Documentation... (was Re: Article in Wired)

Peter Crowther Peter.Crowther at IT-IQ.com
Fri Jun 11 07:43:54 UTC 1999


> From:	Russell Allen [SMTP:russell.allen at firebirdmedia.com]
> Server side stuff has requirements of stability and safety which Squeak
> isn't suited for - no matter how well PWS is written it won't approach the
> speed and rock-solid permanance of Apache.
[...]
> The question we should ask ourselves is "How
> can we leverage the amazingly powerful networking standards like TCP/IP
and
> HTTP etc to enable new forms of collaborative creation of great stuff, and
> new forms of communication?"

While Squeak *may* not (yet) be suited to server-side applications, I would
argue that Smalltalk as a language is pretty neutral about the whole idea.
If I'm doing string-slicing, at least I don't have to care about whether the
particular method I'm using is built into the language or derived; I can
even modify the VM/VI to put a new string primitive in if I find something
that I'm doing regularly that's slow in Smalltalk.

Even an 'exquisite personal computing environment' like Squeak actually
works pretty well for the server-side application I'm writing (a MUD).  This
is a server application where the users rewrite how the server works!  It's
considerably faster than the bodged C/C++ system we have now, and its
built-in incremental compilation, garbage collection et al have given us
spectacular savings in development time.

Does this count as 'collaborative creation of great stuff'?  I hope so.  I
hope this MUD will eventually be more functional than a MOO (Multi-User
Object-Oriented system - developed at PARC, interestingly :-) and,
coincidentally, use a more standard language as well.

Now if I could just get Smalltalk embedded as a scripting language in IE6
and NS...

		- Peter





More information about the Squeak-dev mailing list