From: Niels Provos <provos@ws1.physnet.uni-hamburg.de>
Date: Sat, 10 Aug 1996 02:40:02 +0800
To: geoff <geoff@commtouch.co.il>
Subject: Re: PGP Mailer for the masses ?
In-Reply-To: <19960809124443785.AAA218@[194.90.103.93]>
Message-ID: <Pine.A32.3.94.960809153630.15623A-100000@ws1.physnet.uni-hamburg.de>
MIME-Version: 1.0
Content-Type: text/plain


-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 9 Aug 1996, geoff wrote:
> Thanks, for describing the features of Pronto Secure :)
> This is how Pronto Secure matches up to your checklist:
> > - Traverse the web of trust and show how the public key is
> >    related to one own keys to mutual signatures on other public keys
> >    ( For example mean distance to a key signed by the recipient
> >    himself )
> NO  (we handle certification by allowing the user to modify a list of  
>     trusted certifiers for signing keys)
I personally would find it useful if you could get a measure of
trustworthy ness due to keydistance. Like that if I knew that the sender
is only 2 keys away from my own I would most likely trust his public key
to be the original, it would be nice to see the signing people involved
though:
Max Miller 
  |-signed- a friend of mine <-signed- me
  |-signed- Molly Malone <-signed- Someone <-signed- friend <- signed me

According to http://bcn.boulder.co.us/~neal/pgpstat/ there were 19124 keys
in the keyserver but the biggest werb of trust had only 1291 keys and
the next only 16 keys. The mean key distance was between 6 and 7.
That means that you don't have to get too many public keys in order to
find the connection to your key and on the other hand it might show that
calculating trust according to keydistance isn't worth is since the web
of trusts are so small. But I figure if you would show the connection of
keys in your mailer it might encourage people to participate in key
signing parties.

> > Misc:
> > - Passphrase should be kept in memory for a definable time, 0 for
> >  immediate deletion, thus you would be prompted for the passphrase 
> >  each time you use it. Question about Windows Swapspace ? or tag the 
> >  memory as uncacheable ?
> NO (Keyboard sniffing is too easy to do in Windows, This would give    
>    a false sense of security)
How would you get the pass phrase if not via the keyboard ? And if you
keep it in memory till you sign off ( like in premail ) you would only
have to type it once, though capturing the pass phrase once is normaly
enough.

> > I would suggest creating a library with seperate io and gui parts in
> > order to motivate peeple in helping who do not want to support 
> > mainstream products like Windows. Like taking the PGP 3.0 lib ( is it 
> > out yet ?) and modify it a bit.
> YES (Separating UI from security functionality is also the right       
>    way to go for offering plug in security providers)

Though I think that Pronto Secure will help spread the use of cryptography
I would prefer a source code distributed library which could handle most
of the stuff needed including for example preparing encrypted requests to
key servers ( via anonymous remailers or not )( it keyservers will
implement encrypted requests ) or calculating the key distance if possible
with the available keys.

I guess one should wait for the arrival of the pgp 3.0 lib and evaluate
what it can and can't.

Greetings
 Niels Provos =8)

- - PHYSnet Rechnerverbund     PGP V2.6 Public key via finger or key server
  Niels Provos               
  Universitaet Hamburg       WWW: http://www.physnet.uni-hamburg.de/provos/   
  Jungiusstrasse 9           E-Mail: provos@wserver.physnet.uni-hamburg.de
  Germany 20355 Hamburg      Tel.:   +49 40 4123-2504     Fax: -6571 

-----BEGIN PGP SIGNATURE-----
Version: 2.6.3i
Charset: latin1

iQCVAwUBMgtHOcweILHCAJhBAQFtyQQAo+UQF3KmpAIIQ/rEh1JHHAsQUBd9k6dk
OB2lfer/dV+kDUrgpW3CDP/GdlgMIl6LCReJz6pXTA1RShQ74cdB0HokQDfytfJW
pWjHbnUcrfCmotG4KjcWw4MBJLXLbBGY0yqcmhTiOCTpLNuv52Tvtz86vOwe4yxq
ysXIXokGJpw=
=5An3
-----END PGP SIGNATURE-----