[project] Draft for Block Closure Semantics for Squeak v0.6.0

Dan Ingalls Dan.Ingalls at disney.com
Tue Apr 24 03:48:28 UTC 2001


<progress report and suggestion from Andreas R>
>May be!

<various considerations>
>- it makes more fun to realize own ideas instead of trying to mimic other ones.

Stephan -

It's true.  Some redundancy is necesary to spread the joy of creating the universe (and, of course, to find all the best paths).

>But at the end there has to be an evaluation regarding
>- speed,
>- memory usage, and (last but not least)
>- elegance.

Memory usage looks reasonable.
I think your low "parts count" already gives you a decent elegance rating.

Regarding speed, my perspective on this is simply: what will J3 and its successors do with it?  If performance will be good there, then go with it.  Almost by definition, anyone who cares about speed will be using Jitter.

So, Ian and Marcus (or anyone else who is solidly into it), does Stephan's design look like it can be translated to ideally fast code by J3 and its successors?

	- Dan






More information about the Squeak-dev mailing list