Dave Campbell: Inside SQL Services

The Discussion

• 

  Charles please slow down with the posts, I can't cope. Especially if there is goldust by the most venerable Technical Fellows.
  
  I feel like a kid in a sweet shop at the moment, completely hysterical!
  

• 

  Much more to come (but I can't tell what it's about or who's being interviewed).
  
  Keep on watching and listening. Of course, please talk back. We need you feedback. So, what do think of Azure? How will you innovate on it?
  C

• 

  Am I the only one who, when he talked about the underlying filesystem, instantly thought that it sounds kinda "WinFS-ish"?
  
  (just minus the metadata stuff)

• 

  Too bad it's called SQL Services.
  If it's not a SQL Server (which is presumably a good thing), it would make sense to drop the SQL moniker all together and name it something like Cloud Database Services. If it does offer a SQL query interface, that's fine, but it doesn't mean that SQL should be part of the name.
  If it offered a solid relational calculus DSL and no SQL support, no one would complain.
  
  As soon as you guys are building a new database service, why don't you take it a bit further and allow defining entities and relations between them as predicates in a semi-natural language? Relvars are predicates, according to the relational database theory. Why not to preserve the relvar's original meaning in the database metadata? That would be a competitive advantage.
  
  PS: What's that S-LINK thing, anyways? Is it a subset of LINK or a superset of thereof?
  

• 

  
  

  Kudos to the SDS team for releasing Ruby implementations for SQL Services. For the rest of us!

  http://www.azurejournal.com/2008/10/ruby-and-sql-data-services-sdk/" target="_self">http://www.azurejournal.com/2008/10/ruby-and-sql-data-services-sdk

  
  

  -PT.