[uf-discuss] Proposal: wine

[James Jory]

> Correct me if I'm wrong, but isn't everything you both 
> mentioned covered by the much more generic proposal for hListing [1]?

I think you misunderstood at least where I'm coming from. I jumped in on
this thread since it was discussing wine but I was not intending to propose
a wine microformat (as the original subject indicates). Anyway, what I was
asking about was the best way to begin the discussion for how wine
information can be represented using (existing) microformats. Yes, hListing
can be used for listing wines for sale/trade, hReview is already being used
for wine reviews, and I'd like to see wineries start using hCard, geo &
hCalendar on their sites. Part of the challenge in the wine industry will be
simply evangelizing the use of microformats and illustrating their benefits.

However, as an industry (wine, that is), we have separate issues that we
need to address. Perhaps most importantly is how to uniquely represent a
wine. Very few producers use UPCs so we have to rely on the wine name in
most cases. And with wines throughout the world often bound by local naming
a labeling rules which are inconsistent, the problem is not an easy one to
solve. Much of this is not specific to microformats but if we want to
accurately represent wines using microformats the two must come together
somehow. I do not see  anything in hListing or hReview that currently
addresses this. If I missed it, please let me know. As an aggregator of wine
information, this issue is very important to me.

As a side question, at what point is the use of additional class names
within an existing microformat considered a "new" microformat? For instance,
if within the "item" for an hReview additional class names were specified
(say, "vintage" and "producer"), is this considered a mis-use of hReview?
The semantics of the hReview are still intact but now there is additional
definition that may help solve what I described above.

> If we were to go down the route of a wine µF, the next step 
> would be beer, but then the real-ale types would want a 
> real-ale one, and the lager people would want a lager one, 
> and then there's vodka, whisk(e)y, schnapps, and I'm not even 
> out of the alcomahols yet.
> 
> We'd then have eleventy-billion different µFs, one for each 
> and every possible type of product one person in the world is 
> interested in, and then everything on the intertubes would be 
> marked up differently and we may as well have just switched 
> to XML and killed off all hopes for a common base for a 
> semantic web here and now.

Jeez, my head hurts. Is this the way you welcome everyone to this list?