Why [not] X?
Fri, 23 May 1997 14:54:31 -0700
>Date: Fri, 23 May 1997 20:19:51 +0100
>From: Cyber Surfer <email@example.com>
>Subject: Re: Why [not] X?
>At 21:18 22/05/97 -0500, David Gadbois <firstname.lastname@example.org> wrote:
>> Date: Thu, 22 May 1997 16:24:00 -0700
>> From: Mike McDonald <email@example.com>
>> Where's that browser running? If your machine is running a lisp
>> based OS, I doubt Netscape is going to work too well. (Assuming
>> only one machine. Having a second machine on the net just to run a
>> C based browser is cheating!)
>[answering Mike first]
>Speaking for myself only, I don't see a problem with this.
>I _have to_ use a non-Lisp OS. Even if I chose Linux, I'd
>probably still use it via my NT machine, running an X Server.
>Anyone who has doubts about using the web as the user
>interface should take a look at ActiveDesktop. I'm just wondering
>it'll be possible to use traditional CGI techniques to extend the
>desktop. In theory, yes; there are alreay tools that let you do
>this for Java and Perl. Why not Lisp?
I believe that I could build a lisp based system that used a web
browser as its only user interface. I also believe that for a large
chunk of apps, that it would function just fine. However, I haven't
seen how to build such a system that would work well for highly
interactive apps like editors, CAD programs, ... CGI isn't the answer.
It's basicly batch processing. Submit a request and wait for the
answer. It's too heavy weight for interactive tasks. Until I can
figure out how to do these types of apps, I'm unconvinced about using
HTML as the only UI.
>Web based tools make that possible, which may explain why so many
>mutually hostile groups all agree that the web is The Way To Go.
>Fight the web and we might only succeed in uniting them all against us.
Heck, if we manage that, we'll really have accomplished something!
(Whether that'd be good or bad is a value judgement left up to the