This is in response to Diderot's question in another thread, I think this deserves a new one tho...
Hey, Diderot, as you may know, I have been involved in teaching intro to programming courses iRL. So I'm actually really interested in working out something for LSL, but as usual I have time constraints.
I think there are a lot of ways to make in-world LSL classes effective; but it takes time to develop these kinds of classes. For example, the class that I'm involved in now uses lectures and software developed in at least 3 different universities. with at least 4 professors, and half a dozen graduate students over more than a year.
Now we (that is, the CDS) aren't (yet) giving out computer science diplomas, so we can lower the quality of service a little, I suppose. To develop a course here, we need to ask a few questions:
* what level of scripting do we want to teach? The answer ranges from teaching people with basically 0 knowledge of scripting, to people who already have CS degrees, and just want to understand how to apply their knowledge to SL.
* what kind of scripting do we want to teach? Scripting vehicles, for example, is way different from scripting a gambling machine, which is again much different from scripting dance machines. Ideally we should teach certain skills, but also *principles* of scripting such that people can teach themselves everything else after finishing the course.
* how do we want to teach it? These range from traditional classroom type lectures to selfpaced tutorials (which could be via in-world objects, ala ivory tower, or web-based)
* how do we evaluate a person's progress? Do we give any kind of certification at all? Should we have exams? Quick and frequent feedback on a person's progress is good both for the instructor and the student, but how do we do this?
Then, we have some unique SL-related problems too:
* student commitment. Normal courses will build on stuff taught in earlier lessons. Say a full course takes 100 hours, how many students are able to commit that time?
* instructor commitment. Instructors will have to commit the same amount of time (likely more) as the students. While students can drop out without affecting the rest of the class, if an instructor drops out (because of RL conflicts, or whatever), the whole class gets screwed.
* synchronous meetings. getting 5 people to meet regularly at appointed hours is difficult, as we all know.
* etc. (speaking or RL commitments -- I have a lunch meeting)
So, RL classroom techniques are likely not gonna work without RL commitment between instructors and students. We have to consider that almost everyone who does SL, do it on their spare time.
I don't know what the answers are, but I'm interested in other people's ideas. One that has been floating about is a system of aprenticeship, similar to how the old guild was envisioned. But if so, we have to evaluate why the old guild failed, and how we can make it work.
I'd appreciate other people's thoughts on this.
EDIT:
Claude reminded me of the nstadt school, whose premises we could use for lectures, or self-paced learning kiosks, etc.
We also want to give some sort of value added to the [url=http://www.lslwiki.com:xelb9alc]LSL wiki[/url:xelb9alc], which, aside frome being the lsl reference, imho is already a good way to go about learning scripting for the self-motivated person.