| Graeme's profileBooks and LibrariesPhotosBlogLists |  Tools  Help |
|
January 30 In praise of the declarative sentenceAuthors sometimes express some puzzlement or frustration at the process people use when selecting a book in a bookstore. I'm happy to offer them this case study. I was in the local bookstore the other day, waiting for the reading of Blown to Bits to start, and I noticed Watt, by Beckett. I saw this blurb on the back: I could show you a Beckett sentence as elegant in its implications as the binomial theorem, and another as economically sphinx-like as the square root of minus one, and another, on trees in the night, for which half of Wordsworth would seem a fair exchange. The declarative sentence, he makes you suppose, is perhaps man's highest achievement, as absolute as the egg was for Brancusi. — The New York Times Book Review
You're welcome. January 28 "Designing the Digital Experience"This post is less than a review of Designing the Digital Experience: how to use experience design tools and techniques to build Websites customers love, by David Lee King. The book is divided into three parts: Structural Focus, Community Focus and Customer Focus. Early on, in Chapter 3, King focuses on the process of designing and building a web site. Not what the web site will look like when you get there, but the process of getting there. He discusses three process models: Jesse James Garrett's User Experience model, David Armano's Experience Map model, and 37signals' Getting Real model. This is a good place to start. I guess any moderately organized person would start by thinking of the goals they wanted a web site to achieve, but a process model helps to avoid overrunning the goals and requirements in your haste to get to implementation. On the subject of usability, King points the reader to Steve Krug's Don't Make Me Think: A Common Sense Approach to Web Usability. Even without reading the book, I think this is a great principle. Not thinking has many of the same benefits as laziness, the first of Larry Wall's virtues of a programmer. Libraries should design their web sites to be used by lazy, unthinking patrons. No, really. The weakest section of the book is chapter 6, "Emerging Tools for the Digital Community". For example, King spends less than a page on wikis, and makes the same mistake that I've noted previously: conflating the authors of a wiki with the readers. King discusses Flickr, but provides no guidance on whether a library should use Flickr or host its own photo gallery. For me, the key point about customer interaction was made by Jan Carlzon, whom King quotes on page 133: Each point or interaction with your organization can be referred to as a "moment of truth", a concept first introduced by Jan Carlzon, the former president of Scandinavian Airlines, in his 1986 book entitled Moments of Truth. Carlzon defines the moment of truth in business as: "Anytime a customer comes into contact with any aspect of a business, how ever remote, is an opportunity to form an impression."
King also quotes Tom Kelley of IDEO (page 132): According to Kelley, your car-buying experience (and also your experience interacting with the car company) begins before you walk onto the lot and continues after you leave with your purchase.
This is a short book (182 well-written pages), and it seems thin in some areas, but for me the whole is greater than the sum of the parts. Perhaps it's an unnecessarily mischievous thought, but occasionally it seems to me as though librarians write books as annotated bibliographies, as directions onward rather than as containers of content. But building an effective web site touches so many disciplines that this may well be the most sensible approach. January 26 Taxonomy versus FolksonomyOver at the official PLA blog, Lorraine Squires discusses whether tags should be controlled (taxonomy) or uncontrolled (folksonomy). For some reason, this triggered the following not-completely-serious riff on the bibliographic control of user-contributed content. I literally wrote this post as I was falling asleep last night. I guess I shouldn't read anything about cataloging just before bedtime.
I've slowly become convinced that user-contributed content (UCC), including tags, reviews and lists, should be subject to bibliographic control, but further study is needed to determine the right form. For example, should (uncontrolled) tags be classified using (controlled) tags or by using the same subject headings as the underlying catalog. Lorraine apparently sees controlled tagging and uncontrolled tagging as mutually exclusive, but I don't see why they can't co-exist. In particular, if they are going to co-exist, controlled tags could be used to classify the uncontrolled tags, so that several or many uncontrolled tags would be associated with a single controlled tag. But that's just the beginning of the issues to be resolved.
For libraries with foreign language populations, which must be most public libraries in the United States, foreign language tags may be problematic. One answer is to tag the tag in both its native language and in English, but it may not be so easy to determine the language from the one or two words of a tag. You would also need to decide whether to use a foreign culture classification, or merely the (English, controlled) tag translated into the foreign language. Particular care would also need to be taken with disadvantaged language populations, such as the Basques or Kurds. Since tagging takes some time to reach critical mass within a given library or system, pooling of tags from multiple systems has been suggested. In this case, the source system would be identified as the publisher in the record for the tag (either controlled or uncontrolled). Identity control would also be required, both to distinguish different people with the same user id in different systems, and to identify the same person with different ids. When combining tag sets from different systems, it seems inadvisable to simply conflate tags with the same name. The right approach might be to use the FRBR model and treat the tag from each system as a separate expression of the same work. This might be just the push your library needs to move to FRBR. The OPAC would need to be strengthened to allow the user to proceed from a given (tagged) catalog record to either (i) records with the same uncontrolled tag, or the same controlled tag, (ii) using their own tags, or tags from local users, or tags from any user. I suspect that the user interface would be simpler than that explanation of the options. User-contributed reviews seem much easier to handle, since they could be entered into the catalog using whatever ruleset was used for electronic documents, again setting the library or library system itself as the publisher. Since reviews already move from system to system (e.g., the OPAC for a public library pulls reviews from Amazon), here identity control is urgently required. User lists are unfortunately more complicated. Because entries may be copied from list to list, and each entry in a list may have an attached note, which naturally has an author, there seems to be no alternative to cataloging each list entry individually, but then you have to reassemble the list. One possible direction is to follow FRBR and treat a list as a work consisting of an aggregate of works. Copying an entry from one list to another would simply be a matter of copying the individual work (the list entry, not the underlying work the entry refers to) to the destination aggregrate and adding an association (i.e., "copied from") between the source and destination works (again, the list entries). The owner of the destination list could then author a note associated with the new entry. Naturally, the library could provide its own controlled lists to supplement the uncontrolled lists provided by patrons. When displaying a list, the OPAC would need to offer the user the option of displaying only their own notes, or the notes belonging to entries the entry was copied from (available by following the "copied from" association backwards), or any notes associated with the underlying work. Happy Australia Day! January 25 How do you compete with free?I've been a happy user of Library Elf for almost two years. It's a web-based service which monitors your library account and sends you email reminders when a requested book arrives, or a book is coming due, or becomes overdue. That's something that the local library also does. So Library Elf is competing with free. The library will send you a reminder two days before a book is due. Library Elf will send you a reminder when a book is due, or up to seven days in advance, or on a particular day of the week, or every day. The library will send you a reminder when a book is fourteen days overdue. Library Elf will send you an overdue notice immediately, and will optionally keep doing it every day. The library will send you a reminder when a requested book has arrived. Library Elf will send you one reminder, or one every day. If you have two books due, the library will send you two emails. Library Elf will send you one. In fact, Library Elf rolls up all your reminders into one email, so that when you're going to the library to pick up a request, you can see whether you have books to return. The email from Library Elf includes a calendar, so you can see at a glance whether you can wait until the weekend to pick up or return a book. You can associate more than one library card with a single Library Elf account, and one email will include the activity from all the library cards. Your account on the Library Elf web site includes a single page which lists all of your loans and requests, and includes a calendar. The library catalog lists loans and requests on separate pages, without a calendar. Starting this month, Library Elf began charging for premium services, the services I've described here. Some services are still available for free, but not support for multiple cards on one account, something I rely on, so I'm paying Library Elf twenty dollars a year for something the library does for free, just not quite as well. Or the library could get a library subscription from Library Elf, for about a dollar a year for each patron that uses the service. Or the library could improve its current email service. January 23 "Library 2.0 and Beyond"This post is not exactly a review of Library 2.0 and Beyond, which is a collection of eleven articles edited by Nancy Courtney.
Often, discussions of Library 2.0 collapse into a discussion of which Web 2.0 features can be applied to libraries. Then, a catalog from the 1990's, or even the 1980's, is taken as a fixed point around which Web 2.0 features can revolve. Michael Casey's article on "Looking Forward to Catalog 2.0" is an excellent antidote to this reductionism, and sets the bar for next generation catalogs appropriately high. I just want to nudge it up a little further.
Talking about the catalogs of the past (and probably those of the present as well), Casey says (page 17):
Which suggests a thought experiment of sorts. Suppose an author was coming to give a reading at your library in two weeks. This event would naturally appear in the calendar, and perhaps be announced in a blog, on your Facebook page, or on Twitter. But what happens if a patron searches for the author in the catalog? I suggest an X Prize for the first catalog to automatically display relevant information from the library's calendar.
It's important to be clear in evaluating what a patron sees in response to a catalog search. One error to be avoided is to suppose that the data to be displayed is limited to the results of the search. As you can see from the example of the author event, there's really no limit to what might be helpful to the user. Another is that the effectiveness of the display is determined by the data presented. Casey says (page 20):
with which I completely agree, except for the "almost". Casey also has a great description (page 21) of how customizable RSS feeds can be used to deliver catalog updates in particular subject areas to interested patrons.
I'm going to look in some detail at Chad Boeninger's article, "The Wonderful World of Wikis: Applications for Libraries", because wikis are something I've been thinking about, and trying out, recently. Boeninger makes a very telling observation. He has entered information about the Regional Encyclopedia of Business & Management into a wiki (page 30):
Here's my biased gloss on these two sentences: catalog - less useful; thoughts - more useful. Now the thoughts are those of a professional librarian and not patrons, but doesn't this feel like the camel's nose, the slippery slope, etc?
Well, yes and no. Here's Boeninger's description of a wiki (page 25):
It's true that the pages of a wiki can be created and edited by a group of people, but that's not necessarily the same group as the readers of the wiki. So we're left with the somewhat less impressive fact that a wiki is a web-based tool for one set of people to generate content for another (or possibly the same) set of people. And it makes perfect sense that professional librarians should be generating content for patrons.
Boeninger gives three very strong examples from his own experience (although for some reason not in chronological order) but he doesn't go far enough in presenting other ways in which wikis can be used. To give an example I've been thinking a little about, a wiki could be used so that several libraries could jointly develop and share homework resources for teens. Homework resources are the sort of loosely structured content that fits the wiki model pretty well, and libraries in adjacent towns are likely to share much of the same material in common subjects such as American History. The key point is that the "community of users" who create and edit the wiki doesn't have to be an existing community. It can be a group created de novo simply by the existence of the wiki. It's also an example of the truism that the Internet can erase distance.
Boeninger says (page 31), "Self-hosted wikis require knowledge of MySQL and PHP, as well as some experience with web server administration." It happens I've just installed PmWiki on one of my Windows XP systems, and I can tell you exactly how it's done:
Congratulations, you're running a wiki! Like several other simple but popular wikis, PmWiki doesn't even use MySQL, so that's not at all relevant. Now it's true that some knowledge is required — like setting the administrator password (!) in various places — but Boeninger's characterization is a stretch. Ellysa Kroski has an article on "Folksonomies and User-based Tagging". I'm a tagging zealot, so perhaps I'm not the fairest reviewer of this material. While I found the article a useful introduction, it misses one of the advantages of tagging blog posts, and is weak in its defense of tagging. The omission is simply that if a blogger tags their posts, you can normally subscribe (in RSS) to posts with a particular tag. This means that a blogger can post on completely unrelated topics (say, cataloging, world peace, and pictures of his cat) and people can choose to read a single topic or everything. Or a library could post all its news to a single blog, with tags that would allow patrons to filter it if they chose. Vice versa, tags don't seem to be particularly useful for searching blogs, probably because the tags used for blog categories (like "food" or "writing") are too general for that purpose.
(Before I post to this blog, I naturally submit each post to the same rigorous fact-checking as The New Republic. I discovered that subscribing to a single category of blog posts is not something that's easy unless it's a feature provided by the blog owner. However, it is possible.)
Kroski gives several examples of systems that use tags, such as (pages 92-93) Flickr, Delicious and Technorati, but doesn't have anything to say about the differences between the systems. Try searching for "cocker spaniel" (no quotes) on each system, and on Google. Not only are the photos of cocker spaniels cute, but so are the differences between the systems. There's clearly something more to be said.
Here's Kroski on the advantage of controlled vocabularies over tagging (page 98):
Unfortunately for Kroski's argument, the Internet makes this pretty easy to check. Let's say we're looking for books on Drupal. We don't have to commit the LoC subject headings to memory because they're online here, where we find that the right heading is "Drupal (Computer file)". We can use that term to search the library network catalog, where we find that this catalog uses the heading "Drupal Computer Program Language". No matter; we've found three books. Which is great, except a keyword search on "drupal" finds a fourth, which has been classified under other subject headings entirely. Three out of four is not good.
I really don't understand how librarians can continue to talk about "the strength of its controlled vocabulary" with a straight face.
The Library of Congress Flickr Pilot Project put several thousand images from the LoC on Flickr where any registered user could add tags or comments. The result was that "the project has been successful in achieving the objectives and desired outcomes of the Library’s strategic goals" (full report (pdf link), page iv). And "less than 25 instances of user-generated content were removed as inappropriate" (ibid.), which seems insignificant, except that since all of the tags and comments had to be checked in order to find those twenty five, it seems to have taken one full-time person (summary report (pdf link), page 4).
That seems to me to be the most important objection to tagging, and it's something that Kroski doesn't mention. Unless you devote significant resources to checking tags, someone's going to add dirty words and someone else is going to be upset.
Eric Schnell provides a useful introduction to different web services in "Mashups and Web Services". But not all his examples (e.g., page 70) are mashups, which strictly speaking are applications which combine data from different web sites. He's also not particularly clear about the range of tools which can be used to create mashups, focusing on developer concepts like REST and SOAP (page 68) to the exclusion of user tools such as Yahoo! Pipes.
There's a lot more helpful content in the book than I've discussed here, including introductions to podcasting, online social networks, gaming and digital storytelling. I can't recommend the book without reservation, but as long as it's not the only thing you read about Web 2.0, Library 2.0, or about the individual topics presented in the book, I enourage you to read it. |
|
|
