My Old LibTech Blog (2013-2016)

Cloud Solutionism

Author: John Durno
Date: 2014-06-19

If I had a dollar for every time I've heard someone say, "If we only had software x, then we'd be able to do y" ... well I've heard it a lot over the years, but I still wouldn't be able to afford all that software. And all too often, it's not the lack of software that's the problem, it's the lack of someone with the will, knowledge and/or time to do whatever it is that needs doing. I think most of us have figured out that Photoshop by itself won't produce beautiful artwork, but I'm not sure the equivalent is true for our administrative systems.

Paul Signac - The Pink Cloud, Antibes.jpg
"Paul Signac - The Pink Cloud, Antibes" by Paul Signac - Unknown. Licensed under Public domain via Wikimedia Commons.




I came across an example of this the other day, over at Carl Grant's blog. In the post I'm linking to, Mr. Grant talks up the value of cloud-based library services platforms in an interesting way, focussing on their ability to "aggregate usage data globally for all library members" and derive analytics to help libraries inform their business decisions.

Sounds reasonable. But then I looked at the five examples of the kinds of analytics Mr. Grant is suggesting we derive from all this data, and I'll be darned if I can find a single example where usage data aggregated from multiple institutions plays a role. Every example in his list talks about deriving analytics from the usage data associated with individual users and/or an individual institution. I won't list them all here, but they include things like tracking the success rates of students who use library resources against those who don't, or linking library use to successful grant applications.

And the funny thing is, we could do every one of those things now, without shelling out big coin for a library services platform. All we'd have to do is push all traffic to library resources through our proxy server and we could accumulate that data all on our own. We might need to beef up our proxy server a bit, but I guarantee that would cost way less than what we'd have to fork over for a cloud hosted LSP.

"But wait!" some of you might be saying. "You could gather the data, but you'd still need someone to run the analytics, right? Maybe the Cloud could do all of that for you."

Well no, it couldn't. The Cloud is not magic pixie dust. A cloud-hosted LSP would not actually produce any of the analytics listed by Mr. Grant, since they all require interfacing with the other administrative systems at your organization, which would require lots of local elbow grease. The best the LSP could do would be to accumulate the data and make it available for analysis. You'd still need someone onsite to do all the actual work.

If anything, a cloud-based LSP might make it harder to run these analytics, because you'd be at the mercy of your service provider as to what data they let you get access to. Locally hosted systems are usually much easier to extract data from, given that your local admins have access to things like server logs and direct access to backend databases, which are often not available to customers in a hosted context.

But regardless, just having the LSP software isn't going to do it for you. And if you have the capability of doing it at all, not having the software isn't going to stop you.