The Bronze Age Reflects on the Stone Age of Scientific Information Retrieval

Today I came across a reference to an article I wanted to read. For some reason, neither U Mich nor U Victoria’s libraries had electronic subscriptions to the journal it was in. So I actually went to the library, wandered around in the stacks, got lost, asked for directions, found the journal, took it to a copier, realized I didn’t have money on my card, walked to a bank machine, went into a store for smaller bills, went back to the university, discovered that the computer printing cards are different than the library copying cards, went to the desk, waited in line, bought a card, went back up to the photocopier, pressed the bound journal against the flashing glass, spun the alternately upside-down copied pages into the right direction, found a stapler and stapled it together. While I was doing all this I kept thinking “this is how I got every article I ever read in undergrad. crazy.”

I’m routinely irritated by the process of retrieving journal articles electronically. Despite the fact that the one thing researchers want you to do most is read their articles (as a very serious multiple-publication scientist I can attest to that), and despite the fact that journal articles are all tagged with a wealth of formalized metadata which should make it trivially easy to catalogue and search them (title author year keywords abstract journal volume page range) compared to most of the messy and ill-defined stuff floating around on the global networks, getting one’s hands on eprints is a huge pain. Search google scholar for keywords. Or, if you so dare, figure out your university’s tragically complicated keyword search system (choose from dropdown list of database topics, click boxes to indicate which of the meaninglessly named topic indexes you wish to search in. navigate powerfully unclever series of output screens. pray you haven’t missed any relevant journals.) Then, with full and complete citation in hand, return to school’s journal search screen. Enter exact journal name. Receive a dozen possible responses, with yours listed strangely towards the bottom. Click on it. Another window pops open allowing you to choose between accessing it electronically from your campus or some other campus. Click. Another window. Which archive would you like to access your article from? Click. Another window. Optionally you can front load the archive access with the year volume and number of the journal the article is in. But if you do, you will get the same results as if you didn’t. Click. This window shows the journal. You can try their puzzling search interface, but it’s usually just best to browse. Shuffle through all the open windows to find the one with original citation. Read the year. Click back to the ‘browse’ window. Enter the year. Switch back to the citation to find the volume. Back to the browse screen. Oh look! This journal sorts by page, not by volume. Back to the citation. Back to browse. Click on the page range. Goody, another window opens. Scan through the articles, hoping you did indeed choose the proper journal name when confronted with 12 of them. Yes, there is the article. Click on it. Another window opens asking if you want the article in text format, .tiff or .pdf. Gosh I just wonder. Click. Up pops a new window. But hey, there’s you article. Click save. Note that the default filename is an apparently random string of alphanumeric characters. Close the save window. Try to copy the article name. Discover the article is saved as an image and you don’ t have that option. Click save again. Type out the article and author names by moving the save window back and forth across the screen so that you can see the text you’re transcribing. Click OK. OK.

But I have to admit, as ridiculous a way of searching and saving explicitly catalogued and distributed files as that is, it still beats the sneaker net. And I was lucky I could find someone to give me change so I didn’t have to buy $20 worth of copy card. And on my way back up to the copiers I passed the card catalogue rack they keep on exhibit.


Hi-I was in your 531 section last term, and I hope you don’t mind me lurking around here. I’m over at SI (UMich’s library school) and also a fledgling science librarian, and so what you’re talking about here is something I think about all of the time. No, really: some days it’s like, metadata metadata oh my goodness why is it so difficult for people to get ahold of the information they need? What can I do to help someone who isn’t familiar with the specific controlled vocabulary and searching syntax of countless databases find what they’re looking for, without forcing them to endure a process which occasionally makes me feel like I want to cry? I’m not quite sure how to do that just yet. Here’s what I do know, so far–subject specialists have awesome powers few people know about.

Awesome powers? Damn, like what? Flight? Advanced keyword search? Seriously: what?

The great thing about google scholar is that its easy: you just type whatever is in your head into the little box, and chances are reasonably good that you will get something like what you want out the other end. And that is a strong power. The down side of course is that it’s wickedly imprecise. You don’t know what indices it’s searching, or more importantly, which it isn’t searching. And there is no way to tell the little box what part is the author and what is the title. And yet, that random shotgun approach seems to work, 80% of the time. And the time and clicking it saves seems to be wildy worth that 20% over the more precise, unrememberable methods offered by all the specialized academic search options.

(and hey Dianne, lurk lurk! and keep saying smart things! see you round.)

leave a comment