Interesting blogs I’ve read this week

I have been slowly emerging from a coma, at least in terms of my public presence.  I posted on twitter today for the first time since August, and I actually browsed a couple of topics on a few SQL Server-related forums.  I have been nose to the keyboard for 8 or 9 weeks on a very intensive integration project, and have barely had time to breathe, never mind entertain you folks.  Now that the “heavy” phase of the project is winding down, I thought I might kick up this blog to blurt out some things that DON’T fit within 140 characters.

My first post (by the way, Hello World!) is just a collection of links I found interesting this week, and why.

• Michelle Ufford (who tweets as @SQLFool) wrote a great post on Partitioning Tricks.  This was interesting to me, and the timing was perfect, because I was starting to contemplate how I could use partitioning in the next phase of the aforementioned project.  Michelle’s explanations of replicating and compressing partitions turned on a few light bulbs for me, and I think I am well on my way to coming up with a plan.
   
• Jonathan Kehayias (who tweets as @SQLSarg) warned us about a GDI+ vulnerability, and explained some of the things that aren’t very intuitive in the actual security bulletin.  I’m not rushing out to apply this fix to all of my production servers until I really understand the scope of my vulnerability, but kudos to Jonathan all the same.
   
• Aaron Bertrand (who tweets as @AaronBertrand) started a tremendous series last week, entitled, “Bad habits to kick.”  Over a series of at least a dozen posts, he goes over several bad habits that a lot of us have, and explains both why they are bad and how we can overcome them.  Sadly, a lot of the comments are just inviting a religious battle about whether CustomerOrders is better than customer_orders, or vice-versa.  Still, the series is very entertaining, and I am embarrassed to admit that I am guilty of more than one.
   
• Speaking of religious battles, Mladen Prajdić (who tweets as @MladenPrajdic) wrote a post on surrogates vs. natural keys.  A lot of interesting comments (but no fisticuffs, yet) have emerged from this emotional topic.
   
• Paul Randal (who tweets as @PaulRandal) wrote a detailed post on which index SQL Server will use to count rows.  Personally I typically use sys.dm_db_partition_stats to get row counts, but I found it interesting (and in hindsight, obvious!) that SQL Server is much smarter these days about optimizing COUNT(*) queries.