A couple of interesting things I remembered and will probably be blogging about once the SQL Server 2008 CTP5 bits are available.
- CTP 5 should be released by the end of this month (more information should be available this Monday).
- Filtered indexes are cool and allow you to mimic an Oracle behavior as a side effect. You can now create a unique index but allow more than one NULL.
- Sparse columns and column sets are really cool (especially for data warehousing).
- Change tracking, which is used for sync services, can also be used for other purposes by yourself (and maybe one day for Merge Replication?).
- Filestream will be a valuable addition but should be carefully considered. One of the biggest drawbacks in my opinion is the lack of database mirroring support.
- There are actually names for a lot of things we have implemented.
- Data mining can be used for purposes you might not think of at first as was displayed by Rafal Lukawiecki.
- Many people still do not understand that a database engine is not a psychic, if you have no way of logically telling it which records you need it will not use the book of Nostradamus to guess. Oracle and Delphi were actually psychics by the way.
- Belgian fries still rule and yet they are called French fries. If you want to find out why check out Wikipedia.
I hope to get back to you soon because that would mean CTP5 was released ;-)