The Blog Libific

ICOLC posts statement on the economic crisis and its impact on libraries

ICOLC, the International Coalition of Library Consortia, has just released an interesting statement titled “Statement on the Global Economic Crisis and Its Impact on Consortial Licenses“, and endorsed by nearly 100 consortia (so far). It calls on content providers to be more flexible with their customers regarding how they collect payments and, more importantly, what they expect from libraries over the next several years.

I thought one interesting point was the need for publishers to offer opt-out options for libraries if they expect the libraries to pursue multi-year deals. It’s an interesting problem on both sides. Libraries want to lock in savings for products they value, but given the fact that it’s hard to know what will happen next—and that the recovery of library budgets will lag behind the general economy—they need the flexibility to get out of contracts if they need to take especially drastic actions. But, then, what’s the point of the savings in a long-term contract if the contract isn’t really that long-term? On the other hand, it may well be that libraries won’t subscribe at all, if they don’t have the opt-out options. It’s an interesting problem, and one of many.

Another interesting bit was the part stating “This is not a time for new products.“ Especially since we just launched one yesterday. Of course, now may not be the time for starting to develop a new product (though, on the other hand, it might be; economic downturns do offer great opportunities to start new things), but there’s not much one can do about products that were started many months ago, and are now coming to the market. It’d be a shame if librarians wandered through the aisles at library conferences, looking at brand new products that took 12 to 18 months to build and release, and complain about vendors devoting their resources to those new products, rather than something else. And, if a new product helps a library save money, then now is definitely the time to be releasing it.

When I first saw that ICOLC had published this statement, I was a bit surprised—it seemed somehow totally obvious that libraries are having a tough time right now, and I wondered how a statement from ICOLC would be useful. But after reading it, I found it was useful for crystallizing some of the challenges that libraries and their vendors will be facing in the next few years.

Check out SUMMON

Serials Solutions is launching a very cool new product today, which we call Summon. I think it’ll be a real game-changer in how we present online services to patrons. With an enormous (and, of course, growing) collection of full-text data, holdings information, citation information, local repositories and more, we’re able to dramatically improve the federated search experience: the latency involved in searching is removed, because everything is already indexed. Patrons don’t have to sit and wait while the federated search engine goes out and does the searching at each native database; results come back nearly instantly.

To see more about it, check out our page on it here: Summon unified discovery service.

And be sure to check it out at ALA Midwinter and the Ontario Library Association Super Conference, if you’re attending either.

OPAC FAIL, from my local PL

I was searching my local public library’s catalog last week, and apparently I managed to bring it to its knees:

Sorry about that.

More craziness with IPEDS data

I’ve been doing more work on my research project, and I continue to find more and more challenging data problems.

The latest is that, since I started working on this, the 2006 IPEDS data set has been released, and I felt it was important to use that updated data in my analysis. But as I switch old data with new, I’m finding a variety of incredible changes.

For example, Fielding Graduate Institute went from reporting 3 journals in 2004 to 9755 journals in 2006 - I’m going to guess the latter is closer to the truth, though I counted nearly 15,000 available electronically. Another change was the shift for Columbia University, from 65,117 in 2004 to 102,901 in 2006. Cerritos College went from 10,000 in 2004 to 0 in 2006. I didn’t change the Cerritos number in my data, since that is clearly a mistake, but how can I know which of these others are mistakes? Perhaps IPEDS put in “0” if the institution did not provide a number, but that seems like a poor way of managing data - there’s a big difference between “0” and “no data”. I’m pleased to report that Emporia State’s numbers came back from the stratosphere - from 1,660,173 journals in 2004 to a more believable 665 in 2006.

Another 2004 outlier didn’t change, though: Long Beach City College remained at a previously unbelievable level, dipping just a bit from 193,103 to 191,712 serial subscriptions.

Here are some of the biggest change:

Apropos of nothing: Conversion rates on spam

The BBC had an interesting article today about how much money spammers might actually make. The article is here, and describes how computer scientists at UC San Diego and UC Berkeley infiltrated spambots in the Storm spamming network. They sent out nearly half a billion spam messages in three separate campaigns. In the largest campaign, they sent out almost 350 million messages touting “male enhancement” pharmaceuticals. They created a fake website to sell these drugs, and measured the conversion rate, from sending an e-mail to someone trying to make a purchase online. (When someone tried to actually put in credit card information, the scientists returned a 404 error.) They found that after sending out 350 million e-mails over nearly a month, they got 28 individuals interested in purchasing the drugs. The response rate was less than 0.00001%, but that is, apparently, enough to still make a profit.

The pdf of the complete article is here, and makes for some interesting reading. The authors are very careful not to assign too much value to their research, because of its limited scope, but it is worth noting that this is the first work that has actually measured conversion rates on spam. The researchers also concluded that spammers are not making an enormous amount of money here, and it may be possible to limit spam by raising the cost of distributing the e-mails.