Vancouver TechFest 2009 Recap

Thank you to Rob Chartier and all of the other volunteers and sponsors of Vancouver TechFest 2009.  Due to other commitments, I wasn't able to hang around there all day - although I wish I could have.  My "Data Warehousing and Business Intelligence Primer" session was well attended and seemed to be well received.  (Except for one guy who couldn't seem to help dozing off.. firecrackers next time...)
There were some very interesting questions and perspectives from the audience - and the overwhelming demographic of techies didn't seem to faze the four or five "business" oriented people.  I was heartened to see those particular people nodding subconciously every time I mentioned something to the effect that accumulating cool technology wasn't the best reason to start using BI.  I hope the majority of the audience caught my message on that point - I certainly wasn't subtle.
My presentation materials and a recording of the session will soon be available at the TechFest website.  In case that's not immediately available to you, I've also posted the slide deck on my SkyDrive.  Update 2009-11-25: The presentation recording is now available.
If you attended or heard about the presentation and would like to know more, keep a watch on the Vancouver PASS chapter site.  I would like to slot in a night about the more technical aspects that appear in the middle of this TechFest session - about how a traditional OLTP database isn't appropriate for delivering Business Intelligence.  Since this group isn't the BI-focused chapter, I intend to discuss things from a point of view of a DBA who needs to make both scenarios play well together.  So if you're a interested in what a dimensional model is, why denormalizing is OK in a data warehouse, and what a column-store database is - you don't have to wait too long.

Presenting at Vancouver TechFest on Nov 21

I suppose I can't get enough of this stuff - but this presentation is a bigger step up from those I've done previously.  (Or so I'm leading myself to believe.)  On November 21st, the third annual Techfest will be held in Vancouver - at the BCIT campus in Burnaby, to be precise.  I'll be on stage at 2:30pm, and my topic is...
Business Intelligence and Data Warehousing Primer
You can see the site for the full abstract.  The short version is that I intend to talk to non-BI people about what BI and DW is, and how it's different from normal database work.  If you're a business professional, developer, or DBA who knows very little or nothing about this "new thing" called BI, you should definitely consider coming.
Why should BI be on your radar if you're one of those people?  All of the "analyst" groups (Gartner, Forrester, ...) have identified BI in the top three areas that businesses and/or CIOs are interested in for the last decade almost every time they do a survey.  Sure - sometimes virtualization or SaaS beats it, but it's consistently top of mind for business and tech leaders. 
In just over an hour, I'll let you know what you need to know.  I'll lead off with describing what BI is and what value it's supposed to provide.  Hopefully with metaphors that resonate with the audience.
If you're a Dev or DBA, the next part of the session will talk more directly to you about how you need to unlearn some database "rules" that have been (correctly) pounded into you over years and years of working with OLTP systems - you need to know this stuff so you don't kneecap your BI efforts with poor design.
To wrap up, I'll tell you what a BI project needs to succeed - something everybody there needs to know, regardless what specific group they belong to.
A Bigger Step
I bit off a big chunk when I requested a spot for this topic.  I'm a developer at heart, so it's relatively easy for me to talk about the technical stuff.  Trying to describe a topic that should also speak strongly to business people who are curious about tech, as well as geeks who need to know what their business counterparts are talking about is quite a deviation from that.  On top of that, Rich Baumet let me know last week that there are about 700 people registered for the event... and there's only four tracks.  That could mean (naively) that I could be talking to 175 people - quite a step up from a 50 person user group presentation.  The third thing I'm working towards making this a success is that I'm using all those people as a testing ground for a possible repeat of the presentation - to my own execs.  Sure, it will need some tweaking for the different audience - but I'm really hoping that the pressure for this will reduce the pressure for the second go-around.  And I'm really hoping for some feedback from the audience on what worked and what didn't!
As usual, I'll be posting my slides and a recap of the session after the event - hope to see you there!

PASS Summit 2009 Recap

The PASS Community Summit was an excellent experience for me - and I hope it also was for anyone else who was fortunate enough to attend this year.  Rushabh Mehta mentioned during his keynote on Wednesday that attendance this year was down only 9% - whereas conference attendance in general was down over 30% this year.  That certainly is proof of the value that I experienced - the PASS Summit is definitely the conference of the year for great SQL Server content.
Sessions
Most of the high points for me were actually outside of the excellent sessions - but I'll start with them.  There were way too many sessions for a single person, of course - I didn't get to see all of the ones I wanted to.  On top of that, I volunteered some time which took away from that traditional brain-expanding pursuit.  However, of those that I did get to attend, two stick out from the excellent pack as exceptional.
Paul Randal's session on transaction logs (and crash recovery) was simply superb - although I'm sure it's "old hat" to those of you who are real DBAs.  (I am most certainly NOT.)  It described the complete life and purpose of a transaction log in fine detail while remaining completely clear and focused.  In any technical topic, one of the difficulties is the spiderweb of relationships that need to be explained in a linear fashion - and Paul did that without problem.
Marco Russo and Alberto Ferrari's presentation was also top-notch.  Two Italians playing "good cop"/"bad cop" with Inmon vs. Kimball data warehousing methodologies in "spaghetti English" was simply gut-busting.  I don't know if they intended to be that funny - but they definitely were.  On top of a great overview of both of those architectures, the SQBI.EU pair presented their "hybrid" approach that certainly appears to blend the better parts of both into a very practical industrial-strength, quick-to-delivery methodology.  I recommend that you investigate their ideas and download their whitepaper - I definitely will.
MVP Birds of a Feather Lunch
Many thanks to Mike Walsh for coordinating this event, and the efforts of many PASS volunteers for pulling it all together.  I had a great time, and was very pleasantly surprised that there was more than one person (me) wanting to talk about scripting and custom objects in Integration Services.  I know I learned some things - and got other points reinforced.  I will definitely take advantage of that opportunity, should it happen again.
Meeting Other SQL Addicts
I'm talking mostly about the other Twitter SQL geeks - but there were plenty of non-twits I got to meet as well.
Meeting up with Joy Mundy
I'm a fan of the Kimball group for many reasons - Ralph is a visionary in data warehousing, and he's got a great team of teachers and practitioners assembled there.  I first met Joy at a Kimball course, and I'd last seen her at the MS BI conference in 2008, where I'm sure I was a little overbearing - I just had to give her a demo of the Kimball SCD component I made to address what I saw as the problems in the SCD Wizard.  She certainly seemed receptive and took it all in - but I didn't happen to hear from her after that.  My obnoxious self couldn't not try again this year - so I snuck in to the end of her session and said hi.  This is what I recall from the conversation:
Me: "Hi Joy, I'm Todd McDermid - we met last year at the MS BI conference."
Joy: "Yes, I remember.  You know that part of our course where we spend half a day talking about how to do slowly changing dimensions?  We've cut that out now, and just tell them to use your component."
Wow.

SSIS Community Tasks and Components Site - One Year Later

As far as I'm concerned, the Community Tasks and Components Site on Codeplex is a great success in just one year.  Back last November, I told you how John Welch and I had taken it upon ourselves to catalog the extensions available for SQL Server Integration Services, since there didn't seem to be any good way of finding the amazing things people had contributed - most free to use, with source.
Today, the catalog has over 150 extensions listed, and has been visited over 12,000 times (40,000 page views)!  I caught up with John at the PASS Summit, where we talked about the project with other SQL MVPs and the Integration Services dev team.  With respect to the project, there are a few items on Connect that I'm interested in having addressed to make custom objects cooler, and a little easier to code, and install.
Look forward to year 2 of the Community Tasks and Components project - there are some good things in store!