[TagCommons-WG] Use Cases for Sharing Tag Data

Nitin Borwankar nitin at borwankar.com
Wed Feb 21 08:02:15 PST 2007


Thomas Vander Wal wrote:

>
>I would be careful with this statement as it is from an external view to the
>person who placed the tag.  The tag to each person that places it has
>immense meaning, far deeper and broader than can be discerned by others.
>But, looking at things from outside that person whom placed that tag term we are void of that meaning. 
>

Sometimes, but not always true. The intent is important.

One goes to some effort to use tags that have group meaning when one is 
consciously tagging for group consumption.
One uses far less effort when the tag or tag collection is meant 
primarily as a trigger for personal recall and only incidentally or not 
at all for group use.  If we are to characterize all tagging as personal 
then yes, the above are strongly true.


But tagging is at once both personal and social with the proportion of 
each component varying by venue and application.  Tagging in GMail is 
100% personal - your tags are never consumed by others with whom you 
correspond.
But tagging on Flickr is consciously for sharing ( it is a photo-sharing 
site) and tagging on del.icio.us is primarily for personal use but with 
a strong incidental and sometimes deliberate sharing component.  Some 
tags on del.icio.us are obviously personal and sometime quite cryptic, 
some are obviously shared such as 'nptech'.


In general tag use cases can be classsified by

a) producer of tag - individual or system ( some would argue it is not a 
tag unless produced by a human )
b) consumer of tag - self, other individuals, restricted group, public, 
system

So individual producer/self consumer ( the 'immense meaning' use case ) 
is but one of many, in my mind.




> I would agree with a statement along the lines
>of, "consuming tags placed by others, the single tag would seem meaningless,
>but is imbued with meaning by other tags and/or other uses of that tag term
>by the same identity".
> 
>  
>
Yes, this is correct.  But I see tags also as loose descriptors of 
context, independent of whether I know the identity and/or objects in 
advance.  So 'java, sumatra, borneo, indonesia', 'java, sun, j2ee'  and 
'java, sumatra, dark-roast' are clearly different contexts to me already 
although there are elemnst of ambiguity in there.
These delimit loosely, a-priori, the users I might expect to be using 
such combinations and objects I might expect to be tagged with such  
combinations. 

So yes I do find it useful, for certain purposes, to consider tags and 
tag combinations, external to and without reference to the identity and 
the taggee object.

> Tag combinations have a weaker bond across identities, which means finding tag
>combinations in common across identities seems to be less common that a
>single tag used with the same definition by more than one identity across
>items they have tagged in common.
>  
>
Less common yes, but far more interesting and richer in possibilities in 
my mind.

Another aside -
It would be useful if results if your data were available as RSS feeds 
for practioners, such as on those on this list.  I personally would be 
willing to pay a small monthly/annual fee to have access to large 
databases of actual tagging data.

-- 
Nitin Borwankar
Find, Learn, Act .... 
Greener, the search engine for the planet
http://greener.com
nitin at borwankar.com
510-872-7066




More information about the Wg mailing list