Tag Archive for 'folksonomy'

Quick Updates: 1 April 2007

  • Prompted by What del.icio.us needs to do where Bill Ives wrote:

    Putting my blog posts into del.icio.us also allows me to see who else tagged these posts to determine the ones that others found useful enough to tag in del.icio.us.

    I tagged my own blog posts in del.icio.us to discover that the most tagged by others are What To Do On Behalf of Informal Learning? (8 others) and PLE as Retreat versus Productivity Suite (6 others). No surprise regarding the first. A bit of a surprise on the second other than that this post was picked up by the widely read Stephen Downes in OLDaily.

    More interesting to me, as I looked at what others tagged when they tagged me, was noticing how several others use ‘People’ as a del.icio.us bundle, but then how syntax varied in the tags. I consistently tag as first-name underscore last-name (Donald_Duck), others skip the underscore (DonaldDuck), and at least one inverted the order (DuckDonald), which again highlights a challenge of gaining full value from a pure folksonomy approach.

3 April update: Another alternative from Folksonomy – Tagging Lessons Learned:

I’ve been experimenting with the surname:Link format with good results. All of my bookmarking sites (Diigo, MyWeb and del.icio.us) handle it okay. Flickr removes the colon from the tag – both when I add the tag and when I search for that tag – so it still functions nicely.

  • Follow-ups from my first Quick Updates on 18 March:
    1. Tag Cloud Software. I created a text file from the first two months of my writing in this blog and gave TagCrowd (thanks to Chris Fletcher) and IBM’s Many Eyes (thanks to Jay Cross) a try against that dataset. Both gave better results than my misguided choice described on 18 March. Many Eyes is the more robust choice between the two. Both are challenged with the three word phrase “personal learning environment” until I manually changed the input file to use personal_learning_enviroment. More tinkering for me to do some day here with both.
    2. Resource Pages. If you read this via the site (contrast to a feedreader), you’ll see that I’ve published my IA (Information Architecture) resource page. Still pretty sketchy.

Deeper Dive Regarding Tagging and Such

After my two posts yesterday, I continue to think about tagging and software for tagging. Here is a recap of both what is on my mind and what a tiny bit more research has revealed for me:

  1. Freebase is in a different league than Fuzzzy relative to at least venture capital backing versus student project. From Fuzzzy’s About page…first, the cause for pause:

    The system is a research project and some parts of the system may be very experimental. We are continuously developing this system. If the project is terminated we will inform you some time before the service is shut down. In that case, export mechanisms will be developed so that you can move your data somewhere else.

    Next the cause for excitement:

    Fuzzzy is not only social but also semantic which means that the Tags used have more meaning. When bookmarks are assigned a meaning using a standard like the ISO 13250 Topic Map then people as well as other computer systems can make use of the embedded knowledge in a more meaningful way. This way of categorising content is a middle way between the top-down monolithic taxonomy approach like the Yahoo directory and the more recent social tagging (folksonomy) approaches.

  2. I need to dig more into Topic Maps. I was only first introduced a few weeks back via this wild Alexander Johannesen – Sexier Smarter IA through Topic Maps podcast.
  3. Indeed, I want portability for my tagging data. I know I can export my de.icio.us to bookmarks. But, can I change to a different tagging software (for example Furl, Ma.gnolia, Fuzzzy, etc.) and take my data with me from del.icio.us? Harold Jarche wrote in December about his experience from going from Furl to del.icio.us.
  4. Deep Sea DiverPrompted by this start of a deeper-dive, today I finally got around to creating accounts on Furl and Ma.gnolia and started to experiment to get a more hands-on comparison.
  5. Add to the list FaceTag that I discovered today thanks to washtublibrarian. I am certainly drawn to their tagline, which echoes what the Fuzzzy About page said: “Integrating bottom-up and top-down classification in a social tagging system” as this is exactly what I am reaching for when I first was pointed to that made-up C-word I ranted on yesterday.
  6. For some unknown reason in my adoption of del.icio.us for the last two months, I’ve been resisting tagging lower level pages such as individual blog posts — reserving this only for pages/posts that I really really wanted to be certain to find myself back to; for example, a ‘how to’ on a very narrow topic. I can’t explain my resistance rationally. I don’t know where I came up with the notion that I should do most my tagging for home pages. My thinking is now shifting and I am much more likely to tag a deep-link and I believe that is where the most potential value is. Okay, sometimes I’m just slow to get it.
  7. Either connected to the previous, or alternative to? I want my general tagging software to be integrated with my feedreader. I am annoyed that I am managing my feedreader content through folder structure (yuck) and search only. Here I am reminded of what John Tropea described on his own wishlist over at Library clips.
  8. de.icio.us does admit to performance issues in overdue new year’s resolutions.
  9. Still to dig into, but looks very informative for this dive is Steve Eisner’s five post series on “Enterprise del.icio.us.” I also still need to get up to speed with My Yahoo!
  10. Lastly, I’m feeling a bit foolish about yesterday being so quick to point to David Weinberger’s post versus doing my own research first. Alas, at the end of what felt like a very long work week, I fell into the lazy blogger behavior of pointing to somebody who is pointing to somebody who is pointing to somebody. Hopefully I have received at least some redemption with the above evidence of starting to do my real homework.

Photo Credit: Sean Bellew

David Weinberger blogs Fuzzzy and Freebase

I’m having another of parallel universe feeling. Unknown to me as I was writing my earlier post today regarding some desired evolution of tagging and collaborative directory approaches, David Wienberger also today blogged Gather ye metadata while ye can – via Fuzzzy and Freebase highlighting two new entries to the crowded tagging field.

Regarding Fuzzzy, David says:

[Fuzzzy] lets you tag bookmarks and maintain a social network. The big words come in because Fuzzzy lets you position a tag in an ontology…I’m interested to see how this experiment works out. There’s no question that the metadata it collects — in addition to classifying the resource according to a taxonomy, the site lets you check some boxes to indicate the resource’s “mood,” knowledge type, and details level — would be useful, but experience teaches us — until it confounds all teachings — that people generally resist attaching explicit metadata.

From the Fuzzzy registration page:

You will be able to vote on content, find other users with similar interests, and get personalised views and much more.

Sounds good and fills in some of the missing from del.icio.us for me that I mentioned earlier today.

As I start to use Fuzzzy I’m not so sure about their Mood (Fact, Fun, or Compassion), Knowledge Type (Why and If, How, What Where Who and When), and Details Level (Overview, Detailed) classification schema. It will be interesting how will it works for me over time. Compared to del.icio.us, I do prefer their user interface.

Regarding Freebase, David says:

Freebase will be fascinating to watch. If we do in fact build it, we’ll have a publicly accessible (Creative Commons licensed) ontology populated with tons of stuff we care about that will do much of what the Semantic Web is trying to do: Draw implicit connections, discover context, search better, and just in general be smarter users of a smarter Web.

I submitted my email address to get on the list for an invitation. Stay tuned.

Tagging and Taxonomy Exploration (Part-2)

I am now circling back to Vignette #1 from my 25 February post to compare and contrast the LearningLinks approach to the alternative of using del.icio.us or another social bookmarking application.

  • Both approaches allow for discovery of content beyond one’s own within a defined subject area.
  • LearningLinks supports content ratings, which provides at least an indicator of potential value. del.icio.us displays the number of other people that have bookmarked the same page — another, although perhaps weaker, proxy for value.
  • In social bookmarking I have full control, including creating my own groupings to build hierarchical structure (what del.iciou.us calls ‘bundles’.) In the LearningLinks approach I can influence the hierarchical structure by providing feedback, but I am unable to directly modify the structure to my own liking.

My somewhat obvious conclusions at this point:

  • Hierarchical structures add value (again, I’m no fan of tag clouds or flat lists for navigation or browsing) and either example above can accomplish that
  • Access to what others have thought as valuable (either discovery via common tags or more directly in LearningLinks) is useful and an improvement over plain search, although even internet search has an element of this with the “Google” algorithm using “number links to this” as a determining factor.
  • Collaboration (which I define as shared work on a common work product, and therefore applicable to LearningLinks, but not to del.icio.us) also has the potential to increase value, but slows the creation and adoption
  • Having full control of your own data both increases immediate personal value and speeds adoption

My discussion thus far then takes me to the latest made-up word in the tagging and taxonomy world, which seems to have just sneaked into the learning community this week: collabularies. Here I am with Stephen Downes who on Wednesday wrote:

I would not go so far as to use a word like ‘collabulary’ – that is a ridiculous word, and is not needed to describe something that we already have perfectly good words for, a ‘taxonomy’ or a ‘vocabulary’.

[19 October 2008 Update: removed a "side-bar rant" as contained an incorrect link]

Earlier in his post Stephen nicely makes the distinction of working independently and working collaboratively, even if collaboratively within social bookmarking. Back to the examples above, the del.ico.us ‘elearning’ tag is working independently and thus “folksonomy” as it tends to be defined and the LearningLinks example is closer to “collaborative vocabulary” although really more “collaborative directory.”

Sidebar comment: I notice that LearningLinks doesn’t have a top-level category for e-learning, but does have top categories for Mobile Learning and for Gaming & Simulation. Interesting choice. And, when I search on elearning I get 38 results, but have no way of telling where in the directory these are classified…thus, giving away value from the directory structure.

I’m reaching for something that is a further middle ground:

  • What if LearningLinks was more wiki-like? Even if not allowing for direct rearrangement of the directory by all, at least providing a transparent (visible to all) back-channel as most wikis do with a ‘talk’ or ‘discuss’ page. This would allow for more explicit collaboration among all those interested in participating.
  • What if del.icio.us supported easily creating and sharing a full-blown directory in a presentation form closer to LearningLinks? (perhaps they already do if I get at the API level of things?)
  • What if LearningLinks also addressed vocabulary? For example, providing synonyms for the major directory listings.

Or, coming more “top-down” into either del.icio.us or LearningLinks with an existing authoritative taxonomy — if such exists in the learning area (note to self: dig into this more, the only thing even close to this that I am aware is the Learning Circuits glossary) — to guide the vocabulary and structure to more the ‘standard’. This thought then takes me to this Taxonomy directed folksonomies blog post from December, which a colleague at Novell passed along:

I’ve been exploring how to combine [taxonomy and folksonomy] into something I’ve named a taxonomy directed folksonomy (TDF). Basically we use a taxonomy (in this case ScOT) to suggest terms a user could in a tagging-type interface. I’s our hope that this interface will allow us to use the formally organized taxonomy at the back end to find related resources, while not intimidating users on the front end.

Closing rant: del.ico.us is unacceptably slow. On my shared wireless connection at the moment, a del.ico.us search on ‘elearning’ took an insanely long 30 seconds to return results. By comparison, Google and Yahoo search for ‘elearning’ both returned in sub-two seconds. This alone is about to drive me off of using del.ico.us, or at least look into alternatives. Worse, I can’t even get del.ico.us to return results on a search for ‘e-learning’, it must be getting tripped up by the hyphen? But, meanwhile, it does list ‘e-learning’ as a “related” term for elearning.

Folksonomy and Tagging Introduction from Thomas Vander Wal

Thomas Vander Wal is credited with coining the ‘folksonomy’ term. This one hour “podcast,” recorded at the OZ-IA conference in September 2006, is one of the best introductions to folksonomy and tagging that I have ever heard or read: Folksonomy to Improve IA.

Kicking Off My Tagging and Taxonomy Exploration

Old paper luggage tag This is drill-down #2 on What To Do On Behalf of Informal Learning. This time I’m going after item #5 in the list: Improve content findability. It is both with considerable excitement and trepidation that I start down this path as I know there is several months worth of work and 20 or more posts for me to begin to work through what I either halfway know or I am anxious to learn in this topic area.

Alpabetical File FoldersI’m going to start with a series of posts related to tagging, folksonomy, taxonomy, web directories and the like. For this first post I share three vignettes that had me thinking the most in this area over the weekend:

Vignette #1: Elliott Masie launched one of his many interesting experiments as part of the Learning2006 conference: a project to create a link directory of the body of knowledge associated with the professional discipline of Learning. The LearningLinks site concept is that anyone can contribute a link related to the subject of learning (i.e. a Learning Link) and also guide where this link fits within a hierarchical navigation scheme (i.e. directory). An editor on Elliott’s staff then does the actual mechanics of posting. After the link is added readers can rate the link as in the 5-star system popularized by amazon.com. “Learning Links is based on mutual respect and constructive collaboration.”

Vignette #2: When tags work and when they don’t: Amazon and LibraryThing:

Both LibraryThing and Amazon allow users to tag books. But with a tiny fraction of Amazon’s traffic, LibraryThing appears to have accumulated *ten times* as many book tags as Amazons’s 13 million tags on LibraryThing to about 1.3 million on Amazon.

Something is going on here, something with broad implications for tagging, classification and “Web 2.0″ commerce. There are a couple of lessons, but the most important is this: Tagging works well when people tag “their” stuff, but it fails when they’re asked to do it to “someone else’s” stuff. You can’t get your customers to organize your products, unless you give them a very good incentive. We all make our beds, but nobody volunteers to fluff pillows at the local Sheraton.

This is a great post with over 40 comments so far.

Vignette #3: Graham Attwell’s Social networks based on commonality of interest:

I am increasingly interested in how we can use social bookmarks or tagging as a form of developing social networks based on content. We all know that social networking is a powerful tool for informal learning. But the friend of a friend type approach assumes a commonality of interest which does not always exist!

Tagging has the potential to generate similar social networks – based not on just friendship – however many steps renewed – but based on commonality of interest.

Photo Credits: Zumberto (tag) and Colin Maykish (folders)