admit to feeling a bit daunted by the growing reading list. I’m also uncertain about the many group projects are supposed to work, especially group presentations; will they be more like a barbershop quartet or an improv comedy troupe? But it’s good that I’ll be somewhat outside my comfort zone.

Speaking of comfort, on Tuesdays I’ve got six hours straight with no lunch break. If anyone can recommend a good brand of energy bar, preferably without a crinkly wrapper, let me know.

Indeed, a job search aggregator

Indeed is a recently launched job search site which seems to have been paying close attention to the best practices of specialized search sites in other domains. It follows the Google rule of a clean, simple interface with additional power-user features if you go looking for them.

Indeed aggregates jobs from many sources including newspaper ads and other job-hunting sites. There must be some very interesting data mining and intellectual property work under that hood.

You can subscribe to its results via e-mail or RSS feeds, with one-click shortcuts to send results to your aggregator at Bloglines or My Yahoo. You can even create a “JobRoll” for inclusion in your own site. The uses for that last feature might not be immediately obvious (would you want to post job leads in your personal blog?) but it could be very handy for student or professional organizations or for small employers that list jobs at Indeed.

Searches are geographically based, by city or zip up to 25 miles (like a dating site?) Indeed suggests related search terms and keeps track of recent searches for you, but more cleanly than pushy sites like Amazon.

And for people who enjoy obsessing over minutiae there’s a “JobGeist” feature which tracks the most numerous recent job listings by title, location and company, with up/down trend information soon to come. I note that Dallas and Houston are in the top 10, with Austin ranked at 25. (It would be interesting to see the city numbers proportional to population, and maybe sliced up by occupational field; heck, why not a map?) Among employers, Computer Sciences Corporation and the headhunter Cybercoders are way out front, with more than twice as many jobs as the next employers in line; I wonder whether that simply means that those companies provided Indeed with direct feeds and other comparably-sized employers show up only through intermediate sources? Another interesting exercise would be to try to correlate Indeed’s numbers with those from state and federal agencies which track employment.

In a quick comparison with Monster Jobs I see that Monster claims to have “hundreds of thousands” of job listings, while Indeed says it has 1,691,810 as of this morning. Monster requires a login to search and its interface is noisy, with animated GIFs, widgets all over the place, and — good lord! — display ads. The Monster features apparent at first glance that are lacking from Indeed are resume posting services for applicants, and “networking”, which I guess means discussion boards and such. Leaving those things out seems to fit with Indeed’s model of doing one thing cleanly and well.

I haven’t yet put it to the acid test of trying to find a job through Indeed, but so far it looks to me like they got everything right.

(Full disclosure: founder Rony Kahan is a friend of mine but this post is uncompensated. He didn’t even pay for lunch!)

iPhoto 4.0 gripes

I’ve been unhappy with iPhoto 3 for a while now but held off on upgrading to iPhoto 4.0 because I wasn’t sure that it would address my complaints. When I found a cheap copy of iLife 4.0 the other day I went ahead and did it. I find that I was right to be skeptical.

The biggest problem I’ve had with iPhoto 3 has been that it’s slow. iPhoto 4 seems to help, but I wouldn’t say that it’s exactly zippy now. I think I got as much improvement by the option-shift-launch iPhoto library rebuild trick as from the upgrade. I may get further improvement if I spring for more memory (an expensive choice — my 12″ Powerbook G4 has only one memory slot so to upgrade from the present 640 Mb I have to buy a full gig), but any way you look at it, iPhoto is a hog.

The other major shortcoming with iPhoto 3 has gotten worse in iPhoto 4: bloated image files when you export to the web. I was accustomed to iPhoto 3 exporting images in web albums that were 50% larger than Photoshop would produce at high quality, 250% larger than Photoshop at medium quality. That forced me every time I exported an album from iPhoto to load all the images back into Photoshop and do a “Save For Web” on each of them to bring them down to an acceptable size. I had hoped that iPhoto 4 would do some decent optimization and save me that tedious step. On the contrary: in iPhoto 4 I’m seeing bloat over Photoshop’s high- and medium-quality exports, respectively, to the tune of 120% and 500%! I don’t know what iPhoto is stuffing into those images, but the lack of image optimization makes it essentially unusable as a web publishing tool.

Finally, among the features Apple touted in iPhoto 4 was keyword support. Like everybody else on the and Flickr bandwagon I’m keen for keywords these days and thought that feature might be just the thing to help me sort out my ungainly photo collection. No such luck. There’s a keyword feature in there all right, and there might be a way to put it to use, but the interface to access it is horribly clunky. It doesn’t show up with the other metadata (title, comments, etc.) in image edit mode, nor in the cmd-i “information” dialogue, but in its own third popup dialogue. And it doesn’t allow free typing but instead requires selection from an (unsorted!) scrolling list. It’s so hard to assign keywords to photos that I probably won’t bother.

Now, of course, Apple has announced iLife 5.0, which includes a new major release of iPhoto. Unless I hear that these specific concerns have been addressed, I won’t bother buying it. As it stands I may have been better off saving 30 bucks and sticking with iPhoto 3.