Recently I was reading a book on Jane Austen, which quoted a work by Marvin Mudrick. The quote was intriguing so I took the next step of seeking out the full bibliographic citation:
Mudrick, M., Jane Austen: Irony as Defense and Discovery, (Princeton, Princeton University Press, 1951), Pps. vii +276.
The 1951 publication date sent me to Amazon to see whether the book might still be commercially available. (Clearly, it was still under copyright.) There were two used copies available on Amazon, but no preview of the book’s content. Off then to Google Books, in the hope that there might be a preview option there for purposes of evaluating the book before purchase. There was, but every third page was blocked from view. (Sigh.) Which library near me has this book on the shelf? WorldCat gave me much more information; there had been four or five editions since the book’s original publication, and rights to the text now resided with the University of California Press. Even more helpful was WorldCat’s note that apparently a digital version did exist. As Google had specifically indicated there was no e-book edition, it was a good thing I’d thought to check. To be fair, WorldCat did specify that the digital version housed at the HathiTrust Digital Library only offered a limited view, but I still thought it worth the extra click to see whether additional pages might be accessible. Lo, upon landing on the HathiTrust platform, I discovered that Jane Austen: Irony as Defense and Discovery was available in full text — an open access (OA) monograph in the wild. I could engage with the content (read, navigate, annotate, etc.) online, but the site made it clear that it was also permissible to generate a downloadable PDF of a selected page or chapter, or even the full text of the book.
This is a success story of sorts: from the point of discovery, it took less than 15 minutes to reach an OA monograph. Most readers have no idea of the long-term and communal efforts made by committed industry players to get us to this point. And that access is increasing! The most recent example is an announcement from JSTOR regarding its recently launched OA e-book initiative. It involves four university presses and ultimately hundreds of scholarly monographs that will be made available by the close of 2017.
That said, there are some concerns to be addressed before anyone goes off celebrating. Very real complexities have been introduced into the process of moving the reader from the point of discovery to actual access.
In tracking the Mudrick book, I had to navigate four different platforms and, at every stop, was served inconsistent information regarding the degree of access available to me. Whether that inconsistency was due to discrepancies in how supplier systems handle metadata, the quality of the original metadata supplied, or simply a mundane lag time in updating systems can’t be known for this particular example. Too much is hidden under the hood across platforms and systems.
Second, as someone working in this industry, I knew the available information resources. Like most users looking for a book, my first thought was to check Amazon and Google. However, what proportion of users then think to check WorldCat or HathiTrust? Despite best efforts, neither of those can be deemed to be household brands on par with those of the Internet giants. If I had stopped after checking Google, I’d have never reached the available OA edition.
The third concern has to do with the reader’s understanding of the access being granted. Those who took the time to review the linked JSTOR announcement referenced earlier, will have noted that the open access being offered entails six different Creative Commons licenses, each of which specifies particular rights of and constraints on subsequent usage. That’s to be expected, as the rights holders (all university presses) have legitimate contractual obligations to satisfy — to their authors as well as to their parent institutions. But it still represents another layer of complexity that is being added under the hood to automated systems. How do we convey to readers the degree of access being offered them at any given point? The lack of indicators makes the situation sadly opaque.
Some library records offer at least an indication of what degree of access may be on offer. When I looked at an example from Cornell, the record displayed a subscription service link to a digital edition and noted that license restrictions might limit access (if the user is accessing content off-campus, for example). Displayed immediately below that note was a link to Google Books with the added information that only selected chapters might be visible. (Note: The Cornell bibliographic record was one for a non-OA title, and the search that referred me to that result was run through a link found on a third-party website.)
Using a web-accessible discovery service at a second academic institution, I was horrified to see that the library had, when referring users to OA titles on HathiTrust, added a notation to the individual result that read “(Public Domain).” Granted, that was a form of shorthand that conveyed the degree of access available, but it is still important to note that the phrases, “open access” and “public domain” are not synonymous. With regard to the Mudrick title, the University of California Press may have made the content OA, but the rights have not yet fallen into public domain.
In a third approach to aiding users to discover OA titles, an institution in the UK created a LibGuide. There, the orange OA “Un-locked” icon was displayed on the page (a graphic indicator to catch the user’s eye) and a differentiation indicated between OA journal articles and OA monographs. However, in that instance, the library merely linked to content-provider websites — Open Book Publishers, Knowledge Unlatched, etc. Since no indication was offered as to the disciplines such providers were publishing in, the links were perhaps less useful to the harried searcher than they might have been.
Lest anyone think libraries alone are at fault, I would also point out the failure of a particular OA directory that lists 162 publishers of OA monographs. Seeing a well-known content provider’s name, I optimistically followed the link labeled OA Monographs. However, that only led me to an FAQ on the publisher’s site. But, wait! There was an additional sub-menu option that claimed that it would take me to a page where I could view and download the publisher’s OA monograph titles. Again, I clicked through, only to be told, “You have followed a link to one of our old platforms that could not be redirected.” The platform provider suggested that I should now look for a link to their recently rebranded platform. All reference to OA materials was gone. Now, there are all sorts of technology-related reasons why a system might not be able to redirect a page visitor, but to expect the user to begin again at point A without any further assistance or referral to the material originally sought creates both frustration as well as irritation. At a deeper level, it erodes the reader’s trust in the publisher’s platform and brand.
From what other source might I as a reader expect to discover OA scholarship about Jane Austen? The Jane Austen Society of North America periodically publishes a selective bibliography in its OA journal, Persuasions On-Line. Professionally compiled, the bibliography points readers to critical articles and studies by academics as well as to the theses and dissertations by graduate students. The most recent bibliography was published in 2015; however, it includes no OA monographs and the live links to theses and dissertations lead to a commercially hosted platform rather than to any institutional repository. The question immediately arising is whether there is a lack of OA content in the field of Austen studies (a possibility) or if the problem is due to the pressure of working to deadline. Is there simply a lack of time to identify the location of all those theses and dissertations housed in scattered institutional repositories? Had there been some agreement in place between the society and the platform provider? Bibliographies are a recognized downstream tool for discovery, but under what circumstances should they offer preferential treatment to OA content? What are legitimate expectations to build in the mind of a reader seeking OA scholarly output within a given field? More importantly, whatever happened to our obligation to save the time of the reader?
In a brief talk at the Charleston Conference this year, I noted that what was needed was a combined stakeholder effort to successfully connect reader and book while making clear the degree of access being offered. Why would that be important? Because the reader’s understanding of licensed access is what will inform his or her subsequent attitude toward usage.
In my successful experience with the Mudrick book, there were three different platforms (Amazon, Google Books, HathiTrust) visited by the reader in an attempt to determine the value of the content. Each had a different degree of access built in for even the most benign evaluative task. In the more recent example concerning a subject-specific bibliography, under what circumstance does the information professional indicate appropriate direction to the user (whether via a licensed host platform or an institutional repository)? Affordances for subsequent use are framed differently across digital information environments, and it’s not always easy for the reader to grasp intent. There’s certainly no convenience to the process.
Even once they’ve discovered a relevant title and know it to be openly accessible, readers understand the term “access” to encompass greater meaning. It includes awareness of whether the available format will operate on the device of their choice and which functional tools for engaging with the content (navigation, annotation, printing, etc.) are in place. From their perspective, the key question is what can I do with this?
If content providers are worried about cavalier handling of content by readers, then as rights holders, they must assume some responsibility for communicating in clear and meaningful language the permissions that are in place. If there are Creative Commons licenses in use, which ones are they? No one should just rattle off CC BY NC-ND to a reader and expect to be understood immediately. Whether the patron is accessing material from within the library or arriving via some external referral mechanism, there’s some degree of translation required. Can we create automated systems to do that? Yes, but then we need — for each individual, OA book title — complete, explicit, and accurate metadata. Such metadata is no longer an enhanced value-add; rather, it is a functional baseline.
Those who work at content and platform providers, aggregators, and libraries (any and all of those professionals who may regularly read The Scholarly Kitchen) are groaning. They know the importance of metadata for purposes of discovery. They work at metadata, particularly where e-books are concerned. They supply both pre- and post-publication metadata, enriched MARC records — multiple feeds supplied in multiple protocols to multiple third parties. They supply access information — KBART files, EZProxy stanzas, etc. They do supply licensing information. The problem is that elements of the metadata may drop out as an individual item is subsequently disseminated through the various environments. Lack of awareness within the community isn’t the issue. Service providers are aware. But there are realities to be faced.
One obstacle (particularly with book content) is that many third party platforms were not initially engineered with OA in mind. Such a concept — whether in terms of pricing or in terms of access management — simply was not part of the functional requirements built into the original infrastructure. Plans to re-engineer information architecture or develop workarounds are ongoing but, for the most part, are not wholly in place. (Hence, the presence of gently worded 404 messages, such as the one recommending the user circle back and look for the publisher’s re-branded platform.)
For our part in the information community, it will take a combined stakeholder effort to develop consensus on requirements for:
- complete, explicit, and accurate metadata for OA monographs;
- consistent messaging about OA licenses; and
- commonly accepted formats across operating systems and platforms
Only when content and platform providers, in conjunction with libraries, have a reasonable system in place can they help readers grasp the meaning of OA terminology and adopt the appropriate behavioral norms.
For the record, the National Information Standards Organization (NISO — full disclosure, I am NISO’s Educational Programs Manager) is bringing together a range of interested entities for the purpose of exploring answers to the OA discovery and access puzzle. Already, there is a NISO Working Group looking at the development of recommended practices for e-book bibliographic metadata. The NISO Discovery to Delivery Topic Committee is tentatively examining approaches that might be useful in lowering barriers to discovery of open content — open journal articles, open educational resources, and the openly accessible monograph. Certainly, industry engagement is welcome.
None of our activity in supporting OA in advanced information systems and services is particularly transparent to readers. That is okay. They don’t want to be inundated with details; they just want to engage as fully as possible with the content itself. Let’s let them do it.
20 Thoughts on "Monographs, Transparency and Open Access"
I am impressed by the extent of your explorations, but they raise another disturbing question. How many of these indexing and cataloging sites did you either pay to access or gain access through your institution? For most, I presume you were a public user expecting a free service. If so, it seems you got what you paid for!
As you correctly imply, indexing content requires skill and dedication. What is the business model to provide a viable indexing service?
It’s worthwhile noting that I did not initially expect to be able to find an open access version of Marvin Mudrick’s book. In visiting Amazon and Google, I assumed that I would be given sufficient opportunity to examine or evaluate the text before actually buying the book. I only went to WorldCat because I wasn’t able to do so prior to purchase which meant figuring out which physical library near me might have an available copy. I can and do pay for my research texts, but like most educated consumers, I want to leaf through the full copy to gauge what I might be willing to pay for a new or used edition. (Please note that if I had been willing to buy sight-unseen I could have purchased the latter via my initial visit to Amazon.) I am just reminding the various entities in the supply chain (publishers, aggregators, librarians, etc.) that complete and accurate metadata is needed to enable users and potential purchasers to discover content availability and understand the appropriate scope of use. Such metadata is a form of advertising when integrated with the various discovery tools. It ensures that I will find, assess and subsequently re-use the content. I should have assumed that ensuring metadata quality would be a requisite cost of doing business in scholarly publishing.
Jill, I certainly didn’t mean to imply you were trying to get something for nothing. However, the business models of many content providers apparently do not include providing adequate metadata to assess their offerings. It’s sort of like walking into a store where the merchandise is behind dark glass and there are no price tags. Strange business indeed! I think you area doing everyone an important service pointing out these problems.
Using Google Scholar advanced search on “irony as defense” in titles quickly gives links to both the Google Books and JTOR versions, plus a lot of items that cite the book, which might be useful in their own right. However, what is or is not OA is not indicated.
David, what your link provides me with is largely either bibliographic citations (which I already had) and book reviews from scholarly journals. While I might have relied on those book reviews for purposes of assessment, they may or may not actually discuss the specific work by Austen in which I have an interest. It’s not that the book review’s value is negligible; it’s just not adequately specific. (For the record, I am a registered user of JSTOR and I do take advantage of the Bookshelf option for viewing journal articles as needed.)
Jill, thanks for an excellent post highlighting the enormous frustration that may share (including publishers!) in trying to make OA content discoverable, accessible, and usable. As you rightly point out, the infrastructure for books lags behind journals although there are a number of initiatives underway to try to address this both here and in Europe – glad to hear that NISO is also engaging. In the meantime, the best solution seems to be the rather obvious one of hosting content where the users are – we have certainly seen significantly higher usage of our OA monographs in the few months they’ve been on JSTOR than on our own platform.
Alison, I’m grateful to you for taking the time to comment. I don’t think I was presenting *new* information in my posting here, as I know that this is a long standing issue. I was simply trying to relate what my personal experience was, as a means of illuminating what users may encounter at any given point in their pursuit. It’s not just one sector’s problem; effort is needed by the information community as a whole. We’re obligated to find ways where we may better enable discovery, access, and use of solid, published research.
Your post also highlights the need for marketing OA monographs. It’s not sufficient simply to post an OA monograph at some web site; it’s important also to alert those who might be most interested that this new book exists. All too often it seems to be assumed that marketing is not a cost that OA publishers need to take into account. Your post about discoverability underlines the need for this to continue to be a necessary function in publishing. Of course, you’re talking about a monograph that was published way before we entered the digital age, but the problems you highlight point also to the need for publishers of new OA monographs today to pay attention to discoverability issues.
Sandy, thanks for taking time to comment and I’m sorry that it’s taken me this long to respond. One of the interesting aspects of this progress towards broader (open) access to scholarly materials is the full scope of “stuff” that remains valuable as information. I have four or five biographies of Jane Austen and only one of them is currently available in electronic form. Yet that type of historical information is one of the most durable over the long-term. I’m not sure it’s a practicable idea, but it would be wonderful if the information community could start looking at evaluating older content from recent decades and begin to prioritize what is digitized. Academic trade titles, in particular, seem to be overlooked. (At least that’s my impression. I have no firm data from which to draw.)
Jill (and Alison and Sandy), Thanks for an excellent introduction to some real current problems. This issue emerged loud and clear in the work that Michael Jubb and I were doing for the AHRC/British Library Academic Book of the Future project on Discoverabilty, Demand and Access, and specifically the role of intermediaries in the academic supply chain: our brief report might conceivably interest SK readers and is now available at https://academicbookfuture.org/discoverability-demand-and-access-the-role-of-intermediaries-in-the-uk-supply-chain-for-academic-books-richard-fisher-and-michael-jubb/
Jill writes, “The 1951 publication date sent me to Amazon to see whether the book might still be commercially available. (Clearly, it was still under copyright.)” However, there are plenty of books published in the US through 1963 that are in the public domain because copyright was not renewed. Many of these have been made available in HathiTrust and elsewhere in full text. In this case, though, it appears that UC Press gave permission to HathiTrust to make the copy available.
I find that Jill’s search path is a probable one for someone who knows key things about the copyright environment, like the usual cutoff date of 1923. But with so much content after that date made available online, whether through rights clearance or permission of the rightsholder, plus unauthorized copies floating around online, I think readers searching for known items are much more likely to simply bypass channels like Amazon and library catalogs and instead start by putting the title of the book into a search engine to see what they can find. Indeed, a Google search in incognito/private mode in your browser (to avoid the effects of the personal search bubble) pulls up the HathiTrust record as the fourth result. If you click on that, you see that the full text is available. That’s an easier path, and one that I think readers increasingly take.
Hi Jill, Interesting article, one that I have heard discussed in various circles. You say that “Such metadata is no longer an enhanced value-add; rather, it is a functional baseline” (which I would agree with). This suggests that discovery is getting harder and will continue to do so without rich metadata. But given that OA articles/monographs access is usually free, do you think access should be/remain to be free via discovery products (via aggregation service or similar)?
Andy, thanks for continuing the discussion here. From my perspective as to what the appropriate priorities might be, I would hope that, for publishers as well as libraries, enabling discovery in support of subsequent use would be the priority. Enabling discovery however does absorb resources which as you note raises the question of whether there should be a charge for services that handle both discovery technologies as well as the metadata being piped through those technologies. For OA materials in particular, I’d hope that the discovery costs would never be charged back to the user, because again the objective should be to support the broadest possible use. That said, if discovery services or products were to be able to identify and deliver enhancements to either the technology or the metadata, then of course I’d support paying for such an enhanced service. (The enhancement would have to be a demonstrable improvement. But we all know that “better, faster, cheaper – pick two” remains a constant in most things.)
Jill, I like the idea of more information on availability and licensing or ‘openness’ data being included in the metadata for published material. But is there not a problem in principle with getting this to work. Standardly metadata covers properties of the published object that dont change. But the accessability or licensing arrangements change very frequently and rather unpredictably as business models and platforms change. I dont see much realistic prospect of solving this problem. Apart from anything else, metadata itself becomes progressively less useful if one is always going to be in the cleft stick of not knowing whether or not the system that provides the metadata is being kept up to date. The fact that metadata, if it is good, does not change *much* is part of its usefulness. However you are quite right about the problem, it would be great and is surely very necessary that licensing differences be understood and clearly interpreted and positioned.
Adam, that’s a legitimate point to raise — that licensing arrangements shift. Like you, I don’t necessarily see that there’s an easy solution. However, with regard to OA monographs, rights holders can assign metadata that makes it clear. It just needs to be done as early in the process as possible so that, given the time lag in populating updated information across a variety of platforms, users can readily discover the resource at the point of need.
Had you simply Googled author+title “Mudrick, M., Jane Austen: Irony as Defense and Discovery” you would have found the 2nd result = Online Books Page at Penn, where John Mark Ockerbloom has long maintained an excellent referral service for free online works of quality. It would not have taken even 15 seconds to find it. Perhaps you are overthinking it: Amazon, Google Books, WorldCat–be less clever, use the technology.
That’s true, Paul; still, training is a hard thing to shake off. I appreciate that you pointed to John Mark Ockerbloom’s resource page, as that’s another resource to add to the tool box. But if one has no reason to assume that something is available as an open access publication (and I did not have reason to think that something published in the 50’s would necessarily be OA), then it makes sense to adopt the avenue of approach that I used.