[Haskell-cafe] Master's thesis project at Philips Research
hughperkins at gmail.com
Thu Jul 19 17:32:02 EDT 2007
Not applying (probably not eligible), just answering for fun, so feel free
to pipe to null :-D :
Is your target user for the language the end-user of the toy itself, or a
toy designer who will be using your middleware?
I'm guessing the latter. Is there any reason why you cant use Lua? It's an
existing language ,very light-weight, well-known (eg World of Warcraft, TA
Spring, possibly Supreme Commander (unsure) all use it. It's dead easy to
embed securely (you explicitly choose which functions/libraries to expose).
I understand your design goal is "The language should be easily
understandable for a non-technical person", and perhaps you will make this
truly a reality. In general, even in the simplest language, there are
several conceptual hurdles for doing anything much more simple than printing
- the concept that the machine reads the instructions line by line
(actually, in FP this is not so much the case, arguably FP could be better
for non-technical people from this aspect)
- another way of saying this: many beginners think the computer somehow
"understands" the program and somehow miraculously obeys it, rather than
mindlessly reading line by line (again, with Haskell, the beginner's view is
quite close to reality, arguably this could be easier in an FP?)
Probably there's a whole load of psychology research on this already done
As I write this, I'm thinking: if your goal really is to make the language
understandable by a non-technical person, you might want to get some sort of
specialized psychologist on your team???
If you're on a budget and just want something that works, consider Lua ;-)
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Haskell-Cafe