Every Wine Tasting Note Site Should Be Freebasing!
Thats right. I said it. Although it may not be what you’re thinking.
Can you guess what’s wrong with tasting note web sites? Exactly, none (and I mean N-O-N-E) have reached anything close to critical mass of users to make their notes useful. Why is that? Well there are too many wines every year to have multiple reviews per wine. So every wine tasting note site tries to get their hands around an unbounded number of wines and create a tasting note site that is actually useful. NONE have succeeded and even the biggest are only useful for organization purposes (CellarTracker) not for looking up wines.
One approach with promise is Snooth, but they’re actually smarter about it. Its not about tasting notes, its more about personalizing wine selections for you and if there are tasting notes to help then great. I actually like that concept. They’ll bring in a gambit of ratings and notes and attempt to normalize them and match a wine to your liking. This is (obviously) not a tasting note record keeping site but it leverages that function.
So what’s this about “Freebasing”? Well, if you haven’t heard, there is another approach to gathering data out there and they’re gaining steam. Freebase is a massive database that is completely open so that a site can use as its database as a backend. Then anyone can query this DB and get at that information or submit information and contribute to the collective. Also, tags in that information make connection automatically regardless of the original source. The best explanation of this is here, at Tim O’Rielly’s blog (the guy who originally coined “web 2.0”). Its an instance of the semantic web (what some call “web 3.0”). The advantage? Since a tasting notes are not a business but a feature, if all the sites created real business plans with tasting note functions as a part then there wouldn’t be a need to hide the notes in an isolated database. Sure, protect your user DB but submit your notes to Freebase. Gary V can go on ranting and raving with the Vayniacs, Snooth can continue making educated selections for you, WineQ could add value to their custom wine clubs. These are all sites that don’t depend on notes as the core of the business. One thing I won’t get into is this aspect (and the power of Freebase) - if Winehiker were to create an application that was a database of trails in California and some wines he experienced there, then Freebase would automagically create a query result for any other application that connects wines related to the notes Winehiker made about his travels and the wines on each of those trails with other wine notes submitted from these sites. You would start to see a world evolving of things connected to wines and trips and tastes that you’ve never imagined before…but thats a whole different post!
Anyway, Freebase allows sites like these tasting note sites to be built and while they individually create communities for whatever purpose they are all adding to Borg collective known as Freebase.
There is one other approach – creating a micro-format that makes a standard format that allows any note written out there be crawled and scanned into a DB automatically…what-ever. Thats never going to happen unless Microsoft, Apple, and every other user interface company decides they want to support MicroFormat for wine tasting notes. Chances of that happening? Pretty much Zilch…
It would be far easier for other sites that have note functionality to migrate their DB to Freebase, effectively merging all note DBs, and write database calls to the Freebase API rather than their own MySQL “Silo” of information. You think CellarTracker is cool? Imagine every note ever entered into a site on the Internet, regardless of the site, being available to Snooth or WineQ or any other site that wants it!! I’m an Alpha member of Freebase and I can attest that its difficult to explain the potential impact of this site, which brings me to the practical, marketing side of my brain – I’ve seen too many technologies that were just too far ahead and couldn’t survive until the world caught up. I hope Freebase doesn’t go that route…
Every wine note site in the world should be Freebasing!
Enjoy the Wine Life!
Popularity: 6% [?]
Sphere: Related Content
Very interesting discussion on the impact on wine to one's life. For more information about wines, visit http://vino.com 
Joel,
Thank you for the great description of company growth and your objective analysis of CEO transition. I think one ...
Awesome goal and totally attainable. 
I love commenting different articles and information. To be honest with you, it’s my hobby. And, by the way commenting ...
For me it, when making a purchase decision, it is a constant battle to find objective information to help in ...
I didn't even know there was a wine 2.0 group on Facebook. I was too busy throwing sheep at and ...
Rob,
I hope you don't mind my post here as you indicated you'd like to know about new businesses. I have ...
I reviewed this wine last year: http://www.redwinebuzz.com/04_barcel_vintint_cal.htm - when I still used the 100 point scale.
Michael Barrteo does ...
nice work, guy 
Comment by Dr. Debs on 19 January 2008:
Interesting new program, Joel. But I don’t understand how CellarTracker fails to let you look up a wine. It does. And it has more notes than anywhere else. And it aggregates scores. I for one think that CellarTracker is phenomenally useful–and I’m clearly not alone, given the sheer number of users and bottles of wine entered.
I think what we need is for the dust to settle, a few clear leaders to emerge, and then we will get greater consolidation.
Comment by Joel on 19 January 2008:
I’ve seen it hit or miss myself. Others have told me the same. The sheer number definitely helps but I’m not sure its critical mass, it is, however, the closest thing out there. In fact, I’d say they are the clear leaders in that tasting note, with the limitations.
If they DBs were shared on Freebase, then not only would a query that is looking up wine simultaneously check everyone’s DB (of course some standards would have to apply to the entry, maybe CellarTracker-type could force a standard entry by sheer market power)and give you what you need (i.e. better chance of a hit). Then other programs being able to use this open information for “intelligent” correlation.
Having all the databases closed on a bunch of small to medium sites is what is holding back online notes from being as useful as possible.
If CellarTracker is depending on the notes in the DB forever they could be in trouble in the not too distant future.
Comment by Dr. Debs on 19 January 2008:
Again, very interesting. I guess I’ll have to wait and see how the Freebase grows. Already other ratings sites have managed to mine CellarTracker for my rating there and then use them on their site. What’s interesting to me is that CellarTracker doesn’t do this. I’m not sure if this will make them dinosaurs or keep them at #1. During the past two years, I also have only ever added wine data to create a new wine in about half a dozen instances. Every other wine I’ve had has been in the database–maybe no the latest vintage when it’s just been released, but still. And the other thing I notice is that when you Google a wine’s name, CellarTracker, Snooth, and Cork’d are always on the first page, if he wine is in their database. So Freebase is going to have to really push to get folks to see their way is a better way than a simple Google search. Just my .02 on what are a very interesting set of developments. Thanks for bringing them to our attention.
Comment by Kirrily Robert on 22 January 2008:
Hey, thanks for the great writeup of Freebase!
Do you know that someone’s already working on a wine browser using freebase using the Freebase API?
K.
Pingback by The Freebase Blog » Blog Archive » Freebase is smooth and fruity with an elegant bouquet on 22 January 2008:
[…] interesting article from Wine Life Today: Every wine tasting note site should be freebasing. Freebase allows sites like these tasting note sites to be built and while they individually create […]
Comment by Jeff on 22 January 2008:
In my experience, Snooth is useful only if you know ABSOLUTELY NOTHING about wine. If you are aware of the difference between Chablis and Sancerre, Snooth is not much help at all… Plus, the intereface is super clunky (it doesn’t even list Burgundy as a default “region”) and it seemingly adds search terms of its own volition. Cellartracker is clearly far superior.
jb
Comment by Joel on 22 January 2008:
Jeff,
I think Snooth is shooting for the less knowledgeable. I found the same thing as you but given I don’t think I was the target, I wasn’t holding it against them. I could be wrong though.
Kirrily,
Thanks for the link. I’ll definitely check it out!
Comment by Drazen on 1 February 2008:
A few people on this blog and others have suggested that Snooth “mines” data and reviews from wine.com, cellartracker.com and others. Unless there was explicit permission, isn’t this illegal?!
Sure factual data on wine such as its varietal composition is probably not copyrightable, but reviews certainly are! Also, even accessing purely factual information is protected by “trespass of chattel”. Both EBay and ticketmaster successfully sued “aggregators” that were mining their sites.
Am I missing something here… or did Snooth not consider this, and the lawyers for wine.com just haven’t pounced yet?
Comment by Joel on 1 February 2008:
I’m not entirely sure, thats a good question. I know Phillip from Snooth occasionally stops by. Hopefully he’ll comment…
Comment by Philip James on 1 February 2008:
Drazen
Here’s a link to some of our merchant partners: http://www.snooth.com/partners. These companies send us data.
Philip
Snooth
Comment by Robert McIntosh on 6 February 2008:
Very interesting thought. I think what would be most effective in order to ensure something like this succeeded would be for MERCHANTS (and not just online-only ones) to consolidate on this platform. I can’t understand why they do not except that it makes price comparison easier, but that will happen anyway.
If merchants added the information to a standard-ish format, the info would already be available to their customers looking to track, add and save that information. The most useful would then be adding value to their customers.
Non retailing sites would then be encouraged to adopt it to access the wealth of information available, thus also driving traffic to the merchants.
Comment by ryan on 7 February 2008:
I love cellartracker for my personal use - that’s it. No community features and no simple fix for the fact that not everyone thinks in a 100pt scale. They have the critical mass, but seem to be wasting it on poor design and layout. If anyone replicates Cellartracker with basic Web2.0 principles they will win the race.
The way to do this is simple - BE ANAL RETENTIVE WITH YOUR DATA - Make sure that there are no Burgundies and Burgundys as seperate wines. Or other problems. I think freebase is a nice idea, but I really don’t see it working until I can use it with my blog, and other bloggers and web geeks can easily implement it themselves. At that point the knowledge will grow, as others add info and the DB expands. I guess though that this will only lead to many duplicates of wines, and no standard naming solution.
I also agree with Robert - Merchants could make their jobs easier and ours.
Comment by Joel on 7 February 2008:
Well, Freebase is the backend. it would still require middleware to create a standard way to enter the data. Maybe it starts with merchants somehow adopting a framework to enter data into a site or sites using a Freebase backend.
I’m in a meeting writing this comment so I’m sure this comment is clear as mud.
Comment by Robert Cook on 10 February 2008:
Hi guys — I’m a Metaweb employee, and one of the designers of Freebase. I have a couple of comments and a question:
- A key goal of freebase is to “reconcile entities”. In other words, there shouldn’t be a two places to collect information about the same thing — eg., your Burgandys vs. Burgandies — in Freebase these would be a single topic. There’s a process for merging them together.
- Currently Joel is right that we are mostly a back end service. The current Freebase UI is far to general to be useful to a community of wine tasters, but we are working on features that would help non-programmers keep data in freebase and provide javascript tools to bring that data into other sites without technical work.
Which leads me to:
What features would you like to see on your blog that could be backed by a structured database?
Comment by Joel on 11 February 2008:
There has been alot of discussion about problems that need to be solved. i hope more people read this but the main issue is that there isn’t a consistent naming conventions/reviews/wine information convention that allows for wines to be looked up easily. For example - UPC codes for retail items is a universal standard. There is no equivalent in the world of wine. Every retailer, online and off, has their own codes in a proprietary way.
Same is true of wine tasting note sites. Every DB saves the notes in their own fashion. What a site (like Snooth.com) needs to do is work with every individual database (retailer, tasting site, what have you) to integrate the site’s custom DB schema into their own algorithm or database.
The point of my post is that if all these sites could use some sort of framework and periodically upload to a site like Freebase, then, at a bare minimum, FB can start to create a huge wine review DB that can be queried. Not to mention other, more intelligent “Web3.0″ connections it may make.
I really hope other comment on this topic b/c there have been some lively debates on the topic.