I didn't think much of the "Reliability" portion of the  
presentation.  I experienced my typical distaste for static typing,  
overcomplicated syntax, and early binding; this is common when I see  
a C++ programmer's wishlist for an ideal language.

GetElement(as:[]string, i:int):string=
     if(n:nat<as.length=i)
		as[n]
     else
		“Index Out of Bounds”

WTF?  This is a helpful type system?  Why just not write a test to  
make sure that the value is in bounds, since the type definition  
itself is code (with extra new syntax to learn).

Also, what about mobile code?  If I bring a new object (containing  
new code) into someone else's virtual world, will it statically  
compile it against everyone else's code in the environment?  Even if  
it does, how do we know that the semantics match up?  This is a hard  
problem, and static typing doesn't magically fix it (he does admit  
that Haskell-style type inference doesn't scale to "large hierarchies  
of open-world modules").

The "Concurrency" portion of the presentation is more interesting.  I  
like how he breaks down the computations done by a game engine by CPU  
budget and parallism/synchronization requirements.  I've now heard  
about Transactional Memory from a few excellent hackers, and  
definitely intend to read up on it.

Two other things that look interesting are Lenient Evaluation and  
STRef.  My functional programming skills can use some practice, and I  
intend to find some time to learn about these concepts.  However, I  
question whether the general population of programmers is ready for  
the higher levels of abstraction required to program functionally.   
But maybe the massively parallel computers that we'll all have in ~10  
years (depending on what you call massively parallel) will force the  
issue.

Josh




On Feb 6, 2006, at 6:17 AM, goran at krampe.se wrote:

> ...can be found over at Lambda:
>
> http://lambda-the-ultimate.org/node/1277
>
> The presentation is from a game developer and was quite interesting to
> look through (the slides are pretty self explanatory). It would be  
> very
> interesting to hear the reactions on these issues from a Squeak
> perspective - like what do the Croqueteers say and what do you say
> Bryce?
>
> If a future Squeak could be interesting for this segment (game dev)  
> then
> things would really heat up.
>
> regards, Göran
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20060206/ee1701c8/attachment.htm