[WEB4LIB] Re: NOTAmazon OR Google as a "corporate" role model: iTunes

[K.G. Schneider]

> The ability to browse requires rich metadata that iTunes (and Amazon)
> are exploiting.  Browsing by author is moderately interesting
> (everything by the same author).  Browsing by title is less interesting.
> Browsing by album title is okay, but browsing by track title is not
> useful.

Well--as a browse proponent (one who believes that browse is an excellent
post-coordination activity, which is missing from Google et al., and one who
also believes in rich metadata, or even middle-class metadata)--browsing by
track title can be very useful and interesting, if by track title you don't
mean only the titles on that track but all versions of that title. 

I don't think iTunes or Real are good at browse by title, however. So if you
mean poorly-implemented, I agree. :-) I was going to say that's because they
aren't trying to sell multiple versions of the same song, but then, maybe
they are. The Real music store is better at this than iTunes (compare a
search for "Down to the River," no quotes, which I recently did when I
wondered about available recordings for "Down to the River to Pray"). 

Though then I ask myself, "Self, I wonder which versions are based on the
same music? And are some of the other songs I see in this search actually
the same?" Imagine a FRBRized iTunes... anyway. Google is Search Uber Alles,
which is making them money hand over fist, and I wonder if that means it
isn't asking why iTunes works so well. The biggest peeve I have with Google
Desktop is getting back an unfaceted dump that can't be further manipulated.


Karen