So I’ve seen the light of open source code, open source information - Wikipedia and all that goodness. Now maybe this exists - but I had this idea clarify in my head recently. Why not a open database, or a wikidb. Having an xml web service exposing this data would allow this publicly maintained information to be shared among the net for other common uses. The power of XML/Schemas/Namespaces could allow different parties to sculpt and consume this data for their own specific use.
For example, think of how many databases are collecting hockey player stats out there. Now all of each collection is probably using say 80% of the same data for different purposes. I.e. player names, stats current teams etc. Now website 1 may be a hockey player info site where #2 could be a fantasy pool site. If there was a “wikiDB” or something, each site could define their own schemes and queries based on that common data and take what they need.. and leave the rest 
This idea came to me because of my recent project - TownRenowned. We are collecting and organizing local town data that other organizations have already put together. That made me think — would be better for the entire Internet if we localized public data in one place? As a developer I hate redundant code/business logic — thinking beyond myself and my businesses and from the users/consumers perspective it must also be frustrating to see redundant/incorrect data. So who could/would implement this? Perhaps Wikipedia could open a web service to their current data and charge developers for the rights for high-bandwidth dev kits? Who knows, but it sounds like something the net needs.
Brandon
February 18th, 2007
I completely agree with you.
For sports stats, I noticed tsn now has a copyright at the bottom of their NHL standings. it links to stats.com. If you go to their site you can see they have clients like aol, msn, si.com, espn, fox sports … they also do turnkey solutions for stats and fantasy pools. Seems like a really specific wikidb and a good money maker.
I shudder a bit when I think about all the programmers who would use this service and not use data/output caching. The number of database hits to your wikidb would be immense and a lot of them could be reduced … seriously, how many times do you need to hit a database to get cities and states to populate dropdowns? It doesnt change that often!
BPAndrew
February 18th, 2007
“I shudder a bit when I think about all the programmers who would use this service and not use data/output caching. ”
heh yea, I know some profesonal webservices charge on transaction counts/bandwidth. i.e. salesforce.com. So whomever build this would have to build their pricing model around those situations or limit transactions per day like Google Maps.
So are you glad I wrote an “Idea post”?:P I knew there was a dry spell but I can only write about this stuff when ideas comes to me.
Brandon
February 18th, 2007
You’d probably have to build your pricing model around transactions. You wouldnt want to limit the transactions, especially if you’re providing a service that may be mission critical to the customer.
I’d just be worried that customers with money would burn through transactions and bog the server down in general. It wouldnt be long before people started dumping you because of a really slow service.
I’m glad you wrote an idea post! I think my problem is I sit around waiting for sites to be updated and I get impatient… I really need something better to do with my time!