[5897] in www-talk@info.cern.ch
Re: Forms support in clients
daemon@ATHENA.MIT.EDU (Karl Auerbach)
Wed Sep 28 23:07:57 1994
Date: Thu, 29 Sep 1994 03:54:57 +0100
Errors-To: listmaster@www0.cern.ch
Errors-To: listmaster@www0.cern.ch
Reply-To: karl@cavebear.com
From: Karl Auerbach <karl@cavebear.com>
To: Multiple recipients of list <www-talk@www0.cern.ch>
> > I would expect that using e-mail would be difficult to smoothly
> > integrate and probably a pain to administer. Rather, I would want a
> > more unified and consistent exhange mechanism, so that, for example,
> > only one TCP connection need be used.
>
> But this implies, for example, that your Mosaic-like browser is always
> running, doesn't it? It seems to me that if you want the remote agent
> to be able to notify you whenever it finds something, regardless of what
> you are doing, you need an asynchronous protocol for personal
> communication, which to my mind means some variant of email....
Who say's I'm running mosaic or anything like it? Perhaps I have a tool
on my end that knows how to generate these scripts for the server
and anxiously awaits a reply?
Yes, it would be nice if the internet had an asynchronous delivery
protocol. E-mail is being used to approximate it, but it is a noisy
channel with high administrative overhead.
I did an asynchronous text file transfer protocol (way back in the
very early 1980's -- before sendmail) and it served this purpose.
IBM's SNADS also serves this purpose. They could carry e-mail quite
nicely along with other kinds of traffic -- forms, print, etc. It's
somewhat of an inversion of the concept to try to layer general text
files over e-mail.
So, yes, I wan't an asycnronous path, but I don't agree that this
means e-mail.
--karl--
