secure web page ideas

For those people who have Netscape / an SSL-enabled web-browser, wouldn't 
it be useful to have secure web pages that did the following:

(1) An anonymous remailer web page, like the current ones (this one has 
    the obvious advantage that plaintext doesn't travel to the http server)

(2) A pgp-sending web page (type in key id into field, send message to 
    address given, encrypted)  This isn't a bad idea for the same reason that
    (1) above is a much better idea.

(3) Ultimately, a server that did (2), through (1).  Being able to send PGP
    encrypted email to a recipient through anonymous remailers, over the web
    with a secure browser might be PGP's "killer app" in one way or another.

How hard would this be to implement? Would it be worth waiting until the PGP 
3.0 API is released?

Jon
----------
Jon Lasser (410)494-3072                         - Obscenity  is a crutch  for
jlasser@rwd.goucher.edu                            inarticulate motherfuckers.
http://www.goucher.edu/~jlasser/
Finger for PGP key (1024/EC001E4D)               - Fuck the CDA.

On Wed, 14 Feb 1996, Jon Lasser wrote:

> For those people who have Netscape / an SSL-enabled web-browser, wouldn't 
> it be useful to have secure web pages that did the following:
> ...
> (3) Ultimately, a server that did (2), through (1).  Being able to send PGP
>     encrypted email to a recipient through anonymous remailers, over the web
>     with a secure browser might be PGP's "killer app" in one way or another.
>
> How hard would this be to implement? Would it be worth waiting until the PGP 
> 3.0 API is released?

At first I wondered why you'd want to send encrypted mail you couldn't 
sign with your own key, but on second thought, I can think of a lot of 
reasons.

I'd think that much of the hard and unique work would be, first, proper
interface design, and second, putting together an efficient database and
database extraction mechanism for huge key rings (I notice that the MIT
keyserver was recently upgraded; talk to them). While the PGP 3.0 API
would help with the implementation, I see no reason to wait on these
infrastructure steps.  The final version would use the more secure and
efficient API, but quick modular hacks for steps that would be fulfilled
by the API should be all that is needed. 

If you put together a proof-of-concept model, even one that doesn't 
really work, I think you'd see this idea take off. I see no reason to 
wait, and while I'm certainly no CODERpunk, I'd be more than happy to 
play with user interface mockups in my copious free time... 

-rich
 Institute for Ernst Zundel Revisionism
 http://36.190.0.210/~llurch/Not_By_Me_Not_My_Views/