most versus all

We’re changing our borrowing policies to coincide with the start of spring semester next week, and it’s been a long, arduous (torturous), involved process that taught me a lot about the strengths and weaknesses of all of the library’s public services staff, myself included.  A very good learning experience, even if it does drive me to eating Snickers bars and listening to Metallica as coping tactics.  We’ve postponed, restarted, tested more, and rewritten everything ad nauseam, largely so that when users start interacting with the system we know to some degree of certainty that it’s going to work as intended.  We fixed most of the problems in advance, if not all.  It seemed like the right thing to do.  Good service matters, and a well-tested system is good service.

Contrast that with the rollout of the new Proquest interface.  We had a Reference meeting this morning and ran through the challenges we currently have with Proquest:  RILM and MLA are not migrating until mid-semester.  the holdings statements exported to openURL resolvers in the new databases are randomly wrong, the ones that have not migrated yet have lost their openURL links entirely, you cannot browse issues of a periodical, and on.  It’s… not good.  It is, in point of fact, very bad at some of the places where users routinely interact with core functions of the databases.  Didn’t they test it?  Why didn’t they postpone?  Why are the customers the ones finding these flaws?  What the hell is wrong with our vendors that this is acceptable to them at product launch?

And then I think about the administrative nightmare of rewriting our borrowing policies, trying to get and keep a half dozen staff of differing levels of commitment and knowledge onto the same operational agenda, and I have a lot of sympathy for whatever Vice President is in charge of Proquest’s platform migration.  I want to hug this mythical soul and say “I understand.  You tried.  You really did.  It’s not all your fault.”

The thing is, though, that if our policy rollout explodes into fiery pieces of shrapnel, it is mostly my fault.  Not all.  But mostly.  Because as the administrator — the leader and manager — it’s my job to get everyone onto the same agenda, whether they want to go or not. And failure is part of the process of improvement, and sometimes, yeah, we’re gonna fail.  Shit happens.  With the part of my brain entirely separate from my Service Mentality, I have entirely accepted that sometimes you have to screw it up so you can nuke it and start over clean.  Start over better.  Maybe that’s going to happen to us next week.  Maybe that’s what happening with Proquest right now.  So I’m resigned:  If we go up in flames, it’ll be mostly my fault, and it will be all my responsibility to ensure that we fix it.

It’s a hard damn job.  I hope someone at Proquest is doing it, too.  I’d advise that he or she stock up on Snickers, and I recommend “Don’t Tread On Me” or “Master of Puppets”.


  1. We’re (I’m) in the middle of getting MPOW ready to migrate to the new ProQuest platform, and I must admit that I have been more focused on search than on browsing fulltext, so your comment that we can’t browse issues of a periodical freaked me out.

    You’re probably doing this differently, but I CAN browse the issues of a periodical, by starting at the new ProQuest homepage, clicking on “Publications” in the menubar at the top, and then drilling down.

    So there _must_ be some way to link directly to that same page of issues.


  2. It’s apparently different across different databases that are migrating. PsychARTICLES is a big problem for us right now; some titles show the browsing list, some titles do not, and the holdings don’t appear to be correct for ANY of them. Right now, I’m showing browseable issues of the Journal of Abnormal Psychology from 1906-present, but the data pushed out to the openURL resolver says we only have 2001 to the present. David had examples this morning of the reverse holdings problem, and of titles where the browse list was entirely gone. It’s just *buggy*.


  3. Ouch. As a “big school” we got early access, and were supposed to go live over the Christmas break. In November, it was clearly flaky enough, that I pushed our migration out to May. That way, if it’s still bad, then we’ve got the summer to deal with it.


  4. We discovered today that the custom links that are supposed to take you to one database (and either simple or advanced search)appear to not be working properly anymore. Now instead of going to the advanced search screen for Dissertations & Theses (as I set up with the URL), I land on the simple search screen with ALL of our Proquest databases selected. Currently waiting for PQ support to get back to me, but if we’re not the only ones experiencing problems, I don’t expect it anytime soon.

    I’m not sure what’s worse — platform changes in the middle of a term or platform changes at the beginning of the term that are buggy and unreliable.

    Somehow I doubt their leadership will take as much responsibility as you do.


  5. OK, so the last time our consortium migrated, they prgrammed library names as actual names – you know, Manchester or Allen Park or whatever. Logical.

    Only to find out, after implementation, that that’s what pops up on the screen (and prints on the receipt) when you put something in delivery. But the delivery guys don’t use names, they use codes and numbers (we’re MANC 217, for example). So taped to the side of all our circ desk monitors is a list of the 55 or so libraries in our consortium, name-then-code-then-library-number, so that we can look ’em up. Which we do. About 1,000 times a month.

    Sirsi needs to charge about $50K to fix it. I sort of think it’s worth it, but then again, I could spend $1,000 on a lot of other things for my library now. And most of the other libraries aren’t bothered by it.

    Migration is so fun, isn’t it?

    Stay away from the dark chocolate Snickers bars, by the way. They’ve ruined me for regular Snickers bars, which is good in that I can only find the dark chocolate ones in every other gas station, thus making it harder to be tempted. 🙂


  6. One of the things about being Canadian is that the latest innovations in chocolate technology aren’t necessarily available up here.

    So, I do not have to fear the temptation of dark chocolate Snickers, since we don’t have them, and I’m not driving to Buffalo just to get some.


  7. Sorry, Metallica hijack:
    There is no album titled “Don’t Tread on Me” It’s officially self-titled, but commonly referred to as “the black album”.

    Yes they stole the rattlesnake symbol from the Gadsden flag.
    More information @

    All software companies seem to be following the “release imperfect software trend” because they can/maybe/never “fix it later” while charging more for a flawed product now.
    I expect this trend to continue. How many times have you patched a Microsoft, Mozilla, or Adobe program? (If it hasn’t been recently, better get on it.)

    Since we all operate with/within flawed systems, do your best and try to make it better. Paying for software patches/ updates still sucks, but Metallica does make tough situations better.


  8. ooo, dark chocolate snickers. And, Dom, I meant the song, not the album. 🙂

    I actually got a private communication from a PQ staffer because of this, saying essentially “many of us wish it were better. We’re sorry, and we’re going to fix it.” So I have to think, as it usually is, that it’s a management problem.

    Damn managers. 🙂


  9. Anna, I reported the problem about generated URLs not working, and was just informed that

    “Issue #2 is a known bug that our developers are looking into. A single database url that is created from within PAM, if it has a database ID in it, that’s a bad url the system generated. It should have the database name in the url.”

    They also sent me a spreadsheet with the correct DB short names that do work for linking, so ask them for that if you don’t hear anything.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s