Another fun filled day at PASS started with a session from Kalen Delaney.  The room was packed and she was discussing query plans.  I had heard a good bit of the content before but it was still interesting to finally see her present live.  Very much like Itzik, her asides can be as informative as the content in her slides or demos.

The second session was a lot of fun as Paul Nielson and Louis Davidson had everyone join in their personal debate over what exactly is “proper design”.  It was a 300 level session so nothing really earth-shattering came from it but it was really interesting to see the wide range of opinions across the room.  There were even defenders of the dreaded “Muck” table, or the single table housing all look-up data (the horror!).  I tend to fall on Louis’ side of most issues, but Paul made some strong arguments as well.  At the end of the session Paul demoed a project called NORDIC which is an object-oriented database design.  I’ll need to give this more attention before commenting but it certainly left me with questions swirling in my head.

The third session I attended was the most fun for me personally.  Erik Kang, Program Manager from the product team, discussed intellisense and the debugger in 2008.  We talked for a while after the session and I have a much better understanding of where he was coming from back in our early conversations about down-level support for intellisense.  I think there are other pieces of the product that will be much better for the attention devoted to them than the time for implementing down-level support.  I missed two opportunities to meet his team this week which really bummed me out.  Just too many pieces of my schedule have been in play to make it happen.

The last session was a question and answer time with three general managers of the SQL team.  They basically let attendees line up and fire questions at them for over an hour and a half.  Now that’s brave.  Some really interesting questions (some polite others not so much) were asked ranging from dedication to new product features (we hardly knew you Notification Services) to future support for jdbc driver.  The coolest part to me was when a question was trending towards too technical a direction the GMs were able to point into the crowd and say “oh, here’s XXX he/she is the program manager for the area you’re discussing, they would be more than happy to talk more about it…”.  Another benefit of having the event so near Redmond.

All in all, a tremendous day of SQL geekiness.  I have to give some props again to Magenic for okaying this adventure out to the northwest.  It’s been a great time.

After Thursday’s sessions ended I was invited to a dinner thrown by Red Gate software.  I’m in their “Friends of Red Gate” program, which basically means I show their tremendous software to customers or at user groups.  It’s been an easy relationship for me as I am quite a fan of their products.

So the event is packed at a local restaurant, we all sit down and on either side of me I’ve got Peter deBetta and Adam Machanic.  Also at the table is Roman Rehak and Heidi Sweeney.  Not bad company at all, right?  So the conversation runs in every possible direction and a lot of red wine and good food follows.  I begged off the party at 3 as I’m a sissy (or at least that’s what I was told).

Red Gate put on a really great evening and their people are even better than the products they build (if you can imagine).

UPDATE: I added a couple iPhone photos from the evening. 

Note the serious, foreboding face of one Peter DeBetta…

And the rest of the gang…Rowan, Heidi, and Adam.

A quick post about today’s sessions so far…Ted Kummert’s keynote was interesting in how sparse his details were.  Aaron has a good overview of his talking points – I won’t bother retyping.  I guess I was expecting at least one big “woo hoo” announcement out of him.  Oh well.

The most interesting thing I saw was a demo from Dan Jones of a new feature called DAC.  The basic premise is that you have a “fabric” of instances defined and you can then create a deployment package (Dan used the analogy of an .msi for the database) for use against any of those instances.  The tool in SSMS will allow you to apply policy to the deployment and check for successful passage of said policies before implementing on the target instance.  Very cool.

Another great piece of the keynote was a demo of some of the DATAllegro features that will be rolled into the next version of the product.  A demo took place showing data being spread across multiple instances then coming together as it was accessed via an SSRS report.

I went to Mark Sousa’s session on lessons learned by the SQLCAT team around SQL 2008 installations.  The session focused on places where the product team expecting certain behaviors and were surprised by results on the ground (both positive and negative).

I’m about to head to a session with Paul Randall and Kimberly Tripp called “Corruption Survival Techniques”.  I suspect my head will be close to explosion after listening to those two for an hour and a half.  :)

More fun to come…

I’m a couple thousand miles over Colorado as I’m typing this, headed to Seattle and the PASS conference.  Should be a tremendous week of content and networking.  It’s also a bonus to be so close to the Microsoft campus.  I will finally get to put a face with a couple names that have been very helpful to me over the years.

If you are going to be attending the conference drop me a line through the Contact section of this blog and we can meet up during the week.

I recently broke my own blog admonition about Windows 7 and virtual environments.  In my defense, I came across a laptop drive in my home office that worked in my Dell D620.  That’s a sign right?  You’re just meant to install it… :)

So I gave it a whirl fully expecting to either blue screen my laptop (at best) or end up with scorch marks in the carpet and a bad limp.  I was thinking back to the beta of Vista that had all sorts of little maddening quirks.

I think I was struck most by just how uneventful the whole installation process was.  The look and feel was pretty Vista like with some nice differences from my current 64-bit Vista install.  I really liked the paired down feel with the removal of duplicated items like Mail and Photo Gallery (which can be replaced by installing Windows Live versions). 

The largest effort in setup was getting the “superbar” or taskbar to show up like the demos seen at PDC.  If you are interested Rafael has the easy steps to make that happen for you.

Knowing I was dealing with a spare drive I got a bit of a daredevil streak going.  I tried installing the drivers for my broadband modem.  I installed Office, I set up connection to my company’s Exchange server and started dragging email over.  I then kicked off the Windows Live installer and fired up IE.  With all this activity going on I still never saw higher than 1.3 gigs of memory utilization.  Sweet.  I continued to beat on the OS just waiting for the “well there you go…” moment where the beta rolled over.  Never happened.

The only things that really never worked were installing Flash (I never pay attention to how many sites use flash until I’m on a machine without it) and using Live Mesh.  While I can’t truly say Live Mesh didn’t work it did decide to change my color scheme at random intervals.  Perhaps this is just a user interface “feature” I’m just not getting.  I am a data guy after all.  ;)

I even went so far as to use this OS for a week at my current client.  There were no crashes, no problems.  I think Microsoft really has a winner with Windows 7.

So hopefully this will explain a bit of where I’ve been in regards to this blog.  Focus will be returning to SQL Server after this R&D boondoggle, just need to make a trip to Seattle first.