Microsoft doesn't get it. They have an experimental .NET Framework which provides an STM implementation, yet warn of performance degradation by using it. The purpose of STM is not to simplify locking semantics, but to prevent unnecessary locks. If a transaction fails because another resource has changed the memory, the transaction is retried, ( Read more... )

I'm pounding down pina coladas and practicing my coding skills with ASP.NET 2.0. The new Fort Minor album is blasting, not a single light is on in the house.

Soon my wife will be home and put an end to my fun.

The F1 race at Nurburg airs live on Speed at 7:30am. So I'll be getting to bed early tonight.

The MSDN launch event for Visual Studio 2005 was not as useful as the one for SQL Server 2005. The speaker was curiously running a beta version of the software (whose final release was given out for free at the end of the event). Several times during the presentation, a wizard wouldn't work, and at one point, the IDE crashed completely. Aside ( Read more... )

I attended Microsoft's Technet launch event for SQL Server 2005 today. It was an excellent event, hosted by an excellent speaker, Michael Murphy. Notes about the features I found most valueable are inside.

• Availability is a little better this time around. If one of you database files gets busted, you can bring up the remaining database objects ( Read more... )