I spent all of yesterday interacting with visitors to the MATLAB Virtual Conference. It was great fun! A blogger wrote that it was the "high point of geekism and nerdity of this year." (I think that was intended to be a compliment!)
I started out early in the morning in the Image and Video Processing booth, participating in the group chat there. When it got busy it was a bit challenging to follow all the conversational threads, but I think most people got their questions answered, and I came away with a lot of good product feedback.
Then I went over to hear Tom Kush and Roy Lurie give their "MATLAB Universe" keynote address. If you want to hear about the world-wide impact of MATLAB and get some insight into where we think MATLAB is going, check out this talk. (I didn't get a chance to listen to the other talks yesterday, but I will later.)
The experience reminded me of some of the MATLAB user conferences in the 1990s. One of the only times I've ever heard company president Jack Little tell a joke was during his keynote address at the 1995 user conference in Cambridge, Massachusetts. We had been working for several years on MATLAB 5, which was going to be a huge release. Conference attendees had been promised a preview. So at the opening of the conference, in an enormous ballroom, Jack fired up the development build of MATLAB 5 on the big screen and started to talk about all the exciting new features to come. He said there would be a few changes that might take some getting used to. (We developers were sitting on the back row, holding our breath.) Jack explained that we had decided to adopt RPN (Reverse Polish Notation, popular in HP engineering calculators such as my beloved HP 15C) in MATLAB. He proceeded to type something like:
>> 5 3 +
There was a massive collective gasp! from the audience as we about fell off our chairs in the back trying not to laugh out loud.
To any reader who was there (and is still recovering), I can only say, "We're sorry!" ;-)
OK, back to the present. Later in the day I gave a talk on MATLAB array indexing techniques that are useful for image processing, borrowing heavily from material I originally posted here on the blog. I was afraid the material would be too narrowly focused, but the audience stayed with me to the end and then asked a bunch of great questions about my last topic, neighbor indexing, which was pretty advanced.
For the first five minutes of the talk (which I recorded a couple of weeks ago), I was in a rapidly increasing panic, because a 2-second echo was making the audio completely unintelligible! I bolted out of my office to find someone to help. Kevin calmed me down and helped me figure out that I had a second, forgotten, browser window open to the talk, and the echo was coming there. Boy did I feel silly. Thanks, Kevin!
Readers, I have two questions for you. First, did you attend any of the original MATLAB user conferences in the 90s? Any favorite memories to share?
Second, did you attend the virtual conference yesterday? What did you think? If we do it again, what can we do better next time?
Thanks for taking the time to give us your feedback.
To leave a comment, please click here to sign in to your MathWorks Account or create a new one.