[10047] in www-talk@info.cern.ch
Re: Java and HTML and well known socket numbers
daemon@ATHENA.MIT.EDU (Matthew James Marnell)
Tue Jun 4 15:53:45 1996
To: www-talk@w3.org
In-Reply-To: Your message of "Tue, 04 Jun 1996 09:37:57 PDT."
<Pine.LNX.3.93.960604093637.190A-100000@yuan.fv.com>
Date: Tue, 04 Jun 1996 15:48:36 -0400
From: Matthew James Marnell <marnellm@portia.portia.com>
Resent-From: www-talk@w3.org
I've taken www-html out of this.
:>> back, why send one back? Why send a whole big bloody applet back
:>> when you can use any number of protocols to talk back and forth.
:>
:>Well, if you're running a public server, users might want to upload Java
:>applets to be run when forms are filled out, for example. (I.e., I might
:>want to replace insecure CGI scripts with "secure" Java applets, so I
:>don't need to hand-check all user scripts for security purposes.)
Using java applications as CGI scripts is on thing. Who often do you
think your user is going to upload the same app to handle forms? If
it's a java applet embeded in the web page that acts as a form,
parses the data and sends it off to the owner of the page, then we're
not talking about 2-way applets, we're talking about one out-going
applet that either talks back to the SMTP port or Oracle port, or
some such for delivery of the data. Since one of the future releases
of Java will have SSL and other security measures built in, it can
be carried encrypted back and forth. This is likely to carry a heck
of a lot less overhead than sending applets back and forth needlessly.
Matt
