Fuzzzy 2.0 was launched October 1.st. Among the many new features is global tagging support.
Global tagging enables everyone around the world to participate on building tags that are connected. Instead of adding dumb text categories to your Flickr photos, del.icio.us bookmarks or your weblog posts, you can now use a shared pool of tags. By moving away from simple folksonomy tags that are only valid within the single site you can move on to semantic tags that bring with them connectivity and meaning. This increases knowledge sharing by having a shared vocabulary. It also increases findability, learning and consistency.
Try it out for your self:
A. Distributed tagging
If you open the tag "Topic Maps" http://www.fuzzzy.com/tag/?id=1607
you can click the icon that looks like three red connected nodes. Then some dummy related tags from the tag network will be loaded. At this moment there are just two tag servers.
B. Tag server functionality.
On this dummy blog http://om.niscio.us/cowboyroy/ you can tag the blog post with tags from the Fuzzzy tag server.
How was it made possible?
The simple version for those of you who don’t use most of your spare time reading geeky stuff: Global tagging is made possible by the organic tag-set of Fuzzzy which is built upon the Topic Map ISO standard and an underlying infrastructure with web services. Members of the online community can create tags and relations between them. Users can also vote, comment and in a democratic way participate to build a shared tag vocabulary.
Now the geek version: By using Topic Maps for semantic interoperability and the built in identification model with Public Subject Identifiers (PSI’s) tags becomes universally identifiable. The globally connected tags become a shared vocabulary for both people and computers. By using web services with simplified XTM fragments any peer tagging site can request from all other peers related tags for tags that are found throughout the network. If in two different domains the same subject is mentioned then these domains are be connected when the subject has a published identifier. When a user makes a request for related tags at the global level a request goes out to all peers. The list of remote sites are replicated and stored on each peer. A webservice call such as this one http://www.fuzzzy.com/ws/GetTagByPSI/?psi=http://www.fuzzzy.com/tag/3053 is issued and all related tags are returned as XTM. The associated tags at the remote site can then be viewed or imported.
Roy, the web service call actually does not return XTM. It returns some non-standard markup that's probably NetworkedPlanet-specific. That's a shame, as nobody else can read it. XTM would be a real improvement.
ReplyDeleteValid XTM will be supported later along with the currently used simplified XTM topic map fragment format by Networked planet. I'am planning other light weight formats like Atom, RSS and also looking into using more heavy formats like OWL.
ReplyDeleteRoy, a very interesting experiment. I'll certainly be following it to see where this goes. The question of how you close the gap between taxonomies and folksonomies is a fascinating one.
ReplyDeleteI come at this from an enterprise2.0 perspective and as such would be interested in your opinion about scalability or rather de-scalability. On the web the user base is potentially huge, millions if not billions, and you only need a small uptake to obtain the benefits of network effect. This means you potentially get many people voting on the relatedness of each tag and as a consequence, with time, get a nice smooth distribution. Inside a company even a large one, we have ~80,000 employees, the potential user base is significantly restrained. I would suspect in this case we would get far fewer contributors and so the system will not move beyond the immature state and would have a spiky relatedness between tags. If this was the case would you expect to see a significant advantage over a 'classical' del.icio.us type social bookmarking service?