Introduction: Social Tagging and Bookmarking Tools
Lomax is an Assistant Professor at Georgia Tech in the Learning Technologies Division.
Social tagging is new approach to organizing content. It’s about sense-making. Getting people to what they’re looking for.
History. Started with link directories. The hotlists. Bookmarks moved into browser. Then, through Yahoo!, etc., shared technologies for organizing taxonomies.
Features of social bookmarking; organize & categorize content, share information sources, collaborate through folksonomies. Search and browse by topic, tag, or concept. Retrieve through multiple access points. Ubiquitous access through Internet and Web.
“Synonimity” — Lomax’s term for the following:collaborative tagging, user tagging, social bookmarking, social classification, distribute declassification, folksonomy, ethnoclassification.
Why tag? simple, straightforward, decentralized (anyone can tag), reflects individual’s concept of the object — important to learning; allows for disparate opinions (this is a two-edged sword).
Disadvantages: inconsistent, idiosyncratic, biased (personal reflections, political, social, etc.); subject to multiple interpretations (we all don’t agree on what a particular word means); assumes inherent need for everything to organized (quotes Peterson). Some stuff is equivalent of “junk mail” — straight to bin without being used.
Social tagging in academia: CiteULike, Connotea, Librarything, Bibsonomy.
Social Tagging in K-12 Education
What’s important about tagging in K-12 education? Kids have their own language — tagging suits them. Fluid colloquial language. Tagging helps students describe technology in their own words. (Cites Seldow)
Tagging can serve as an adjunct to current teaching practice and provides instructors and students with useful ways to store and retrieve info
Tagging changes “the web” to “my web”.
Myedna — a proposed model for combining folksonomy with taxonomy in educational sector. Australian model. Aims to foster “sharing learning”. A personal learning space. Creates online learning space designed to meet learner’s needs. Allows sharing of information, resources, stories and narrative with other audiences.
Myedna generates a thesaurus based on tags: information about tags, new terms for existing concepts, new concepts and suggested terms. Also information about those doing the tagging. Users can tag, create tags, comment on tags, and comment on resources.
Issues in social tagging: controlled vocabularies don’t reflect user’s actual language, often rigid, often outdated. English evolves very quickly. User-generated tags are more current, richer, multilingual. Tagging provides feedback, indirectly, to existing controlled vocabulary tools.
Cataloging is not opinion. Cataloging and folksonomy exist in parallel — but interact. Folksonomy supplements cataloging, doesn’t replace it. Coexists with cataloging.
Question about authority control in tagging. This strikes me as silly; that’s the point of tagging, after all.
Question about ‘folksonomy-directed taxonomy’. Use folksonomy to clarify jargon; jargon is “official” term, but common language version is what users see. Over time, folksonomy terms enter taxonomy.
Social Tagging and Newspapers
How does social tagging work. How can I conduct research, collect data?
Newspapers and tagging tools in two categories:
- Digg, Reddit, and Newsvine: Users submit things to the service — vote for (and comment on) items. Collect reactions.
- Del.icio.us, MyGoogle, MyYahoo: Users assign indexing terms to articles.
Different papers have different mechanisms for enabling content tagging/voting. NYTimes: register, lets you save to other sites, or save within the site (MyTimes) with broad categories. LA Times: no sign-up, save to MyLATimes with extensive categories, and post to other tools. Washington Post: sign up for account, only save to external tools — no internal collection.
Tracking data in this world is complex: who has data? Publisher? Tagging site? Someone else? Functions are different between different tagging tools. Terminology, terminology, etc. Newspapers’ coverage — newspaper has different access rules than del.icio.us — newspaper might allow access for one week, one month, etc. Tagging sites keep things forever (more or less). How useful is a dead link in tagging tool?
Newspapers internal organization often prohibits external understanding: editors in charge of online version often separate from IT, web publishing staff. Role of online version not necessarily to compete with print. Hard to understand landscape — and it’s not consistent from paper to paper.
We know very little about who is using these social tagging tools provided by newspapers. It’s possible newspapers know something, or social tagging tools, but not clear that either has any information.
Q: Any instances of “tag-bombing” — where a group uses “wrong” tags to hide a web page? Not really, and hard to do.
Tagging and Libraries and Museums
History of tagging in libraries and museums
There are a lot of terms surrounding tagging — bookmarking, tagging, folksonomy, democratic indexing, etc. Symptomatic of that process of building social tagging itself!
Reflects infancy of this tool — around since early 1990s, really popular since early 2000s. Brief (and rough) timeline:
- user-defined descriptors in image retrieval/indexing studies
- del.icio.us, furl, LibraryThing, etc., allow end-user tagging of web sites
- flickr appears
- social bookmarking takes off
- steve.museum.org tagging project. Current research focuses on ways informal tagging has evolved — how do people tag things, and how does that evolve? How can folksonomies inform existing vocabularies? Steve.museum.org — find out how users would represent objects, not how “experts” do. Matching users’ input with formalized cataloging.
- vendors start integrating tagging, faceted classification, etc., into LIS.
- Pew Internet Research study: 28% of online Americans are tagging; 7% are tagging daily.
- SOPAC at Ann Arbor District Library — tagging implemented in library catalog.
- LibraryThing for Libraries goes online at Danbury. Now about 100 libraries using this system.
- Primo, Encore, etc., adding tagging components
- Next big things: tagmashing (in LibraryThing), DIY, Open source packages, etc.
How to do studies in tagging? Hard to find data. Problem is surplus, not paucity: LibJunction lists 7700 libraries in 146 countries doing social tagging. There are a lot of libraries! librarytechnology.org/libwebcats: list of who uses what catalog. It’s huge. Broken down by state, geographic region, etc.
Access to OPAC tagging features often restricted to authorized user — hard to see what’s there from the outside. Tagging features often not prominent (reflecting their “beta” status?) and ill-documented in many cases. Workflow unclear — how to know what happens after a user enters a tag. Few mentions of tagging in libraries and museums in literature.
Given these challenges… Used a random sample from lists mentioned above (every 10th item from WebJunction list). Public, academic, and museum sites. Not K-12 schools. Abbas intends to create a wiki for this. Had 700 different sites that she reviewed. About 8% of sample was using tagging.
Results by kind of institution
- Live tagging Anything is taggable. Could see others’ tags. Some users could *delete* other tags. Search by tags, or not. Some appeared as keywords, others they didn’t. Tags may or may not appear instantly; some were queued for review (or some other system process). Only latin-alphabet tags (in most cases) were allowed.
- Suggestion boxes: Help others find this object. Users could submit tags — but no explanation of how tags were used by system. No tag searching. Things are vetted, apparently, for later inclusion. Logistical problem for libraries/museums — if we have these tags, how do we use them? Many major museums in sample have no tagging at all — yet. But clear from literature that they’re thinking about it.
Live tagging Lots of this in OPAC, but not much on the web site. No searching by tags. Can choose from tag cloud(s) of various kinds (AADL). Users can review books. Also “users who checked this out also checked out…” LibraryThing — lots of public library implementations. As of 20 October, LibraryThing says about 100 libraries participating. Tags are not part of OPAC.
LibraryThing just started “CommonKnowledge” — communal taxonomy developed by users.
Many libraries using del.icio.us — most popular so far. Using it to organize online resources, reader’s advisory lists.
Only two have live tagging: PennTags, Midlands Technical College (Columbia, South Carolina). Users can tag, search by tags, but tags don’t immediately appear. All of these are restricted to authorized users. Midlands is open to everyone, but not much there yet (it’s new).
Del.icio.us — used for pathfinders, organize links to internet resources. Some libraries have bookmarked thousands of items, others only a few. Connotea and CiteULike seem to be more common in literature, but not in Abbas’s sample.
Closing thoughts: Technology can’t get in the way of finding information. At the same time, we need to be cautious that tools don’t obscure research. Also, need to make sure libraries work the way Millenials think — they are the community, to whom we need to react.
Q: What to do about “inappropriate” tags?
A: Lots of approaches; white-, gray-, and black-listing; moderation; other users can report.
Q: What is a tag (word, phrase, etc.)
A: Its linguistic/cultural; in U.S., LCSH is learned; other countries (Asia), there is not such a tool, commonly — phrases are more important.
Discussion of whether tags are part of catalog record, or independent — contentious.
Q: Privacy and ethics: What happens to traditional library privacy values when users are exposing themselves?
A: Challenging. And unknown. Rules may be different in public, academic, and K-12 spaces.
Q: What about credibility and authority?
A: Authority control will always be there