Skip to content

Pegasus Librarian Posts

The User is Not Broken

Published on Upworthy

Every few years this picture makes the rounds on the internet, and every time it does the booksellers and librarians of the world nod in understanding. Public librarians and booksellers get this question day in and day out – academic librarians less so. (Most of our books in academic libraries don’t have their fancy covers on anymore, so this question would be pretty much impossible here anyway.)

Here’s the thing, though. It’s not that “people annoy us” with this question. It looks to me like there was some frustration behind this particular display, sure, and the result is wording that pokes fun at patrons in a way that makes me squirm, but that’s not the norm for people I’ve talked to. Most librarians and booksellers I’ve encountered are proud to know their collections and clients well enough to be able to match people with the books they want!

No, the thing that’s frustrating is that it’s hard for us when we can’t figure out the answer to the question (we’re very all-in on the identity of being able to answer questions), and our systems don’t let us search by cover design features. That mantra of librarianship, The User Is Not Broken, means that if we’re getting tons and tons of the same question, that means that the system isn’t set up right.

Of course, another piece of the frustration is that we often don’t have a ton of control over our search systems. Don’t get me started on the refrain from vendors about “what users want” that almost never matches what the people I work with every day ask me about… And, sure, often what people want and need aren’t technically feasible, or aren’t feasible within shrinking budgets, or aren’t feasible with reduced staff or with staff that gets less and less training.

But if we actually can’t change the systems, we can embrace the fact that we can foster humans who can say “Oh, maybe you’re thinking of this red book?” Humans are pretty incredible!

Leave a Comment

Farewell, dear LIShost

Well, let’s try this again. DNS confusion meant that last Friday, when I thought I was all set, I actually wasn’t. But apparently NOW I’m all set, so here’s the post that I thought would get posted last week. (Many thanks to Reclaim Host tech support for all their extremely helpful help.)


I’m writing now from my new home on a new server and a new domain registrar for the first time in my blog’s history. Up until now I’ve been happily living in my own little corner of the LIShost server, carefully tended by the wonderful Blake Carver. Sadly, LIShost is shutting down after 18 years of dedicated service to library folks. I can’t thank Blake enough for his approach to hosting, giving me as much leeway as I had the skills for, and as much help as I needed to cover those areas where my skills were lacking.

But welcome, now, to my new home, here in my own corner of the Reclaim Hosting servers. If things continue as they have begun, I think I’ll be quite happy here in my new digs. But if you notice anything wonky do let me know. I’m still unpacking and figuring out the light switches and hanging pictures, and there’s no guarantee I’ll notice everything.

Leave a Comment

Preparing for your database lists to malfunction on occasion

If you’re like us, you have a LibGuides-based list of databases for your library, and it went down for a bit this morning. If you’re like me, this pretty much cripples you until the list is back up and running. This kind of downtime could happen no matter your platform, so if you’re like me you might also want to have an option available to get around your database list during downtime. Here’s what I came up with about a year ago and was happy to have in place this morning.

Each month I download a CSV file of our databases – LibGuides has a nifty “export all” option and I click that. Then I paste that data dump information into a hidden sheet on a Google Sheet that I’ve made available to anyone with a link. From that messy data dump, I built formulae onto a visible sheet in that Sheet (really, Google has GOT to get better about its names for apps so that I don’t have refer to a sheet on a Sheet). These formulae help me display a full alphabetical list of all of our databases, their descriptions, their base URLs (a note on that in a second), vendor, and whether or not they need proxy access from off campus. Basically all the formulae say are “if the corresponding cell on that base sheet is blank, don’t put anything here, but if there’s information there, then display it here.” It looks like this:

=iF(isblank(Raw!B2),"",Raw!B2)

Then I use a slightly fancier formula to build a proxified version of that base URL into column dedicated to off-campus access. It goes like this:

=if(isblank(F2),"",if(F2="yes",CONCATENATE("http://ezproxy.carleton.edu/login?url=",C2),C2))

Translated, that formula means “If the cell in Column F that says whether this database needs a proxy string is blank, leave this cell blank (this just makes for a cleaner spreadsheet without a lot of error cells where the formula is there even if there’s no database listed). If that proxy cell is set to YES, then put together our proxy string and base URL, and put the resulting URL here. If the proxy cell is set NO, just put the base URL here.”

Then I hid the Proxy check column (column F) because nobody really needs to see that if they’re using the spreadsheet. I just needed it for calculation purposes. (Sure I could have referred to that proxy check cell on the base sheet, rather than bring it to the visible sheet and then hide it, but sometimes I just feel like doing things easiest way that occurs to me in the moment. Don’t judge!)

Finally, I gave this back-up spreadsheet a nicer URL: http://bit.ly/libedatabases And I posted this URL in places where librarians can find it when needed (such as our documentation for QuestionPoint cooperative librarians, our intranet, etc).

So now if our proxy server goes down (rendering our database list mostly useless), we can use the base URLs from this spreadsheet, at least from on campus or in combination with a VPN. It’s better than nothing. And if the whole database list goes down, we have access to all of our databases and their URLs from this list.


Edited to add: I forgot to mention that I’ve also built a script into Google Docs to unmerge cells. For whatever reason, exports from web-based products like Springshare tend to have random merged cells, which I don’t want. The only way I know of to get rid of these (other than looking for them all and then unmerging them individually) is via script.

So! In Google Sheets, click on the “Tools” menu and then “Script Editor” and then paste in the following:

function myFunction() {
   var breakRange = SpreadsheetApp.getActive().getRange('A:T');
 for(;;) {
   try {
     breakRange.breakApart();
     break;
   } catch(e) {
     breakRange = mySheet.getRange(
       breakRange.getRowIndex(),
       breakRange.getColumnIndex(),
       Math.min(
         breakRange.getHeight()+5,
         Sheet1.getMaxRows()-breakRange.getRowIndex()+1
       ),
       Math.min(
         breakRange.getWidth()+5,
         Sheet1.getMaxColumns()-breakRange.getColumnIndex()+1
       )
     );
   }
 }
 }

This will look at columns A through T (you can edit that in that second line if you need to) and unmerge any merged cells.

When you paste in a new export, just click “Tools” and “Script editor” and then the little “play” arrow to run the script, and it’ll unmerge all those merged cells for you.

Leave a Comment

2020 Census: the demo software, data, and more on the debate about differential privacy

Since last I wrote, I’ve been able to learn more about a question that had plagued me: what of IPUMS, that unparalleled resource for census micro-data? For one thing, I was sure they must be thinking about privacy already – micro-data must be handled with care. For another, the 2020 Census “Privacy Budget” work is likely to make IPUMS’s work pretty complicated or even impossible.

Turns out, they are of course way ahead of me. Here’s their page on the topic, which also links to the Census’ released software and documentation, as well as to data available from IPUMS that researchers can use to test that software.

I also happen to know someone who works at IPUMS, and he says they’ve already been doing record swapping and injecting “noise” into their data to protect unique cases.

I certainly hope that everyone involved can figure out how to protect people’s privacy while still allowing vital research. This seems like an impossibly tangled knot to me.

Leave a Comment