A Librarian Reads the News

Publications and Presentations

In my classroom...

Proving non-existence, and other impossible things

If someone knows something has been published, either a specific thing or a category of things on a topic, I’m pretty decent at pointing them toward that thing. I’ve had a whole lot of practice with this set of skills — I’ve been doing this day in and day out for twelve and a half years, now. I’ve done this more often than I’ve done nearly anything else in my life, including things like “go to bed at night” and “eat a meal.”

But for a whole category of things, all this experience simply means that I have a slightly better sense of when to give up and suggest that maybe the thing we’re looking for doesn’t exist. I’ve looked enough times for biographical information or for the personal papers of even relatively famous people to know that many, many people just don’t have much of a textual/audio/video/pictorial footprint after they die, even in relatively recent times and in pretty developed countries. Lots and lots of organizations don’t maintain extensive records/archives, especially not records we have access to. For that matter, many governments don’t share much information, or if they do it’s not guaranteed that it’ll be available online.

But this is such an uncomfortable outcome for a research question because how can I be sure that the thing doesn’t actually exist? How do I know I’m done searching? How do I know we’ve looked in the right places? It’s utterly nerve-wracking to be faced with potential non-existence, especially these days when we’ve grown so used to more and more and more being available. We’ve grown so used to living in a world where we have to deliberately keep personal information private that the world of our grandparents, let alone our great grandparents, is almost incomprehensible. Depending on particularly socially conscious family members to submit birth, wedding, and death announcements to local newspapers, or partnerships between families and archives to preserve select family papers — all of this simply doesn’t make sense to most of us in our current lives.

So I spent quality time last week trying to figure out if the blind reviewer on a faculty member’s manuscript simply mis-remembered the number of articles by a particular author in a particular publication, or if our lack of indexing for that publication was keeping some of this potential treasure trove of publications from surfacing. This week I’ve spent a few hours trying to figure out if a relatively famous Jesuit priest and Hiroshima A-bomb survivor left behind any letters or other writings before he died. In each case I found something, but nothing like the amount I had hoped for, which felt like failure, but also felt like the kind of failure I’ve experienced over and over in situations like these.

I wonder what other librarians’ criteria are for declaring a thing “likely not to exist.” I wonder how much subject knowledge vs how much knowledge of publication processes someone needs to do this well.

I know that for me there’s a weird circular logic involved in which if I can’t find the thing and I also can’t find people citing the thing, then I’m more likely to declare it non-existent. This is true even though I know that there are plenty of things that haven’t been cited but that do indeed exist. So I’ll start with the likely places (depending on the thing, this could be one or more research databases, or my good friend Google, or with a human expert). Then I’ll move to the less likely places and/or strip away anything in my searches that might have been mis-remembered or misspelled. (At this point there’s usually a whole bunch of paging through really weird result lists in hopes of finding anything that would point me in the right direction.) And I’ll search for anyone writing about whatever I’m hoping to find, hoping for clues about what they saw and where they saw it. And finally I’ll search through more context-less citations, first trying Web of Knowledge citation searching, but always also moving through Google Scholar, Google Books (you get different results if you go to Google Books rather than relying on just regular Google), and our free-text-search databases like JSTOR and Project Muse and the like. And at this point I always strip things down to just the author’s last name (if known – some citation styles don’t include first names) and other things that are less likely to be changed or abbreviated. This usually includes paging through tons of completely wonky results.

So basically, this mopping up of possible citations can take a whole ton of time, most of which is spent wondering whether you just have the wrong information about the thing, or if the thing actually doesn’t exist. And I haven’t figured out any way of speeding up this process or of predicting the outcome in less than several hours. So I guess in the absence of better strategies the only thing I can do is reframe the experience for myself. Maybe I can start thinking of the skill it takes to decide a thing isn’t findable rather than concentrating on the failure to find the thing.

What works for you, fellow searchers?

Microfilm Whimsy

Last year one of our microfilm reader/scanners bit the dust, and we know that our ancient analog reader/printer is one service call away from the junk yard. And oddly enough, our microfilm gets fairly regular use here, so being down to one machine was going to be a pinch. So now, as of last week, we have a new machine.

Our dear old T-Rex

Gemini wasn’t exactly this, but it was a lot like this. (Image from here.

This means that we had to come up with a name for our machine, of course. We’ve been naming our machines for at least the last decade when a former colleague said that our first reader scanner should be called Gemini, because it looked like twins.

Then I said our ancient analog machine should be called T-Rex, because it’s ancient but powerful and a fan favorite.


My first year here, we got a fancy new machine¬†that had a silhouette that reminded me of Yoda meditating on a rock, so we called it Yoda. This machine was also mostly a mystery to all who tried to use it, not least because it had no less than 7 on/off switches. Plus a light made to look like a mouse, but wasn’t a mouse, and was actually completely unnecessary unless you were looking at opaque microcards. But nobody could ever remember that. So being simultaneously powerful and utterly mysterious also fit with its name.

You can see Yoda sitting on a rock, right? Right??

Time passed. At some point Gemini was beyond support. And eventually we got a new machine.

Our next machine reminded one of our campus IT folks of the character Crazy Frog, and the name stuck.

And now, as of last week, we have this new machine which looks almost exactly like Crazy Frog, but it isn’t Crazy Frog, and it will be taking over for both deceased Yoda and (eventually) T-Rex. So we spent some quality time this week naming it, and we ended up merging ideas from another campus IT person and from a librarian in my department. The new name? Yandu-Bot. It’s a robot version of a Yandusaurus. It merges the old and the new. It’s way more automated and powerful than our other machines. And as an added bonus, because it has the letters Y and D in that order in its name, the desktop systems administrator doesn’t have to make any changes to the computer and related systems it’s hooked up to, which all have the letters YD for Yoda.

Result lists as a genre of writing

Idea – by me

I’ve been having a bit of an up-and-down teaching term this Fall — some classes going really well and some falling flat — but one thing that I’ve really enjoyed is that so many of the classes I’m working with are in new subject areas for me, or are taking different approaches than I’ve taken before. It’s felt like everything is new, an experiment, and if I look at it in that light I feel just a little better about the term as a whole.

One recent experiment consisted of actually speaking words in a lower level class that I’ve been mulling over off and on for years but have only ever uttered once in an advanced seminar several years ago. And I think it was ok. I think it was worth the few minutes of time it took, and I’ll think about where I can work it into other teaching I do.

The words? “Result lists are a genre like any other genre of reading. They may look different from tool to tool, but they all conform to certain conventions, and you can read a result list like a type of document, applying genre-specific reading strategies just like you’d approach an article differently than a mystery novel.”

The class I was teaching was of a type that I generally really enjoy: teaching students how to read instrumentally in order to do better research. And we talk about different kinds of reading: skimming, deep reading, and reading instrumentally. In lower level classes this generally involves me passing out a short reading (usually a newspaper or magazine article) and having students work together to generate lists of topics, key terms, and names associated with those topics by reading the article carefully. Sometimes I have them do this with the aid of a worksheet and sometimes I just do it with them on the chalk board.

And this time I added reading result lists as a type of reading that has its own specific place in a research strategy. Result lists come in many forms, but they will all help reveal the range of questions authors seek to answer that involve the search terms you use, patterns in authors or publications that revolve around the terms you used, and clues about the vocabulary of your topic which you can then take note of and use to revise your searches. They are highly condensed, jargonized reference “entries” that teach you a lot about patterns of publication, about vocabulary, and about where you can go next with your searching.

In this particular class I didn’t elaborate on speech genres in general, or explain that they’re “relatively stable types of utterances” that operate within a particular context and reflect “specific conditions and goals” (Bahktin 60). I didn’t even indulge in a geeky digression into the ways that “secondary” speech genres “arise in more complex and comparatively highly developed and organized cultural communication” (Bahktin 61-62). Does this remind you of scholarly communication pathways and norms? Disciplinary discourse conventions? Yeah, me too. But in a 45 minute class with first year, first term students I thought maybe Bahktin was a bridge too far.

Even so, understanding result lists in this way has really helped me, over the years, to get away from the frustration of “failed searches” and become far more comfortable with the idea that spending time opening results here and there and quickly gathering vocabulary and a sense of publishing patterns is one of the quickest ways to arrive at useful results, even if it at first feels like taking detours through a swamp full of weeds. I hope it will help those students, too.

Bakhtin, Mikhail. 1987. “The Problem of Speech Genres.” Speech Genres and Other Late Essays. Edited by Caryl Emerson and Michael Holquist. Translated by Vern W. McGee. Slavic Series. Austin: University of Texas Press. pp. 60-102.

A librarian plays with javascript for the first time

A few times this summer my work with libguides left me turning to people for help with what felt like basic stuff, and each time the answer was “Oh, here are 2 lines of javascript that’ll solve your problem.” So I got jealous. I want to be able to know when two lines of javascript will solve all my problems, and what two lines those might be. Plus, I hate bothering busy people for tiny stuff.

Meanwhile, the other librarians and I were trying to figure out a clear and consistent way to link to catalog records from our Book assets in Libguides. Sure, there’s a URL field there, but over and over we’ve seen people think that if the title is a link, that means that the full text is behind the link. So they open the catalog, see what is clearly information about a book, and then proceed to use the catalog’s search bar to “search inside the book.” Then they’re confused when they just get results for more books. So they open their book again, and try it again. Endless frustrating loop. Also, we actually do have quite a few ebooks, so it’s extra confusing when some of those linked book titles go to ebooks and some go to catalog records. Not a great usability scenario.

So I thought “Well, we can just put a little HTML link in the book’s description box that makes it clear that this link, unlike the title link, leads to a catalog record. But 1) writing a bunch of HTML by hand can lead to errors, and worse, 2) not everyone is comfortable writing HTML by hand. But we did really want a second URL field with clarifying language, reserving linked titles for ebooks.

This, then, was the perfect project for me to learn Javascript. What could possibly go wrong? Here are my two methods of teaching myself from scratch:

  1. W3schools¬†(especially fiddling with their “try it” setups to see if I can figure out how and why things work)
  2. Googling everything I want to do, one thing at a time, and hoping someone else on Stackoverflow has asked a similar question, and that whatever answers they got there will work if copied and pasted into my code.

I started thinking I could get away with making a bookmarklet, but since it turns out that browsers generally don’t allow you to access your clipboard except in very limited circumstances, and that many websites (including our catalog and libguides, the only two sites I cared about) don’t let me inject HTML to build a popup form to get around the lack of access to the clipboard, I finally had to admit that a bookmarklet would not work for me.

So then, according to the developers who helped and encouraged me along over on Mokum, it was time to start thinking about coding a browser extension. It took me a couple of days to admit this to myself because it sounded scary and complicated, but eventually I took this tutorial and its sample files by the horns and dived in. Along the way I got valuable hints from a friend to pay attention to the order of operations (at this point that means I just try things in various orders, but some day soon I hope to actually know what I’m doing), and to use the built in “Inspect” console in Chrome to debug the Javascript (which until now I’d only thought about as being a tool for HTML and CSS stuff).

And today, nine days later, I published my extension to Chrome and to Firefox! Now adding links to catalog records is quick and consistent for everyone in my department. (And if you want the unpacked files to mess around with, here you go, though I must warn you that I am NOT an expert so there may be stuff in here to make an expert weep, or laugh. Any and all suggestions for improvement are welcome.)

At this point I’m only just barely good enough to try lots of things in hopes that one of the things will work, and then to move on to the next thing. And I’m still very hampered by only having the vaguest idea what’s possible and (almost more important) what’s impossible. But who knows, maybe sometime soon I’ll have another tedious workflow that I can speed up by getting my computer to do a bunch of the tedious part for me. The world is my oyster!