John Meacham wrote:
<ramble>
On Fri, Jun 09, 2006 at 02:16:47AM -0700, Clifford Beshers wrote:
Interesting. I just gave a talk to the SGVLUG (San Gabriel Valley Linux Users Group, which is centered at Cal Tech). It was the first time I've given such a talk, half about Linspire/Freespire, half about Haskell features, and the other three halves were technical problems.
Oh, I live a block from Caltech, I didn't know there was a Haskell talk there.
Dang! I referenced your 'small but featureful grep' as an example of how Haskell should help reduce the need for `little languages' and two-level languages. I promised to come back in six months or so and talk about the progress we've made on development tools for Freespire. I'll let you know.
Writing the slides, I found that it is hard to disentangle all the concepts and build from the ground up. Those of us who use it have forgotten just how many new concepts there are and how tightly bound together they are in Haskell. As always, when you try to teach something you get a deeper understanding of it. I'll see if I can't clean up some of the examples with hindsight and send it along to you and see what you think.
I always prefered using a chalkboard (or whiteboard, or overhead + markers) instead of a pre-prepared slideshow when giving talks. it lets me change the focus depending on audience reaction and questions more. If you can get away with it, I'd recommend it for future talks, ignore anyone that says it is not profesional, they wouldn't have paid attention anyway to anything other than your font choices and choice of screen-wipes between slides.
If I had been doing a tutorial, I might have done that, but I was doing a whirlwind tour where the goal was to get people excited. Also, I find that the first time presenting some information, I do better if I lay it out before hand. Explanation of code is a different beast than creation of it.