The Long Tail and Libraries

To date, Lorcan Dempsey's Libraries and the Long Tail has offered the most insightful analysis of the Long Tail's importance in libraries. As I've written before, the Long Tail is an effective strategy to utilize when implementing Library 2.0 for the modern library. The question is: could it be implemented without a huge overhaul of most existing libraries? These are some points that Dempsey argues:

(1) Transaction Costs - The better connected libraries are, the lower the transaction costs

(2) Data about choice and behavious - Transactional behavioural data is used to adapt and improve systems. Examples of such data are holdings data, circulation and ILL data, and database usage data.

(3) Inventory - As more materials are available electronically, we will see more interest in managing the print collection in a less costly way. Although historical library models have been based on physical distribution of materials, resources are decreasingly needed to be distributed in advance of need; they can be held in consolidated stores

(4) Navigation - There are better ways to exploit large bibliographic resources. Ranking, recommendations, and relation help connect users to relevant material and also help connect the more heavily used materials to potentially useful, but less used, materials

(5) Aggregation of Demand - The library resource is fragmented. In the new network environment, this fragmentation reduces gravitational pull, which means that resources are prospected by the persistent or knowledgeable user, but they may not be reached by others to whom the resources are potentially useful. What OCLC is doing is making metadata about those books available to the major search engines and routing users back to library services

Google = God?

Maybe Google got it right all along. But is it God? That often appears to be the way that most people do their searching online nowadays, expecting to find the answer to just about anything. Yihong Ding calls this kind of searching, "oracle-based" web searching, which search engines such as Google are assumed to know everything. But this worked relatively well in the early days of the Web because it a pragmatic and affordable strategy; at that time, the quantity of web resources was comparatively small. We rarely searched for meaning. Based on this premise, to build a semantic oracle (i.e. Semantic Google) is equiavalent to create a real God (who knows everything) to human beings.

Perhaps, according to Ding, a better alternative is collaborative searching. Since current answer-based search strategy is motivated by questions, collaborative search is motivated by answers. In our answer-based search model, the ones who answer questions may not have passion (or enough knowledge) to questions. But an inanimate search engine such as Google doesn't know this -- nor does it care.

However, Web 2.0 is slowly changing this course of searching. Already, search engines such as Cha Cha are harvesting collective intelligence and wisdom of the crowds to retrieve more "relevant" results. Ding goes one point further: Web 3.0 will be based on community-sensitive link resources. It will reverse the relation between horizontal search engines and vertical search engines. The current model of vertical search engines being built upon generic search engines are not working well because they are too immature to provide communicate-specific search by themselves. (Just look at the limitations of Rollyo). What will the Semantic Web search engine look like? Maybe something like this.

The Future of I.S.

Meet Ramesh Srinivasan, professor of Information Studies at UCLA. During my trip to Los Angeles, I met with the IS faculty and visited some of the libraries there at UCLA. My conversation with this up and coming academic star was fascinating to say the least. Ramesh's interests includes exploring connections between diasporic/indigenous communities and new media and how information technologies shape, transform, and differentially impact nations, cultures, societies along educational, political, health-related, social, and infrastructural dimensions.

Among his more interesting projects is the Emerging Databases, Emerging Diversity (ED2): National Science Foundation-funded initiative to study methods by which digital collections can be shared via systems that maintain diverse tags, ontologies, and interfaces. In collaboration with Cambridge University's Museum of Anthropology and Archaeology, and the Zuni community of New Mexico, the $300,000-funded project inquires how digital access to ancestral objects affects diverse communities. Ramesh's work involves extensive field work in places like Kyrgyzstan and India. (Exciting!)

The faculty at UCLA represents Library and Information Science's gradual shift towards the iSchool movement. Academics such as Ramesh Srinivasan represent the new face of LIS. This has important implications for librarians, who will ultimately be bred and nurtured by these new scholars nontraditional perspectives to LIS. Rather than basing their studies on users of libraries, newer scholars such as Srinivasan, whose background is as diverse as his research (his PhD is in Design), go beyond the traditional domain of LIS. Inevitably, librarianship will change because of this new approach. New ways of thinking and research will be injected into the profession -- perhaps this is where the source of innovation in libraries will come from as well. From the classroom.

Metcalfe's Law

As I had opined in previous posts, the next stage of the Web will be built on the existing infrastructure of Web 2.0. One of the foremost thinkers of the Semantic Web makes an insightful analysis of the progress from Web 2.0 to the Semantic Web. Along with Jennifer Golbeck, James Hendler puts forth the idea of Metcalfe's Law, arguing that value increases as the number of users increases. Because of this, potential links increase for every user as a new person joins. Not surprisingly, Metcalfe's Law is the essence of Web 2.0.

As the number of people in the network grows, the connectivity increases, and if people can link to each other's content, the value grows at an enormous rate. The Web, if it were simply a collection of pages of content, would not have the value it has today. Without linking, the Web would be a blob of disconnected pages.

As information professionals and librarians, we shouldn't miss out on the obvious links between Web 2.0 and the Semantic Web. Social networking is critical to the success of Web 2.0; but by combining the social networks of Web 2.0 with the semantic networks of the Semantic Web, a tremendous value is possible. Here's a scenario from Tom Gruber which I find very compelling:

Real Travel "seeds" a Web 2.0 travel site with the terms from a gazetteer ontology. This allows the coupling of place names and locations, linked together in an ontology structure, with the dynamic content and tagging of a Web 2.0 travel site. The primary user experience is of a site where travel logs (essentially blogs about trips), photos, travel tools and other travel-related materials are all linked together. Behind this, however, is the simple ontology that knows that Warsaw is a city in Poland, that Poland is a country in Europe, etc. Thus a photo taken in Warsaw is known to be a photo from Poland in a search, browsing can traverse links in the geolocation ontology, and other "fortuitous" links can be found. The social construct of the travel site, and communities of travelers with like interests, can be exploited by Web 2.0 technology, but it is given extra value by the simple semantics encoded in the travel ontology.

Genius.

Pragmatic Web as HD TV

The Pragmatic Web: A Manifesto makes a return to simplification. For all the hype about Web 3.0, we've still seen very little substantial evidence that it exists. Schoop, De Moor, and Dietz proposes a "Pragmatic Web" as a solution which does not replace the current web but rather, extend the Semantic Web.

Rather than waiting for everyone to come together and collaborate -- that could take forever or worse yet . . . never -- the best hope might be to encourage the emergence of communities of interest and practice that develop their own consensus knowledge on the basis of which they will standardize their representations. Thus, the vision of the Pragmatic Web is to augment human collaboration effectively by appropriate technologies. Thus, the Pragmatic Web complements the Semantic Web by improving the quality and legitimacy of collaborative, goal-oriented discourses in communities.

I liken this scenario to High-definition Television. By 2010, the majority of programming in North America will move to HDTV specifications, thus effectively removing other TV formats such as plasma TV's from competition. In the meantime, consumers are free to continue using their existing TV sets. The Web could very well employ this model, as it's logical and crosses the path of least damage. Using the HD TV scenario, Web users can continue using their current browsers and existing ways of surfing while those who want to maximize the full potential of the Web will use Semantic Web browsers (e.g. Piggy Bank) that are designed specifically to utilize the portion of the Web that is "Semantic Web-compliant."

Meanwhile, in the background, semantic annotation will be slowly integrated into Web pages, programs, and services. As time progresses, users will eventually catch onto the "rave" that is the Semantic Web . . .

E-Commerce 2.0

Web 2.0 has been quite the hype over the past few years, perhaps too much. Much of it pertains to best practices using blogs, wikis, RSS feeds, and mashups. But not very much has been discussed - well, not enough in my opinion - about practical commercial applications other than the ubiquitous eBay and Amazon. Not anymore. Meet Zopa, the world's first social finance company. In 2005 Zopa pioneered a way for people to lend and borrow directly with each other online as part of our continuing mission to give people around the world the power to help themselves financially at the same time that they help others. According to Kupp and Anderson's Zopa: Web 2.0 Meets Retail Banking, here's how Zopa works:

(1) Zopa looks at the credit scores of people looking to borrow and determines whether they're an A*, A, B, or C-rated borrower. If they're none of the those, then Zopa's not for them

(2) Leners make lending offers such as "I'd like to lend this much to A-rated borrowers for this long and at this time

(3) Borrowers review the rates offered to them and acept the ones they like. If they are dissatisfied with the offered rates on any particular day, they can come back on subsequent days to see if rates have changed

(4) To reduce any risk, Zopa spreads lender capital widely. A lender putting forth, for instance, 500 pounds or more would have his or her money across at least 50 borrowers

(5) Borrowers enter into legally binding contracts with their lenders

(6) Borrowers repay monthly by direct debit. If repayments are defaulted, a collections agency uses the same recovery process that the High Street banks use

Mashups for '09

It's already been two years since my publication of an article on library web mashups. There have been developments, but still no breakthroughs with that killer application that could popularize mashups for the masses. The main challenge with mashups is that they are still a programmer's world. In merging two or more web programs together, web mashups are the next stage of Web 2.0 and are changing the way that the web is being used. There are already several mashup editors that help user create or edit mashups. Yahoo pipes, Google Mashup Editor, Microsoft Popfly, and Mozilla Ubiquity. But they require some programming skills. I believe mashups are the next stage of the web, the Semantic Web. Why? Because mashups open up data, breaking down the information silos.

I've updated my last article with Mashups, Social Software, and Web 2.0: How Remixing Programming Code Has Changed The Web. In taking a look at mashups, I think libraries need to pay attention, as they open up virtual information services to a much larger audience.

When Times Are Tough . . .

I love libraries, everything from the smell of books, to the warmth of staff, the comfy carpets, to the great DVD collections that are all free to borrow with just a library card and nothing more. But we are in tough times lately, and the downfall of the economy has proven just how useful libraries are to society. As the Los Angeles Times has reported, that although retail stores may be quiet these days, but libraries are hopping as people look for ways to save money. The Los Angeles Public Library is “experiencing record use,” said spokesman Peter Persic, with 12% more visitors during fiscal 2008 than the previous year. At the San Francisco Public Library, about 12% more items were checked out in October than a year earlier. The Chicago Public Library system experienced a 35% increase in circulation. The New York Public Library saw 11% more print items checked out (a spokesman said that could be partly explained by extended hours) . . .

And I`ve begun to experience this myself. Patrons are starting to use collections more, and realizing the financial pinch that the economy has given us. Fear not. The library isn`t going anywhere anytime soon.

Mashups for '09

It's almost two years since I first researched on web mashups. I still remember having a working draft of an article I had been doing for the Journal of Canadian Health Libraries on New Year's Eve. (Hey, it was a slow day). Lo and behold, two years later, and there still have only been a handful of articles on mashups. My idol Michelle "The Krafty Librarian" Kraft has written an excellent chapter in Medical Librarian 2.0 which is perhaps the most concise to date.

I've recently written another entry on mashups, Mashups, Social Software, and Web 2.0
How Remixing Programming Code Has Changed The Web. The challenge with mashups is that it's still unfortunately a web programmer's tool. However, the next stage of the Web will be mashups. It's about opening data for others, and breaking down information silos.

11 Ways to the Library of 2012

Don't blink. It's only five years away. Inundated with the day-to-day duties working in a large academic library has sometimes removed me from the "larger" picture of what libraries look like not only to users, but ultimately how libraries will look like in the future. I've written a great deal about the Semantic Web and Web 2.0; but how do they fit libraries: physically and conceptually? Visions: The Academic Library in 2012 offers a meta-glimpse of how libraries might look like in 2012. As you'll notice, some of the features are suspiciously Web 2.0 and Library 2.0? Let's take a look, shall we?

(1) Integrated Library System - the system will recognize the patron and quickly adapt and respond to the patron's new questions and needs (A Semantic Web portal?)

(2) Information Available - collections will undergo dramatic transformations, as they will be largely patron-selected, featuring multi-media resources and databases, many provided collaboriatvely through extensive consortial arrangements with other libraries and information providers (Think longtail?)

(3) Access to Information - print-on-demand schemes will be developed utilizing the dissertation production experience of UMI but providing mechanisms by which the user can return the fresh, undamaged manuscript for credit, and for binding and future use (Kindle?)

(4) Study Space - Space for work and study will be adaptable, with easily reconfigured physical and virtual spaces (Information Commons? Learning commons?)

(5) Information Instruction - Training and learning support, delivered both in person and through appliance-delivered (desktop, hand-held, and small-group), videoconferencing, will characterize all this

(6) Information Printouts - Articles, videos, audios, an on-demand printing of various formats will not only be commonplace, but displays of titles will be coordinated with publishers and booksellers to enhance information currency, to market small-run monographs, and to generate revenues

(7) Organizational Aspects - Library staff will be engaged, networked, matrix-structured, and largely "transparent" unless the patron is standing inside the facility facing the individual

(8) Orientation - Library's perspective will be "global" - ubiquitous automatic translators will facilitate truly global information-accessing programs

(9) Computer Access - From OPACS to wireless access for collapsible laptops and personal appliances

(10) Financial - the viable library will have developed dependable revenue streams to facilitate ongoing innovation and advancement (Library as Bookstore model?)

(11) Consortia - Collaborating to create and publish academic journals and resources, particularly e-journals, e-books, and collections of visual resouces in various media (Open Access?)

Happy Holidays and Seasons Greetings

Seasons Greetings to all. It is indeed a wonderful holidays as the Google Scholar has published an important piece to the Semantic Web literature. He's done it again, writing an concise and cogent piece on the key elements which differentiates Web 3.0 from Web 2.0. In other news, a reader recently made a comment from a previous entry which I found to be very interesting. Here's what he said:

I (as a librarian) found the article and the whole topic very important. I especially enjoyed the conclusion. You wrote that "Web 3.0 is about bringing the miscellaneous back together meaningfully after it's been fragmented into a billion pieces."I was wondering if in your opinion this means that the semantic web may turn a folksonomy into some kind of structured taxonomy. We all know the advantages and disadvantages of a folksonomy. Is it possible for web 3.0 to minimize those disadvantages and maybe even make good use out of them?

My response? It'll sound cliched and tired: it's really too early to tell. But although it's murky as to what the Semantic Web will look like, all directions point to the possibility that folksonomies will play a key role. Here's why:

(1) Underneath the messiness of the Web, is a fairly organized latent structure, whose backbones are web threads. A scale-free network is significantly dominated by few highly connected hubs.

(2) What this means is that folksonomies and tagging are in fact controlled vocabularies in their own right. Lots have been written about this. Recent studies have shown that the frequency distribution of tags in folksonomies tends to stabilize into power-law distributions. When a substantial number of users tag content for a long period of time, stable tags start appearing in the resulting folksonomy.

(3) Such a use of folksonomies could help overcome some of the inherent difficulties in ontology construction, thus potentially bridging Web 2.0 and the Semantic Web. By using folksonomies' collective categorization scheme as an initial knowledge base for constructing ontologies, the ontology author could then use the tagging distribution's most common tags as concepts, relations, or instances. Folksonomies do not a Semantic Web make -- but it's a good start.

Information Science As Web 3.0?

In the early and mid-1950’s, scientists, engineers, librarians, and entrepreneurs started working enthusiastically on the problem and solution defined by Vannevar Bush. There were heated debates about the “best” solution, technique, or system. What ultimately ensued became information retrieval (IR), a major subfield of Information Science.

In his article Information Science, Tefko Saracevic makes a bold prediction:

fame awaits the researcher(s) who devises a formal theoretical work, bolstered by experimental evidence, that connects the two largely separated clusters i.e. connecting basic phenomena (information seeking behaviour) in the retrieval world (information retrieval). A best seller awaits the author that produces an integrative text in information science. Information Science will not become a full-fledged discipline until the two ends are connected successfully.

As Saracevic puts it, IR is one of the most widely spread applications of any information system worldwide. So how come Information Science has yet to produce a Nobel Prize winner?

But the World Wide Web changed everything, particularly IR. Because the Web is a mess, everybody is interested in some form of IR as a solution to fit it. A number of academic-based efforts were initiated to develop mechanisms, such as search engines, “intelligent” agents and crawlers. Some of those were IR scaled, and adapted to the problem; others were a variety of extensions of IR.

Out of all this emerged commercial ventures, such as Yahoo!, whose basic objective was to provide search mechanisms for finding something of relevance for users on demand. Not to mention making lots of money. Disconcertingly, the connection of the information science community is tenuous, and almost non-existent – the flow of knowledge is one sided, from IR research results into proprietary search engines . The reverse contribution to public knowledge is zero. A number of evaluations of these search engines have been undertaken simply by comparing some results between them or comparing their retrieval against some benchmarks.

As I've opined before, LIS will play a prominent role in the next stage of the Web. So who's it gonna be?

The Semantic Solution - A Browser?

In a recent discussion with colleagues about Web 2.0, we ran into the conundrum of what lies beyond Web 2.0 that would solve some of the limitations that it has. I offered the idea of an automated Web browser - a portal - one that would not be unlike an Internet Explorer browser with which a user could just sign in, and enter his or her password, and then freely surf the the Semantic Web (or whatever parts of it exist). It would be an exciting journey. Dennis Quan and David Karger's How to Make a Semantic Web Browser proposes the following:

Semantic Web browser—an end user application that automatically locates metadata and assembles point-and-click interfaces from a combination of relevant information, ontological specifications, and presentation knowledge, all described in RDF and retrieved dynamically from the Semantic Web. With such a tool, naïve users can begin to discover, explore, and utilize Semantic Web data and services. Because data and services are accessed directly through a standalone client and not through a central point of access . . . . new content and services can be consumed as soon as they become available. In this way we take advantage of an important sociological force that encourages the production of new Semantic Web content by remaining faithful to the decentralized nature of the Web

I like this idea of a portal. To have everyone agree about how to implement W3C standards - RDF, SPARQL, OWL - is unrealistic. Not everyone will accept the extra work for no real sustainable incentive. That is perhaps why there is no current real invested interest by companies and private investors to channel funding to Semantic Web research. However, the Semantic Web portal is one method to combat the malaise. In many ways, it resembles the birth of Web 1.0, before Yahoo!'s remarkable directory and search engines. All we need is one Jim Clark and one Marc Andreeson, I guess.

(Maybe a librarian and an information scientist, or two?)

"Web 3.0" AND OR the "Semantic Web"

Although I have worked in health research centres and medical libraries, I have never worked professionally as a librarian in a health setting. That is why I have great admiration for health librarians such as The Google Scholar, who can multitask, working as a top-notch librarian while at the same time keeping up with cutting edge technology. The Google Scholar recently made a wonderful entry about Web 3.0 and the Semantic Web:

In medicine, there is virtually no discussion about web 3.0 (see this PubMed search for web 3.0 (zero results) and most of the discussion on the semantic web (see this PubMed search - ~100 results) is from the perspective of biology/ bioinformatics.

The dichotomy in the literature is both perplexing and unsurprising. On the one hand, semanticists are looking at a new intelligent web has 'added meaning' to documents, and machine interoperability. On the other, web 3.0 advocates use '3.0' to be trendy, hip or to market themselves or their websites. That said, I prefer the web 3.0 label to the semantic web because it follows web 2.0 and suggests continuity.

I find it perplexing, too, that academics tend to subscribe to the term "Semantic Web" whereas practitioners and technology experts tend to refer to "Web 3.0." For example, the Journal of Cataloging and Classification recently had an entire issue devoted to the Semantic Web - without one mention of the term "Web 3.0."

Although the dichotomy in the literature is apparent, it's interesting that for most of us, we associate Web 3.0 and the Semantic Web together. It's not unlike a decade ago when we used the terms "Internet" and "Web" interchangeably -- even though they are not.

Tim Berners-Lee and the W3C envisioned for the Web to eventually progress to becoming the Semantic Web. Standards such as RDF and DAML+OIL emerged as early as 1998, long before Web 2.0. Web 2.0 is not even mentioned in the W3C because it has no standards. In my opinion, Web 3.0 and the Semantic Web are separate entities. Web 3.0 goes one step further in that it will extend beyond the web browser and will not be limited to just the personal computer.

It is important that medical librarians -- all librarians for that matter -- join in (and even lead) the discourse, particularly since the Semantic Web & Web 3.0 will be based heavily on the principles of knowledge and information organization. Whereas Web 1.0 and 2.0 could not distinguish among Acetaminophen, Paracetamol, and Tylenol -- Web 3.0 will.

Google and End of Web 2.0

Google Scholar recently celebrated its third birthday. There were some old friends who showed up at the party (the older brother Google arrived a bit late though) -- but overall, it was a fairly quiet evening atop of Mountain View. So where are we now with Google Scholar? Has the tool lived up to its early hype? What improvements have been made to Scholar in the past year? In a series of fascinating postings, my colleague, The Google Scholar, made some insightful comments, particularly when he argues:

What Google scholar has done is bring scholars and academics onto the web for their work in a way that Google alone did not. This has led to a greater use of social software and the rise of Web 2.0. For all its benefits, Web 2.0 has given us extreme info-glut which, in turn, will make Web 3.0 (and the semantic web) necessary.

I agree. Google Scholar (and Google) are very much Web 2.0 products. As I had elaborated in my previous entry, AJAX (which is Web 2.0-based), produced many remarkable programs such as Gmail and Google Earth.

Was this destiny? Not really. As Yihong Ding proposes, Web 2.0 did not choose Google; rather, it was Google that had decided to follow Web 2.0. If Yahoo had only known about the politics of the Web a little earlier, it might have precluded Google. (But that's for historians to analyze). Yahoo! realized the potential of Web 2.0 too late; it purchased Flickr without really understanding how to fit it into Yahoo!'s Web 1.0 universe.

Back to Dean's point. Google's strength might ultimately lead to its own demise. The PageRank algorithm might have a drawback similar to Yahoo!'s once dominant directory. Just as Yahoo! failed to catch up with the explosion of the Web, Google's PageRank will slowly lose its dominance due to the explosion caused by Web 2.0. With richer semantics, Google might not be willing to drastically alter its algorithm since it is Google's bread-and-butter. So that is why Google and Web 2.0 might be feeling the weight of the future fall too heavily on their shoulders.