Let Them Eat... Everything: Embracing a Patron-Drive Future by Rick Anderson, University of Utah

1. Let Them Eat... Everything: Embracing a Patron-Driven Future Rick Anderson Associate Director Scholarly Resources & Collections
2. J. Willard Marriott Library Toward Greater Sanity in Scholarly Communication Less sane  Interlibrary loan  Big Deals  Subscriptions  Approval plans  Reference/Bib instruction  Redundant cataloging  Print runs More sane  Article purchases (document delivery)  Wikipedia  Shared cataloging  Ease of use  PDA (for books)  Print on demand
3. J. Willard Marriott Library Up through the 19th century, a library was... “... a building, room, or set of rooms, containing a collection of books for the use of the public or of some particular portion of it, or of the members of some society or the like; a public institution or establishment, charged with the care of a collection of books, and the duty of rendering the books accessible to those who require to use them.” (OED) “... a place in which literary, musical, artistic, or reference materials (as books, manuscripts, recordings, or films) are kept for use but not for sale.” (Merriam-Webster)
4. J. Willard Marriott Library (Up through) 19th -century model
5. J. Willard Marriott Library A more recent definition: “... a collection of sources, resources, and services, and the structure in which it is housed; it is organized for use and maintained by a public body, an institution, or a private individual.” (Wikipedia)
6. J. Willard Marriott Library 20th -century model
7. J. Willard Marriott Library In other words, when we say library we’ve usually meant: a structure, filled with a collection.
8. J. Willard Marriott Library Then comes the internet.  Buildings start mattering much less  Collections are diffuse and hard to define  Access is available from anywhere in the world  Access can be purchased virtually immediately  Collection size is potentially limitless  Economies of scale make vast purchases affordable ... and therefore  Libraries can cast a huge net rather than carefully craft artisanal collections; or  Libraries can put off acquisition until need is demonstrated
9. J. Willard Marriott Library 21st -century model
10. J. Willard Marriott Library Today, when we say library we increasingly mean: a “structure,” “filled” with a “collection.”
11. J. Willard Marriott Library Game-changers in the next five years:  Continued budget declines (or, at best, flattening)  Google Books  Radical discoverability  Radical availability  Hathi Trust  Robust, trustworthy archiving with effective metadata ( = even better discoverability)  8 million books in 2010; 14 million by 2012  Patron-driven options  Ebook/article PDA  Print book POD (Espresso Book Machine)
12. J. Willard Marriott Library Espresso Book Machine: The UU Experience  Basic design  Two printers  Saw  Glue pot  Complications  Dry climate  Ink systems  Still waiting for color printer  What works:  Physical processes  What doesn’t:  Discoverability of content (metadata)  A certain creakiness (glue pot, etc.)
13. J. Willard Marriott Library Espresso Book Machine: The UU Experience  Surprises:  Demand for self-publishing  Demand for blank books (!)  Opportunities for commercial publishing  Plans for the future:  U of Utah Press backlist  Unique digital collections  Hopefully, more ODB content with better discoverability  ODB’s plans:  Advanced search by end of 2010  Adding new in-copyright content (with metadata) at a rate of roughly 10k ti/month  Currently working on deals with Internet Archive (1.8m titles) and Bibliolife (1.5m)
14. J. Willard Marriott Library To summarize: this is our new context:  Documents tend strongly to be available online  Documents are radically more discoverable than they have ever been (even if only available in print)  Print-on-demand (whether outsourced or insourced) is an increasingly available option  Our budgets have been/are being dramatically cut  Waste is decreasingly acceptable to stakeholders  When we try to guess what patrons will want, we’re wrong nearly half the time
15. J. Willard Marriott Library Why we are still building collections anyway:  Not everything needed is available (or even discoverable) online  Some essential documents require physical curation  Watch for growing bifurcation: library as archive (special collections) vs. library as information resource (general collections)  Not everything can be shown to patrons before purchase and then purchased immediately upon demand  Budget management: easiest way to control spending is to keep control inside the library
16. J. Willard Marriott Library The Unattainable Ideal (or North Star Approach)  Every book ever published is easily and immediately findable  Any book ever published can be purchased by library for patron immediately upon realization of need (purchase or borrow)  Every article...  Every data set... This ideal does not have to be attainable in order to be useful.
17. J. Willard Marriott Library What can we do in the meantime?  Share. (Ugh.)  Books: expose everything we can and buy when the patron points  Ebooks (MyiLibrary, NetLibrary, EBL, Ebrary, etc.)  Print books (LightningSource, OUP, etc.)  Print books (Espresso Book Machine)  Journals: by-the-drink purchasing  Remember: patrons don’t need journals; they need articles  This is the opposite of the Big Deal: it’s the Tiny Deal  Problem: publishers don’t want to sell that way
18. J. Willard Marriott Library Contact: Rick Anderson University of Utah rick.anderson@utah.edu

Editor's Notes

This year’s program is absolutely chock-full of excellent programming on the topic of PDA. But since I’m going first and since this is a plenary session, I have the luxury of talking in bold philosophical generalities while letting the folks in more focused sessions deal more with nitty-gritty details.
(click to next slide)
So, in that spirit: as I get older, I’m finding that I tend more and more to like things that are fundamentally sane, and to dislike things that are fundamentally insane. And as time goes on, more and more of the things we do in libraries seem to me to fall into the latter category. Not all of them, by any means, but more of them.
Now, I want to be clear about this: I’m not saying that it has always been insane to do them. In very many cases, we had no choice but to do them in the past because of the constraints that our information environment placed on us. But in a good number of those cases, I think we’ve started to suffer from a kind of Stockholm Syndrome – we’ve come to confuse onerous necessity with professional virtue. Let me give you some examples of what I mean. (continue with notes on next page)
Less sane
ILL involves shuttling documents around between libs. Hugely wasteful and inconvenient for patrons, and the more it’s needed the more it reflects our failure to build the right collections. My ILL department is fantastic, but they are fantastic at doing something we really should be figuring out how to AVOID. This is a standard problem in our profession: we get used to doing fundamentally insane things when we have no other options, we get really good at them, we lose sight of the fact that those things are necessary evils, and we start thinking of them as core services.
Big Deals, subs, and approval plans are all ways of buying stuff you don’t need in order to get what you do. (A journal sub is just a small Big Deal.)
Ref/BI are great services, but they’re not scalable if your library is serving a large population. The only way these services can function is if they’re not actually reaching the vast majority of your constituents.
Cataloging – how many good records do we need? (One.) How good does it need to be? (Pretty good.) Although we’ve made significant strides in the direction of shared cataloging, there are still way too many librarians creating way too many records for each book.
Print run is not just unsustainable – though it is that – but more importantly, it makes no sense.
* Note: the problem here is not that these practices are “old”; it’s that they don’t make sense.
More sane
Since buying journals means buying what we don’t want in order to get what we do, buying articles makes much more sense. But remember her that “sane” doesn’t necessarily mean “affordable.” I’m talking about models here that make sense at a fundamental level; what make article purchasing sustainable or not are the pricing models.
Wikipedia – Why would I put this here? Because Wikipedia is not just a product; it’s a model of information creation and distribution, and it’s a fundamentally sane one. Criticisms about the “hive mind” are not entirely relevant, because they go only to the content of Wikipedia entries. What students tell us they use W for constantly is not so much for authoritative information, but for links to authoritative information – the more you talk to users, the more you hear the phrase “jumping-off point.” Wikipedia is becoming the preferred alternative to A&I databases – in part because, unlike full-text articles databases Wikipedia returns manageable and reliably relevant results.
Ease of use, as an alternative to BI, seems obvious, doesn’t it? And yet how much time do we spend arguing with colleagues who seem to think that if it’s easy for patrons to find the sources they need, we’re somehow failing to provide them with a real educational experience?
PDA and POD: the fundamental sanity of these twin approaches is obvious, because in the first case we’re talking about buying only what’s demonstrably wanted, and in the second we’re talking about printing only what’s demonstrably wanted. Up until very recently, though, the fundamental sanity of these approaches hasn’t made them actually available. PDA and POD are where I’m going to focus my attention today, but I want to start quickly reviewing what’s happened over the past hundred years or so to get us to a place where it actually makes sense for us to start moving away from the idea of traditional collection building in favor of just-in-time resource provision.
Here are a couple of very traditional and noncontroversial definitions from a couple of venerable and respected sources.
Notice the fundamental elements cited by both of these definitions (the library is a place with a collection).
Here’s my attempt at a visual representation of the reality that made these definitions accurate:
This is a crude visual representation of the scholarly information environment up through the 19th century. The oval represents the universe of available documents, and the colored dots are the documents themselves. The ibrary is relatively small, and its walls are relatively solid (meaning that the documents it acquired were only available from inside the library itself). It contains only those documents that librarians go out and grab and bring back. Because documents are relatively expensive and the process of going out and getting them is also expensive, they are carefully chosen for match to mission. (Notice that this reddish library is grabbing only reddish documents.)
Now here’s a more recent definition of the word “library.” Notice how the language has changed. The OED said that the library is a building full of books; Merriam Webster said that it’s a place that contains documents in various formats. Wikipedia says that the library is first of all a collection of sources and services, and also a physical place.
You can see the definition of the library getting fuzzier from the OED to Merriam-Webster to Wikipedia.
Here’s an image to represent this evolved reality:
Here we see an explosion of available documents, because in the 20th century there was significant growth in scholarship, research, educational funding, and library resources; thus there are more documents to choose from and libraries had more money with which to buy them and more librarians who could go out and get them. During the 20th century, the unit price of a scholarly document dropped dramatically.
Consequently the library is quite a bit bigger, but its walls are also becoming less solid. Although the documents are still almost all in physical formats they are no longer completely contained in the box; they are starting to become available to patrons who are at some distance from the library (via telephone, or home delivery, or ILL).
But the library’s walls are still fairly solid, its collection is fairly well-defined, and you could argue that an individual library is at least partly defined by what it doesn’t contain (either you have an architecture collection or you don’t)
But all the way through the 20th century, these fundamental definitional elements apply: we generally thought of the library as a physical place that contains a collection. It’s actually what the word itself means – in French, Spanish and Latin the word “library” is cognate with the word for “bookshop.” To call something without books a “library” would be a bit like calling something without wheels a bicycle.
And it’s important to bear in mind that we became really good at this kind of librarianship. Over the course of the 20th century we figured out how to construct good library buildings, how to move books and journals around fairly efficiently, and how to build excellent collections. In fact, collection development became a whole subdiscipline of librarianship. We figured out how to craft approval plans that would allow us to stop using up our time making the most obvious selection decisions so that we could spend more of our time applying our hard-won expertise in the pursuit of more obscure or hard-to-find or ambiguous materials. We learned how to talk to faculty and analyze syllabi and watch for new programs so that we could anticipate the needs of our patrons. Anticipating their needs was essential, because if we failed to guess ahead of time what they were going to to want, they would come to the library and be disappointed.
And then, of course, in the early 1990s everything changed.
Now, it’s worth remembering that the internet didn’t emerge in the 1990s – it had been around since 1969, and by the late 1980s there were lots of people (including many of us in this room) using it for email and discussion lists. But the internet didn’t hit the world of scholarly information in a really radical way until graphical interfaces started being developed for it in the early 1990s. That was when it first became practical to start publishing scholarly content online rather than in print. And obviously, twenty years later we’re still absorbing the shockwaves of that development.
At first, I don’t think we understood how radical this change was going to be – we figured that an online journal was pretty much just like a print journal, only in a different format. What many of us didn’t anticipate was that the Web would become the default location for virtually all information seeking and for an increasingly large majority of publishing.
One thing the internet made possible was the acquisition of enormous just-in-case collections of content. So we didn’t abandon the idea of collections – we only moved away from the idea of carefully crafting them. But now that ebooks can be acquired at point of need, journal content can be purchased by the article, physical books can be printed on demand, and budgets are dramatically tighter than they used to be, it’s much harder to justify that approach.
Now that the internet has radically changed the world of library collections, what does our environment look like?
Now the library’s walls are really fuzzy. Technically they’re still there, but from the patron’s point of view, it’s very often unclear whether a document is part of the collection or is just plain available. In fact, I think our patrons are less and less aware of the library collection at all; they divide the world of information into what’s available and what’s not, and while they have a vague sense that some things are available to them because of their institutional affiliation, they don’t think of available online resources as “library resources.”
Notice also that the library is huge; its acquisition patterns are decreasingly selective and more inclusive, partly on purpose but largely because of the growing availability of bulk-purchasing models that make the unit cost of a document extremely low. Documents that might never have been acquired on purpose in the past may now be acquired almost by accident. This isn’t only true of large research libraries, either – because of bulk-purchasing options and the emergence of regional consortia, even small and mid-size libraries tend to have much larger collections today than they would have dreamed possible in the 1970s and 1980s.
Notice that there’s still a “core collection” there in the lower-left corner of the library. And there are still librarians reaching out for relevant documents that fall outside the collection.
But what all that fuzziness means is that our definition of “library” has gotten decidedly postmodern; almost ironic. (Illustrate with scare-quote fingers.)
What it also means is that the ground is very soft right now; I don’t think we can safely assume that the fundamentals of the future scholarly information economy are settled.
Many factors are going to contribute to the continued unsettling of that foundation.
The impact of ongoing budget cuts (whether in allocated dollars or, in the case of flat budgets, as a function of price increases) is obvious and doesn’t need much discussion.
Google Books changes everything. Absolutely everything. Libraries have never offered radical discoverability or radical availability; we’ve always offered moderate discoverability of a highly selective collection to a restricted population of users. Now, I know what you’re thinking, because I’m thinking it too: Google Books won’t offer radical discoverability until its metadata improves, and we won’t know whether it offers radical availability until the terms of the settlement are finalized and the settlement is approved. To which I would say: “Yeah, okay, that’s sort of true, but those are minor details: what matters is that the system is in place.”
Now, how can I say that lousy metadata and legal inaccessibility are minor details when we’re talking about, for crying out loud, discoverability and access?
Think about it this way: suppose you have in your parking lot a huge, gleaming, beautifully maintained eighteen-wheeler with a full tank of gas. Now suppose that its tires have all been stolen. Is the fact that it has no tires a minor detail? Well, yes and no. You sure can’t drive the truck until the tires are replaced, and replacing them is going to cost a lot and involve a lot of work (outfitting an 18-wheeler with a new set of tires is going to set you back roughly $5,000, according to a truck mechanic I talked to). But in a very real sense it is a minor detail: we know how to buy tires and we know how to put them on. It just takes money and effort. Once done, that truck is ready to roll.
Hathi Trust offers something that is simultaneously not quite as good as and almost infinitely better than Google Books. Not quite as good because not quite as comprehensive; infinitely better because the content is of more concentrated quality and the metadata is good enough to realize the potential of radical discoverability. Together, Google Books and Hathi Trust make a collection that effectively knocks down the walls of every research library in the country; they undermine the claims to unique wonderfulness of elite research libraries while increasing by an almost quantum factor the functional collection size of thousands of smaller libraries. With these two developments, suddenly your local collection matters much, much less than it used to. Hathi currently includes 8 milion volumes; they’re projecting 14 million by the end of 2012 (which, by the way, is not very far off at all).
Patron-driven selection erases what has been a core function of librarians: that of guessing ahead of time what patrons are going to want. Print-on demand, whether outsourced (think of LightningSource) or in-sourced (think of the EBM) erases a longstanding problem of the publishing industry: trying to guess how many copies of a book are going to sell.
The Marriott Library has had an EBM onsite for about a year now. As clunky and ugly as it looks, it is actually unbelievably sexy. What makes it so is the fact that it creates a book essentially out of thin air, virtually instantly, in response to a clearly demonstrated need. Being able to do so calls radically into question virtually everything about traditional librarianship – from the process of selecting (why pay librarians for their ability to predict future patron needs when we can meet patron needs in real time?) to the necessity of the collection itself (why buy books on spec at all when they can be acquired at point of need?)
Obviously, the reality is far more complicated than that – not “everything” is available; the metadata for many of the books that are available for printing is so awful that they’re virtually undiscoverable; the machine itself is still primitive and clunky. But these are minor details. We know how to buy tires for this truck, and we know how to put them on.
The EBM has already taught us several very valuable lessons, one of which is that we are not very good at anticipating what patrons are going to find compelling. Early on, we used the EBM to print a bunch of 80-page blank books, with images from our digital collections on the covers. We used these blank journals as incentives for patrons to take a library survey. We were immediately inundated with requests to purchase these blank books. Now we sell them at the Reserve desk, where they are very popular. We would never have guessed that empty books would turn out to be such a draw.
We’ve also discovered a significant self-publishing market in our area. Utah has a population that is unusually interested in personal and family histories, and we find ourselves printing and binding these (for a fee) on a regular basis. We are also printing, binding and shipping a two-volume conference proceedings for a scientific society.
Our campus has recently moved to an electronic thesis/dissertation system, and we’re using the EBM to print up archival copies of the ETDs. (“Shelf copies” will be .pdf versions stored in USpace, our institutional repository.) And we’re moving carefully in the direction of making the University of Utah’s Press’s backfile available on a POD basis via the EBM.
Each of these realities militates against the logic of building collections based on speculation. All of them together make a powerful argument against traditional collection development practices.
Of course, we’re not yet at the point where we can just cancel our approval plans, fire our firm-order vendors, cancel all our subscriptions, and turn completely from a library-focused collection-building operation to a patron-focused informatin brokerage operation.
When members of my staff get nervous about setting ambitious goals for themselves, worrying that they’ll be penalized for not reaching them, I tell them that I’d much rather see them set a goal of 100 and reach 80 than set a goal of 50 and “succeed.” I think this kind of thinking is essential if we’re going to remain mission-critical to our sponsoring institutions (whether the institution in question is a polity or a university or a corporation).
I think of it this way: I don’t aim at the North Star because I think I’m going to get to the North Star. I do it because if I keep myself aimed at the North Star it will keep me going in the right direction. Traditionally, the research library’s North Star has been the Great Collection. I think this needs to change. I think our new North Star should be “every patron gets everything s/he needs, with zero effort, in the moment s/he needs it, in the format s/he prefers.” Attainable? Of course not. But this goal will keep us pointed in the right direction. There is a great danger that we’ll forget that the collection is a means, not an end.
That said, I do think it’s important that at least some libraries continue building what I call Monument to Western Civilization collections; someone needs to be trying to preserve, in a permant, robust, and reasonably comprehensive way, the intellectual heritage of our culture. But just because you’re a library doesn’t mean that’s your role – obviously, very few of us are funded to do that, or even explicitly charged with doing it. To the degree that the realistic option exists of buying only what’s needed, at the point that it’s needed, most libraries would be well advised to shift in that direction and away from the traditional model of collection building, which is built on librarian speculation about future needs.
Obviously, such options are neither universally nor comprehensively available. Yet.
I hate sharing. Not as a matter of principle, of course, but I hate it as a solution to the problem of access – at least when it comes to print documents. Shuttling books and articles around between libraries is sometimes necessary, but it is not a fundamentally sane solution to the access problem.
However, there are sharing options emerging that actually are fundamentally sane: HathiTrust is one; what’s “shared” isn’t so much access, since access is public. What’s shared is the cost of archiving the content robustly and providing access to it reliably.
In fact, this is true of online resources generally: if two campuses “share” access to a database or journal, the sharing is really happening on the cost side, not on the service side – for patrons, the reality is simply that everyone on both campuses has the same access.
On the book side, we need to move as quickly as we can from the print to the online environment; not because ebooks are always preferable to printed books, but because the upsides they offer (24-hour access, remote access, full-text searchability, etc.) so dramatically outweigh the downsides. And once you’re in the ebook realm, buying on spec becomes absurd. Many PDA options are available from a variety of publishers and aggregators. Even in the print realm, PDA options are proliferating quickly.
The journal subscription is a fundamentally insane method for acquiring journal content; it’s just a small version of the Big Deal, whereby you buy tons of stuff you don’t need in order to get what you do need at a decent price. We have no choice but to move in the direction of per-article purchasing. Unfortunately—and I mean that sincerely—an article-based information economy will not support as many publishers as a journal-based economy did, for the same reason that a song-based music economy can’t support as many record labels as an album-based one did: you can’t make as much money selling the songs people want for 99 cents apiece as you can selling them the songs they want alongside songs they don’t want at a bundled price of $15.99. Some journal publishers are going to go out of business as the marketplace becaomes more rational. I wish it weren’t so, but it is.

Let Them Eat... Everything: Embracing a Patron-Drive Future by Rick Anderson, University of Utah

You are reading a preview.

Check these out next

Let Them Eat... Everything: Embracing a Patron-Drive Future by Rick Anderson, University of Utah

More Related Content

Featured (20)

Let Them Eat... Everything: Embracing a Patron-Drive Future by Rick Anderson, University of Utah

Editor's Notes

Let Them Eat... Everything: Embracing a Patron-Drive Future by Rick Anderson, University of Utah

You are reading a preview.

Let Them Eat... Everything: Embracing a Patron-Drive Future by Rick Anderson, University of Utah

Recommended

More Related Content

Featured (20)

Let Them Eat... Everything: Embracing a Patron-Drive Future by Rick Anderson, University of Utah

Editor's Notes