One of the concepts I wanted the students to grasp was “lock in,” and I frequently referred to social-networking platforms as, possibly, an environment in which positive network externalities (the service got more valuable as more people signed on) were extremely strong but where the lock in to the service might nonetheless be fairly weak. The reason for this, I argued, was that switching costs to other networks were low. Or, perhaps more to the point, that the costs of joining a new service and duplicating your network of friends was so low that people would just participate in multiple venues, migrating over time — moving from, say, Friendster to MySpace, to Facebook, with a few Ning networks or the like along the way. What, I wanted them to ask, was the glue that would really make any particular network sticky?

Obviously, I underestimated the relative durability of Facebook — which was brand new during the time I was teaching that course — along with the sheer ubiquity that it would attain. But Facebook has also become sticky for reasons beyond sheer size and the value to the user of large networks. As the network has become more complex over time, and come to consist not only of linked profiles of individual people but of fan pages, groups, institutional presences, and applications, it has become stickier. The development of applications – including games – has also contributed to stickiness, as has the use of Facebook’s authentication (originally branded as Facebook Connect) for customized access to other content sites and web applications.

Most importantly, Facebook users have long used the platform as an information sharing and information seeking environment — passing links, news, and comments to their friend networks along with social updates. As Google was inspired by academic practices of footnote-mining and journal cross-references in its construction of automated web search, Facebook is borrowing practices more akin to dropping a marked-up xerox under a colleagues door. Except each service is extending these metaphors to much broader communities than ever used those particular analogs. Lock in works here because Facebook has the audience, so it gets the sharing/linking — and because it has the sharing/linking, it gets the audience.

Farhad Manjoo’s Slate article, “Facebook’s Plan to Take Over the Web,” gets, in the course of discussing the (relatively) new “like button,” at the heart of what I was missing about Facebook. Far from the “walled garden” that social-networking sites once represented, Facebook has become a portal to the world that renders the whole Web “social” — users of Facebook have begun in some respect to cede control over their experience of the universe online to their “friends.” Our social networks – or at least one particular social network – now plays an enormous role in directing our attention around the Web, and for many of us plays perhaps an even more central role in determining what online content we encounter than do search engines.

But, on a (perhaps) lighter note there’s a brand new intruder into the library space. Starbucks has just made Internet access, via its in-store Wi-Fi, free. But there’s more. As a sweetener in the offering, some paid online content will be made available for free to Starbucks-connected Web users – most notably, this will include “pay wall” content from the Wall Street Journal. Hmmmm…. Isn’t collective pooling and mediation of the costs of acquiring and finding information a pretty fundamental definition of what libraries do? Granted, Starbucks will cover the costs in a very different way than any traditional library – either via advertising or through selling coffee and food to those drawn by the free online content. But much bigger intrusions into the traditional market space of libraries — e.g. novel delivery structures for ebooks — will also be funded in very different ways (and, the danger, accountable to very different masters).

(For what it is worth,the WSJ is obviously offering Starbucks something different than they offer us — paywall content on the WSJ web sites is presently only available through an individual subscription with login/password authentication. Proxied group access through, e.g., campus libraries has only been available through intermediaries that aggregate the Journal’s articles.)

The product, Google Editions, has been much-discussed for some time, but I believe this is the first solid announcement of a time frame for launch. According to Jessica Vascellaro and Jeffrey Trachtenberg of the Wall Street Journal:

Google says users will be able to buy digital copies of books they discover through its book-search service. It will also allow book retailers—even independent shops—to sell Google Editions on their own sites, taking the bulk of the revenue. Google is still deciding whether it will follow the model where publishers set the retail price or where Google sets retail prices.

While this announcement isn’t about books in the controversial library-scanning project, we may soon have a glance at parts of the distribution infrastructure and book formats that will also be used for books in the scanning program.

From the standpoint of an observer of eBook formats and models, and of their likely impact on library services and business models, the most interesting thing may be that Google seems to intend to make book content available “live” through the browser, rather than by means of some downloaded (presumably DRM’d) file. As a reader, “your” books will “exist” in the cloud, rather than in storage on a particular local device. I haven’t seen any indication of whether this will require a persistent live connection to the Internet for reading, or whether (through a Google Gears-like mechanism?) off-line reading will be possible. If either a conventional download or use of local data via something like Gears is available, I’m not sure how Google will meet demands for copy-protection/anti-piracy measures. And Gears would limit access to devices with conventional operating systems (Macs and PCs) whereas Google presumably wants a service accessible to any device running a browser. So my (completely uninformed) guess would have to be that we’re talking about a product that assumes a persistent connection, whether WiFi, wired, or cellular, to the Internet.

Certainly both much-expanded WiFi networks and the (rather new) ability of the cellular wireless network to effectively handle Web traffic – supporting the proliferation of browser-enabled mobile devices – have changed the landscape from the one confronted by the earlier library ebooks that reference librarians failed to “sell” to most of our patrons (who considered them a low-tier choice for their lack of printability or downloadability). But have we progressed so far that books requiring a reader to be “plugged in,” if that is what these will be, are viable with the mainstream?

Another important attribute of such a service will be that the contents of all these books, including those sold by publishers or authors selling “Google Editions” on their own sites, will be hosted by Google and (like older materials out of the scanning project) float among Google’s data. This will mean, presumably, that they will be not only crawlable for search tools but available to Google for all manner of further semantic processing and data mining. Ad-serving is, of course, an obvious application of this information. But a Google book database of this sort, combined with data on user access to the books, could hypothetically go far beyond even the capacity of an Amazon for building recommender services and other tools to direct users from one resource to another.

This is an expanded version of a post that appeared on the CWRU Law Library’s blog here.

EDIT: Sarah Peretz at ReadWriteWeb quotes Amanda Edmonds of Google as indicating that a cached version of books will be stored locally once they are first accessed. Presumably this means Gears, so I’m still not clear on how it will work for the full range of mobile devices.

Greg Lambert at 3 Geeks and a Law Blog took the idea and ran with it — expanding Sarah’s chart to show how information products beyond traditional publishing have also been gobbled by the giants, and creating a Google Docs version for others to add information.

One thing that leaps out from the charts is how many publishers that proliferated the late-19th and early-20th centuries (which, I like to point out to my students, was another great era of perceived information overload)were consolidated into our current big players, beginning in the 1980s. So the explosion of research tools and publications that really created “modern” book-based legal research in the beginning of the 20th-century left a lot of competitors (though obviously West, with its Digests and National Reporters, had a special role). It will be interesting to think about how that compares with innovation in the legal research and legal information space in the last few decades. Obviously there has been some (uh… huge disruptive change, one might interject), but a lot has been subsumed through acquisitions or internal development into a really very consolidated industry.

Of course, by taking the perspective of expanding the family tree out backwards, these kinds of charts show us only how much a Thomson or an Elsevier have acquired, not how much they haven’t. In the “traditional” law publishing space, that may not matter — we all know there are precious few independents now and precious little market share for them. But in the information marketplace defined more broadly there at least might be other stories to tell… Maybe?…

(this post also appeared on Just in Case: News and Notes from the Judge Ben C. Green Law Library.)

It is an interesting, although incomplete, argument. I particularly find value in the first part, urging a reconsideration of both print and, more broadly, library metaphors for online, inter-linked, legal information. And I would agree that law libraries as institutions have often been too captive, if perhaps inevitably so, to incremental improvements in traditional services and resources, such that we are probably doomed to miss the really big disruptive, transformational, changes in the ways information (including legal information) is constructed and consumed.

While borrowing Jason’s disclaimer to state that I, also, am not a legal philosopher, it is by his analysis of law and legal change that I’m unconvinced. He seems to be writing off the construction of legal research tools around silos of authority (statutory codes, case repositories, etc.) as a rusty and outdated shackle that we’ve retained only from our habitual over-reliance on library metaphors. Instead, I see the tendency of online tools to erode the visibility of hierarchy and structure in the law as one of the greatest dangers of online legal research (to its practitioners) and, perhaps, even a danger to the broader nature and quality of legal work and discourse.

I should say the tendency and potential of online tools to erode the visibility of lines of authority and context — authority and context don’t disappear (replaced by some incompletely-defined LEGO-block of a legal meme) just like that – even if the collective cognitive structure of “law” is to break down over time it will be a slow process, leaving residues and traces, and the practitioner who understands those residual patterns is advantaged over the one who doesn’t. Actually, I’d say that so long as law is law our ability to make reference to these structures and contexts is essential. We’re all legal realists now, and we ‘play the game’ with self awareness, but nonetheless judging and other law-making activities are bounded and legitimized only by the necessity to make legal argument within the bounds of established hierarchies of authority and precedent, and with shared reference to external authority (both sovereign and cognitive) as defined and accepted within communities of practice. It is that “cognitive” part where I worry about researchers who’ve lost touch with context-giving research tools – it seems like there is value in crafting an argument, not just from a grab-bag of authorities, such as what court p has said about similar-scenario x, but also in knowing how legal issue y has traditionally been understood as a close cousin of legal issue z and how the law in a subject area y has tended to show more favor to analogy a than to analogy b because of a‘s relevance in the law of z.

The research component of this, and the law librarian’s stake, is that tools can either make those contexts and networks of authority transparent or not. I’d agree that we may reach a point where book-metaphor and library-metaphor representations aren’t the best way to clarify those connections — something “webbier” may be emergent. But context-providing research tools, from primary sources constructed as statutory codes to secondary sources like well-accepted narrative treatises that give voice to a practice community’s understanding of how its subject area is contextually organized, make it easier for practitioners to make good legal arguments and, presumably, for lawmakers to make good law. Tools that obscure context make it harder.

The existing generations of online research tools have obscured context — their focus (technically necessary at their inception) on literal, Boolean-based, textual retrieval targeted to an individual “document” removed all visible context. This is critically limiting as soon as search turns to something like a statutory code, that does have internal order, and accesses it using tools and paradigms (e.g. text-based query with each section defined as a ‘document’) that actively hide that internal order and context. For cases, bypassing a Digest does remove one external giver of order/context, but at least judicial opinions, in some sense, really do “float out there by themselves,” each an atomized document, just as they appeared in early Lexis or Westlaw, connected to one another primarily by way of citation and cross-reference (i.e. links). So a web of “links” — cross-referencing documents over time — isn’t actually a bad way to represent judicial law. New tools, like Westlaw Next, that (finally) replace both Boolean literalism and the crude relevance ranking of their old ‘Natural Language’ with the connection-seeking, link-parsing, logic of post-Google search engines are important, and will doubtless change the way we search, read, and write law. But I still have trouble seeing how such tools would replace some need to communicate structured context to the researcher. And so I remain skeptical of the logical next step, which would be a collapsing of any remnant of the “library metaphor” database directories into one big wide open Google-esque search box.

I have more to say about Jason’s second post, where he grapples toward a sort of object-oriented approach to legal argument (using a LEGO metaphor). So far as this seems to be a recognition of the inadequacy of some CALR equivalent of the basic Google search, while still looking for some alternative to the old database silos, I’m in sympathy. But that’s a whole other discussion…

It was a semester during which we saw the expansion of Google Scholar results to cover case law and had our first glimpse of Westlaw Next. One theme I try to get across to the students is that we now work in a research environment (in fact we’ve been there for some time) in which you don’t have to work with tools that convey and communicate structure – you have to provide the structure, and be alert to which resources (treatises, statutory codes used correctly) continue to tell you something about the context and organization of the law and the ways in which some community of practice has constructed a broader notion of legal structure. What will be the long-term effect on the law of legal research conducted (and therefore legal analysis informed) primarily within an organic “web” of resources, rather than bounded by ordered, classified, finding tools (like the case law Digests)? I don’t know. But I do know that, in the transitional environment of the foreseeable future, the inherited cognitive structures of the law aren’t going to simply collapse — they’ve merely become easier to miss for the casual user of unstructured online search tools. And that law school legal research instruction (which once could perfectly appropriately focus on the limited, mechanical, skills tailored to a closed universe of discrete tools) must instead now focus on training students to be conscious of and attentive to their own interactions with retrieved legal work-product. It has, I think, become much harder (if it ever was easy) to separate research skills from some broader idea of knowledgeable information seeking.

I still think most of the questions I asked in July are the right ones for libraries and electronic reading devices. There has been a certain amount of publishing industry hand-wringing about digital books and in the rising popular interest in ebooks prompted by these developments on the device side. I remain surprised that there is less interest – or at least less interest beyond gadget-based curiosity – in libraries, at least in my law-oriented corner of the library world. While many libraries do offer ePub books for “loan” (at least for the Sony Reader and other readers compatible with both the ePub format and Adobe’s DRM) this is a pretty artificial model of “lending.” I think there’s a real question of whether online, as opposed to paper, distribution of book length materials doesn’t fundamentally undermine the central market failure on which lending library business models are based – and I find very little discussion of that.

The radical thing here is the untethering of digital texts from the networked PC. Unless we instead end up with a truly ubiquitous, fast, free, universal, seamless wireless network in the near future (which seems less likely for multiple technical, economic, and social reasons) I would expect that portable devices that can ‘carry’ an un-tethered digital version will be an important part of text and reading ‘going digital’ any time soon (whether through dedicated digital-ink devices or not). I’d put the likelihood that the “flip” will happen soon moderately high.

Suppose that what we are seeing now follows the pattern set with music, movies, digital photography — i.e. the digital form of the medium goes through something of a slow build for a number of years ( already far more than a few for text, of course), there’s a lot of skepticism, the economic sense of the transition gradually appears to key participants, and then when a critical mass is reached the “flip” to digital dominance is very rapid. If so, in a very few years we may see most personal, consumer, book consumption “gone digital” and the analog medium reduced to a niche product — maybe a bigger niche than vinyl records or slide film, say, since there are plenty of special cases for books (art books and ‘coffee table’ books for example, and the very existence of book collections as decorative objects in the way few other analog containers have been) — but a niche all the same. There’s a lot yet to happen on the technology front in terms of format standardization, etc., but as library leaders and managers we should be prepared for this transition to happen, and possibly quite soon.

If this is right, what does that mean to us as libraries? We come from a background of an at least partially failed model of “ebooks” for the library — the kind that required sitting in front of the online computer and that I’ve rarely been able to get faculty and students to consider consulting in lieu of a printed copy. They work for some kinds of works (the O’Reilly stuff on Safari), and do the trick in a pinch, and my anecdotal sense is that general acceptance has (slowly) increased over time but it is still pretty low. Treatise content in Lexis and Westlaw is another model, but an unusual one and perhaps the frustrations of legal-research instructors in urging students to use it in a print-like, contextualizing, manner indicates a ‘failure’ of a different kind.

I’d be interested in having a discussion and hearing about any work you all have put into some of the ensuing challenges/tensions. Points that occur to me (in addition to formatting/footnote/etc. issues) include:

1) Foretaste of the death of the traditional library “intermediary” role for ‘ordinary’ (popular, mid-price or lower) books? A big challenge for public libraries, a medium-sized challenge for special libraries like “us.” A move to publication of ebooks in either a very locked-down format (Amazon Kindle) or through a move to sale/dowload of DRM-free packages (as has ultimately happened with music files) focuses on the end-user in a way that removes a library-type role. Is there a library model? A very big “maybe” for some of the Kindle-competitors with lendable DRM models… interesting note that (contrary, I think, to much of the ideological leaning in the library community) library ebook models all seem to *require* robust and functional DRM. And even then, the removal of physical objects and associated constraints of place could allow an emergence of non-localized subscription libraries or any number of other lending or rental business models to supplement and compete with the traditional localized library.

2) Books are closer to the heart of the traditions and analysis of copyright than any other media and for books to ‘go digital’ will heighten the tensions in the law and policy of content distribution — e.g. amplifying what seems to be the decline of the relevance of the distribution right and its “first sale” limitation.

3) What is the model for text works that have never been economically produced and sold to individuals? In law, this would include the full blown multi-volume treatise, for example. Are they a dying medium (there’s been commentary/discussion about new models for casebooks/textbooks — maybe for the more scholarly long-form work as well)? Will they still be “ok” with the online tether for authenticated, library-based, site-licenses and other group subscriptions? Susceptible to ‘lendable’ DRM on individual chapters/units (is it robust enough for the vendors? capable of offering adequate user protections to prudent purchasing libraries?)? Or are these just not really “book like” any more in the way they are used — more susceptible to integration in the online legal research systems with predominantly non-linear access and use steered in significant part by behind-the-scenes Lexis and Westlaw algorithms.

4) PDF libraries (Hein Online, JSTOR, etc.) — most devices can show pdfs, but the hardware-reader ideal would be a “reflowable” style document not offered by these repositories for the very good reason that it would be at odds with one key purpose of their fixed, document-image, pdfs to provide an authentic photographic reproduction of a definitive, print, “original.” Will they need to offer both? Must our concern with (or belief in) the sanctity of paginated print-derived “original” versions die out?

5) What’s Google up to? It seems like whatever comes of their intention to sell e-book files of all of the public-domain content in Google Books and in-copyright books from participating publishers will be a potential game-changer in terms of defining formats and user expectations.

6) Or, does this even affect our content/users at all? After all, the only radical thing here is the removal of the tether to the Internet. But most of us and of our users sit most of the day in front of, and access all of our work content on, networked PCs anyway (though it doesn’t stop us/them from printing out articles by the ream). Some of the frustrations researchers rightly express with online formats as versus “book based” research are not helped by portable readers — e.g. you can’t even imperfectly emulate having a bunch of volumes open on the desk in front of you at once unless you have a lot of open browser windows on a distinctly non-portable multiple-monitor computer. E-reader hardware devices are even more compromised in this respect than is the big-screen desktop computer, and are no where near cheap enough (yet?) to use multiple devices to stand in for our multiple print volumes open on the table (which would pose other difficulties anyway). Even so it seems like the things our users most want to have ‘on the run’ (journal articles and other periodical articles, single-volume monographs…) will ultimately be heavily used in this format. In a weird sort of way, if portable reading devices pull users away from spending the day in front of the PC with 43 browser windows open, could it even help bring a return of sorts to linear reading and analysis? Will study tables of the future consist of large display panels in which to plug ones personal portable device in order to flexibly display and arrange textual/literary content?

7) Another “blow over” possibility — will e-reading devices never adequately substitute for printing it all out? After all, the reason those networked ebooks are hugely unpopular with our users when compared to print books, but that electronic journals are just the opposite (with most users unwilling to even look at a printed journal volume) is that journal articles are available for printout and networked ebooks typically have major printing constraints. Are e-reading devices that scholarly readers (who mark up and take notes and highlight on those voluminous printouts) will accept still farther away than my other comments/questions account for?