SAS® USER PROFILE
Names: Tom Kari
Location: Ottawa, Ontario
Company: Statistics Canada
SAS History (years using SAS, platforms, primary products, etc.): I started using SAS somewhere around 1980; I remember that my first manual was for SAS 79, and being very excited when SAS 81 (82?) came out, with TWO manuals (Base and Stat) needed to cover this enormous software product. Needless to say, this was on the mainframe.
Humorous interlude: we decided to use SAS for a project in 1987; we thought we'd use the mainframe, but we decided to try out the new PC version. Unfortunately, we had 5,000 records to process, and that killed it. I was involved in SAS on and off over the years, but really linked up again when we introduced Enterprise Guide for the Census around 2004. I've been drinking from the fire hose ever since.
Hobbies: I absolutely love to travel, and I'll pick up my suitcase at the drop of a hat, if I may mix a metaphor (frequently heard around the office: "Are you going somewhere AGAIN!?!")
I've also been studying Judo since about 1993. I'm not "ten fingers of death" yet, but I'm up to about six and a half on a good day.
Ideal weekend: Waking up at the cottage in July, lots of coffee, some bagels, and then out to ski or wakeboard. Unfortunately, our ski boat died in 2007, so I'll have to stick to canoeing for now. But coming soon, a new boat!
Thoughts on winning the SAS User Feedback Award at SAS Global Forum this year: My heavens, it's been so overwhelming! This is absolutely a career high note for me, and I'll treasure all of the kind comments forever. Sigh, unfortunately I'll be back at work soon, and the carriage will turn back into a pumpkin. But it was fantastic while it lasted!
If I could be anything at all (besides a SAS programmer), I would be: I have developed a great interest in international affairs over the last few years, and I hope to make some kind of contribution in this area after my retirement from Statistics Canada. (Just in case anybody from CIDA is reading this!)
One thing every SAS programmer should know: Sorry, two things. 1. PROC MEANS: Over the last few years, I've been working on various projects involving very large volume tabulations. PROC MEANS is an absolute beast; we've never broken it yet. 2. PROC SQL: Anyone who is good at DATA step programming should add a rudimentary knowledge of SQL to their toolkit. The two are very complimentary; things that are hard in one are easy in the other.