Categories
Library Issues MPOW OCLC Web Scale

Are we live?

Today is the scheduled Go Live date for the first implementation of the OCLC Web Scale Management services, and we are so close that we can almost touch it. We spent the day doing massive testing of the various pieces of WMS: Circulation, Acquisitions, Cataloging, and our WMS-driven Worldcat Local, with the OCLC team working with us to tackle anything that popped up. As you might expect, being the first to implement this radically new a system brings things to light that neither we nor OCLC entirely knew about or understood…the infrastructure for this is only now really being tested with live data from a working library. I know that a lot of eyes are on this to see how it works (or doesn’t work), so here’s a quick rundown of where we are today.

Circulation could, as far as the system is concerned, go live now…it’s a functional system at this point, with all of our policies in place and working for all of our patron types. However, we now have a backlog of circ data to catch up on (the delta between the last massive data load and now), and we’re proceeding with it, and with further data-verification testing at the same time.

Acquisitions/Cataloging is so, so close. We successfully received and cataloged books today from scratch for the very first time, and walked them all the way through the process of preparing them to circulate. As a part of this process we discovered that we are going to have to re-evaluate our workflows even more…even we, who have been preparing for this for 6 weeks now, didn’t really grok the degree to which this alters our traditional workflow. So over the next few days, we’re going to be taking a closer look at those, and see how we take advantage of the massively streamlined workflows that are possible with WMS. Acquisitions is still very much a module in progress…it’s much ahead of schedule as far as the initial install schedule goes, and I’m impressed with how far it’s gotten in almost a ridiculously short timeframe.

The last piece of the puzzle is Worldcat Local, which is about 95% of the way there for our needs. The one major display issue cropped up last week as we started to see our serials holdings in Local, and that display issue is already scheduled as an out-of-cycle fix. We’ve seen the fixed display, and it works really well.

I can’t really explain how hard many people have worked to get us to this point. Andrea Schurr, our Web Technologies and ILS Librarian has worked multiple 80 hour weeks to get all our data out of our current system and into WMS. The entire staff at Lupton Library have laid other needs aside to help out with every little piece of this project, completing things in weeks that were timelined for months of work yet. And the team at OCLC has just kicked ass responding to questions, pushing for updates, and generally being amazing in helping us understand how this beautiful beast is constructed so that we can get the most out of it.

So with all that said, it comes down to: did we make our implementation date? Are we “live” with WMS? At this point, for my purposes, we are. We have about 98% of our data in the system, and the few straggling records are ones that require some serious cleanup…which we’re doing, and those will be done in the next couple of weeks. We are are working towards catching up our circulation activity of the last 2-3 weeks, which will take a few days. As well, we are going to take a few more days to examine our acquisitions and cataloging practices in order to make sure that we’re using the system and our staff as efficiently as possible, and then begin the process of catching those data bits up…if you aren’t processing as you go, it’s amazing how quickly things fall out of sync. And our patrons are already using Worldcat Local, driven by WMS data, for discovery and resource location.

Are we as live as I’d hoped? Honestly, no, we’re not. But as my Dean keeps reminding me: it’s not about us, it’s about the patrons. We have to get just a bit more testing, more data verification, and more workflow examination on our end in order to make sure that what we’re delivering to our patrons is the best possible experience we can for resource discovery at UTC.

Here’s my takeaway from this whole process up to this point: we now have the first working system of this kind in any library in the world. We moved from our previous ILS to this completely new model and set of data structures in just 6 weeks, or roughly 1/6th the time of the average ILS migration. We are working very hard, still, to make this as good an experience as we can for our students, faculty, and staff. Who can complain if we take a couple of extra days to triple check a few things? 🙂

Categories
MPOW OCLC Web Scale

Web Scale launch delay

In our original timeline, we were supposed to have a soft-launch of our OCLC Web Scale Management system tomorrow, August 20, where we went live with Check in/Check out functionality. We are very, very, very close…but it’s unlikely that we will actually launch tomorrow. Here’s the short list of issues we are still working through:

  • Currently, the locations in our Worldcat Local instance aren’t accurate. This should, literally, be fixed tonight. But we’ll need some time to spot check to make sure they came through accurately.
  • Ditto with the Local Holdings on our Serials…they are starting to show up today, but we will need a bit of time to check some of them.
  • The loan rules for circulation went live today as well, but moving into circulating, even dual with our old catalog, without the chance to really run them through their paces makes my Access department want to strangle me.
  • There is one issue on the backend having to do with reporting that we have to have a solution for in order to satisfy our state auditors. Said solution is, we think, there…but not demonstrated. The state will hunt us down if we don’t demonstrate before going live.

Our data (bib, holding, patron, circulation) has been successfully pulled and imported into WMS. We’re missing a small handful of records that we’re still hunting down OCLC Numbers for…but we’re talking about 2000 or so records out of 600K+. We have pushed more data into the system than anyone else, and it is still responding very well…the performance is impressive, given that it’s operating over the web.

All in all…we are very, very close. So close that it pains me deeply to think about NOT going live on schedule. But schedules are flexible, and if we slip by a week, I’m not going to cry failure. So we implement a complete ILS transfer in 7 weeks instead of 6 weeks…no bigs.

TL;DR: We’re a little behind. But we’re almost there. We are still completely on target for a full live launch on Aug 30th, and we will almost certainly have the public catalog up and running well before that (technically it’s running now, but with some UI issues). I still expect us to start the limited-live before that as well for circulation…but it won’t be tomorrow.

Categories
MPOW OCLC Web Scale

Plowing ahead

I can’t believe that it’s been 2 weeks since we announced that we were going to be transitioning to the OCLC Web Scale Management ILS, and that I haven’t blogged in the meantime. Although to be clear, the second is a direct result of the first.

We have been working like mad to make this admittedly insane timeline work. I’m pretty sure that my ILS Manager/Queen of All Data Andrea worked 100 hours that first week, moving data around and massaging it into what OCLC wanted from us for Web Scale. As the first live site, we volunteered to take upon ourselves a huge amount of the data manipulation, so Andrea has been moving MARC out of Virtua, into Access to manipulate, and then finally through an XSD to provide OCLC with the final mapped XML for all the fiddly-bits of data. We’re also dumping huge amounts of MARC directly, but for non-bib records (holdings, items, patrons, etc) we’re doing a ton of conversion.

This isn’t to say this is the way that everyone will do it…but with our somewhat aggressive schedule, we were determined to give them whatever made it easiest to make WMS happen.

I’d like to publicly thank everyone at MPOW who have really put themselves out on a big limb to help with this implementation. We’ve got three major working groups doing different parts of this massive job, and all of them are digging in and getting stuff done. I’m so, so proud of the team that I work with at UTC…serious, I couldn’t have hand-picked a better group of librarians. They freaking rock.

We’re still on track for an August 20th launch, and we’ll be rolling WorldCat Local to our patrons before the full go-live….so we’ve got just short of 3 weeks to finish this thing off. Between now and then we’ve got to finish the little troublesome data sets, get an updated patron file ready, start a marketing blitz on campus for our patrons, get a redesigned index page up for the website that highlights WorldCat Local, and kick the hell out of some really, really shiny new tires.

Categories
MPOW OCLC Personal Web Scale

OCLC Web Scale Management

I am very pleased to finally be able to announce that the Library at the University of Tennessee at Chattanooga is scheduled to be the first live implementation of a product that has been talked about for years: a web-based, collaborative, modern library system that does away with silos of data. We are implementing the OCLC Web Scale Management library system even as I type, and will be going live with the system for circulation on August 20, 2010, and with circulation, cataloging, and acquisitions on August 30, 2010. A wiki page documenting the process, working groups, and more is available, and will continue to be updated as the process continues.

I could talk for a long time about how excited I am about the possibilities of this system…and probably will for the next few months at least. I’ve been pursuing Andrew Pace about this product for what feels like years now, and after seeing it and understanding what may come as a result of this…well, I can’t wait.

This is a major shift in the library world, and it’s one where I think the repercussions will take years to really be felt. The simple time-saving that will be immediately felt for libraries in their processes are enormous…the workflow necessary to get something from order to shelf is so straightforward and fast that I feel strongly that we’ll save several person-years of staff time in short order. In addition, there is a shared-plugin architecture for the staff-side that combines with the open API calls that give incredible access for mashups of data that directly interact with the system. One example that I’ve seen is a plugin that combines the New York Times bestseller API with the acquisitions module in Web Scale to allow for single-click ordering from a list of bestsellers that is a live call from the NYT.

Going the other direction, the architecture allows you to pull your own data out and impose it on other pages. An example of this would be a Firefox plugin that shows you realtime budget information while you shop on Amazon.com…complete with recalculation as you add things to your Cart.

This isn’t to say that there aren’t a ton of questions still. We’re the earliest of adopters on the first product of its kind…to say that I am a bit nervous would be an understatement. But the potential and promise of Web Scale is something that make it worth it.

We are literally implementing Web Scale Management in 30 days. To my knowledge, I’ve never heard of another ILS migration even approaching this level of speed…so if I’m a little out-of-touch for the next month, you won’t be terribly surprised.