Andreas Raab wrote:
> 
> John,
> 
> > I'm not entirely clear what the benefits of true "headlessness"
> > are on the Windows platform. It seems to me that one can always
> > make the Squeak window quite small (to use less Display memory),
> > then minimize it so it is out of the way. That way, if there are
> > any errors, one can re-size the window and debug the problem.
> 
> Not if you're going to install it on any jointly used machines.
> Think of a Swiki web server running as a service under NT.
> 
> > Headless-ness seems to make a bit more sense on a Unix box.
> > But there, again, one could open Squeak's display as a remote
> > X window on some other machine, just as a debugging console.
> 
> Yeah, this would be cool. Install a Swiki demon, telnet on a special port
> and run your debugging session remote without even letting others know
> that something happens right now. WHOW!
> 
> Andreas
> 

Basically right.  One benefit of true headlessness
under UNIX (or NT maybe) is the ability to run on a machine
with no window, not having to export the window.  In some 
cases exporting the window becomes a security risk or an
extra headache you don't need (if you are trying to start
your web server over a text only slow phone connection for example)

If you want to be able to use Squeak as a web content
generator (and I do, eventually) or a web server etc. then
running headless is in some computing environments a necessity.
If anyone wants to know the gritty details of why just email me...

-- 
Dana Lynne Goldblatt Anthony
Internet Publications Technology at SAS Institute
World Headquarters Cary, NC, USA