Wednesday, May 31, 2006
The Data Management/Discovery Connection
Jim Rapoza of eWeek has a great article on how desktop search is not as effective as his metadata-enhanced MP3 Collection. To quote, "Because of the way that music ripping and creation encourages proper tagging, it's very easy for me to find and sort music based on, for example, artist and musical style."
Thus, my favorite example of metadata (music metadata) strikes again and continues to make headway into the public consciousness...
Thus, my favorite example of metadata (music metadata) strikes again and continues to make headway into the public consciousness...
Comments:
<< Home
Musical metadata is interesting, but it isn't a good basis of comparison for anyone else's metadata challenge anywhere.
The reason is that music metadata is very insulated. With services like CDDB (and similar clones), it's possible to have a complete, accurate repository of all metadata (or at least a substantial percentage). The "tagging" that happens when you rip files is actually just copying from CDDB, or copying known values from the back of the CD cover. There's no judgment call about how others might search for the information (the need to come up with keywords, descriptions, etc) that pretty much all other metadata challenges need.
It's also static. "Born in the USA" by Bruce Springsteen is always going to be called that. Its track number relative to the album isn't going to change, and while there may be separate assets that are re-recordings of the song by other artists, its genre isn't going anywhere either. That's nothing like enterprise data assets that may change many times per day.
ID3 tagging in MP3s provides a notional example of how tagging works and what its value might be, but it isn't a very good analog for how things work in real enterprises.
The reason is that music metadata is very insulated. With services like CDDB (and similar clones), it's possible to have a complete, accurate repository of all metadata (or at least a substantial percentage). The "tagging" that happens when you rip files is actually just copying from CDDB, or copying known values from the back of the CD cover. There's no judgment call about how others might search for the information (the need to come up with keywords, descriptions, etc) that pretty much all other metadata challenges need.
It's also static. "Born in the USA" by Bruce Springsteen is always going to be called that. Its track number relative to the album isn't going to change, and while there may be separate assets that are re-recordings of the song by other artists, its genre isn't going anywhere either. That's nothing like enterprise data assets that may change many times per day.
ID3 tagging in MP3s provides a notional example of how tagging works and what its value might be, but it isn't a very good analog for how things work in real enterprises.
# posted by 
: 8:56 AM
: 8:56 AM
While I agree that there are certainly differences between music metadata and enterprise data assets, I disagree that music metadata is insulated or static.
Instead of insulated, I would consider certain aspects of metadata both mature and maintained. Those databases like CDDB and MusicBrainz sprang up to fill a need because people care passionately about their music and want to get it right. I assert that this passion for music metadata is the distinction and if such passion existed for enterprise data assets the same quality of metadata would exist.
As far as being static, certainly music genre is not static. Additionally, there are constantly new metadata fields being added like composer, instruments, and tempo.
So, If you have examples of enterprise metadata assets that are not at all analogous to music metadata, I would love to hear about them.
Regards,
- Mike
Post a Comment
Instead of insulated, I would consider certain aspects of metadata both mature and maintained. Those databases like CDDB and MusicBrainz sprang up to fill a need because people care passionately about their music and want to get it right. I assert that this passion for music metadata is the distinction and if such passion existed for enterprise data assets the same quality of metadata would exist.
As far as being static, certainly music genre is not static. Additionally, there are constantly new metadata fields being added like composer, instruments, and tempo.
So, If you have examples of enterprise metadata assets that are not at all analogous to music metadata, I would love to hear about them.
Regards,
- Mike
<< Home
