Participated in a panel at ALA last Saturday:
“Hiding in Plain Cite: The Growing Importance of Content Neutrality in Library Discovery Services”
Roger Schonfeld served as moderator. Joining me on the panel was Lisa O’Hara, University of Manitoba; Todd Carpenter, Executive Director of NISO; and Amira Aaron, Northeastern University.
One of the questions I took the lead on was, “What does Content Neutrality Mean to You?” Here’s my response:
I’m sure most have heard the phrase “net neutrality” — a network model that says bandwidth providers should treat all data that moves across their network in the same way. It is certainly true that many ISPs are just in the bit-moving business..providing network access..but a smaller percentage also provide content. It’s that vertical integration (meaning significant parts of the supply chain fall under the same owner) that gives rise to trouble.
For example, in a net-netural world:
- Comcast as an access provider shouldn’t shape traffic in such a way that Netflix video streams end up slower than content flowing from Comcast’s own Xfinity platform.
- Verizon shouldn’t count Amazon video streams against a user’s data allowance while exempting the same user from cap charges on a FIOS video stream.
“Content neutrality” is a similar idea. Our “access provider” in this instance is the discovery platform vendor. The analogs to traffic shaping or billing distortions occur instead around the metadata that’s being searched to “discover” relevant content. As with ISPs and net neutrality, there are some companies that just provide a discovery platform and others that are also in the content business. As before, vertical integration and perceptions of competitive advantage are problem incubators.
When you pay for a subscription to an e-content silo but metadata in that silo is not also represented in your discovery platform…well, you may have a content neutrality issue. I say “may” because there are some end-of-chain content providers who do not provide metadata to any discovery platform vendor (e.g., SciFinder Scholar).
How about an example?
Here’s one I saw just a few days ago. It highlights several dimensions of the issue and adds an ironic twist:
A student walks into our library, looking for a PDF of a dissertation that was completed by another doctoral student at George Mason University. She searches our collections via our discovery product and concludes it doesn’t exist. Well, yes, she did find a record to a print copy that’s shelved in our dissertations area, but she wanted a PDF version–to keep with her other e-research resources. She leaves the library dissatisfied.
We’re in an urban area and belong to a regional consortium. Later that evening our student continues her research at another university’s library. This time around she quickly finds that elusive dissertation and in just a few seconds downloads the linked PDF from Digital Dissertations.
Can’t you can just see the thought bubble over her head?
“…this library is so much better than Mason’s…”
What just happened?
Turns out we bought our discovery product from a vendor who doesn’t have access to the metadata from Digital Dissertations. Yes, we subscribe to that content—dissertations and metadata—but for us, it’s only available in a ProQuest silo. A standalone database. From the perspective of a user of our discovery platform–we’re wasting collections funds on that silo-bound subscription.
Across town they’re using a vertically-integrated discovery product. Summon™ (the discovery system) and Digital Dissertations are both ProQuest products. If I were to tell you that Digital Dissertations is available exclusively through the Summon™ discovery product would you be surprised?
Ironically, the PDF our student downloaded from Digital Dissertations was supplied by a Mason student when he sent his dissertation to ProQuest.
Two workarounds come to mind:
- Spend your time explaining to that small percentage of users you actually see at the reference desk that they need to enter the Digital Dissertations silo for that sort of information (which surely makes them wonder why we’re promoting our discovery service).
- In this particular case, as we are now doing, require ETDs which we can find in our discovery system thanks to an OAI harvest from our DSpace instance.
In this age of open access and digital repositories I have to wonder why we’re promoting ProQuest dissertation services to our new PhDs in the first place. To take it a step further: if you assume there’s a cost to SerialsSolutions for indexing and providing Summon™ discovery access to Digital Dissertations, could you argue that we are helping subsidize that free-to-Summon™-users discovery access with the money we pay for our silo-only subscription?
If you believe that withholding metadata from competing discovery vendors might ultimately lead to subscription cancellations, you might wonder why content vendors don’t see it in their advantage to make their siloed metadata available to all (e.g., if you can’t discover it, it doesn’t get used and if it isn’t used, it isn’t renewed).
I have a somewhat depressing theory and it suggests we won’t see much change in the near term:
Libraries are a niche market. There aren’t an infinite number of ways for a vendor in that market to turn a profit. You grab the opportunities you see. Discovery products are “hot” and represent a new income stream. Rather than fight the battle over software features and price, content providers recognize that they can leverage their propietary content to differentiate and advantage their discovery product in head-to-head comparisons with competitors. If we, the customers, don’t do something to change this reality (e.g., start demanding access to metadata in our licensing agreements), we shouldn’t expect a lot of change.