Mailing List Archive

A few more GnuPG / NAI questions
Excuse me if these questions have been posed (and answered!) previously. I
have looked at the Mini-FAQ and other documentation etc.

I am very new to GnuPG but have used NAI PGP extensively. I think GnuPG is
FAR better than PGP because of its versatility, but have a couple of
technical queries:

a) Will GnuPG support "as standard" RSA (with or without IDEA) from Sept
2000?

b) I think a strong argument can be made for allowing a user to "get at"
(e.g. print on the screen) the decrypted session key in case of a court
order rather than having to give up the asymmetric key. A corresponding
feature could then be added to decrypt a message with a given session key.
Are there any problems with this approach and if not why hasn't it been
implemented? I've seen the November discussion on the development list RE
this and Werner seems to not like this approach - but I can't think why.
Surely it is empowering users? I appreciate that I could implement this
myself trivially, but IMHO this is such a potentially useful feature that it
should be a part of the standard distribution.

c) Are there any downsides to using ElGamal encrypt + sign keys (apart
from the lack of interoperability w/PGP users and the size of signatures...)

d) Can I create an RSA key with GnuPG????? I've got the IDEA/RSA modules
compiled and installed but can't find a way of creating an RSA key?

e) Question on key prefs(gpg --edit-key x| pref)....NAI/PGP created keys
report "S2 S3 S1" whilst GnuPG keys (of either type) report "S10 S3 H3 H2 Z2
Z1". I guess Sx refers to symmetric cipher whilst Hx is Hash and Zx is
compression....My questions on this are: Is the first item in the list the
"prefered" option? Why isn't S3+S4 included in the GnuPG list? Why does
NAI/PGP list only symmetric ciphers whereas GnuPG offers a more complete
list?


I think GnuPG is the sexiest thing since sliced bread. TIA,

Sam Simpson
Communications Analyst
-- http://www.scramdisk.clara.net/ for ScramDisk hard-drive encryption &
Delphi Crypto Components. PGP Keys available at the same site.
Re: A few more GnuPG / NAI questions [ In reply to ]
[I removed the crosspost to -devel]

On Fri, 7 Jan 2000, Simpson, Sam wrote:

> a) Will GnuPG support "as standard" RSA (with or without IDEA) from Sept
> 2000?

Sure. However I still think that DSA keys are better. No IDEA of
course - ask again in 2007(?)

> b) I think a strong argument can be made for allowing a user to "get at"
> (e.g. print on the screen) the decrypted session key in case of a court
> order rather than having to give up the asymmetric key. A corresponding
> feature could then be added to decrypt a message with a given session key.
> Are there any problems with this approach and if not why hasn't it been
> implemented? I've seen the November discussion on the development list RE
> this and Werner seems to not like this approach - but I can't think why.
> Surely it is empowering users? I appreciate that I could implement this
> myself trivially, but IMHO this is such a potentially useful feature that it
> should be a part of the standard distribution.

I still don't like to support govermental spying. It has never been
proven that decrypting a message has influenced a court decision.
Traffic analysis is much more powerful in most cases. But, I won't
debate over this theme.

You know that we received some governmental funding; and there has
never been a discussion about such a requirement!

If you need this, someone will probably be able to implement it for
you and Shashdot will get a long thread ...

> c) Are there any downsides to using ElGamal encrypt + sign keys (apart
> from the lack of interoperability w/PGP users and the size of signatures...)

Not that I am aware of except that the computation takes loner.

> d) Can I create an RSA key with GnuPG????? I've got the IDEA/RSA modules
> compiled and installed but can't find a way of creating an RSA key?

If you get back to a very early CVS revision or look at those
g10-0.0.?.tar.gz files, you might find some code. I have removed this
intentionally.

> e) Question on key prefs(gpg --edit-key x| pref)....NAI/PGP created keys
> report "S2 S3 S1" whilst GnuPG keys (of either type) report "S10 S3 H3 H2 Z2
> Z1". I guess Sx refers to symmetric cipher whilst Hx is Hash and Zx is

This means: Get the cipher algorithm by doing an intersection of your
software algorithm list with Twofish, CAST5, 3DES and use the first
match. Same for hash algorithms and compression algorithm

> compression....My questions on this are: Is the first item in the list the
> "prefered" option? Why isn't S3+S4 included in the GnuPG list? Why does

You mean S3, S1? S1 is IDEA and not supported by GnuPG so it does not
make sense to create keys which advertise that IDEA may be used.

New GnuPG keys have Blowfish again in their list, as it has turned out
that it will take some more time to resolve some problems with OpenPGP
extensions.

> NAI/PGP list only symmetric ciphers whereas GnuPG offers a more complete
> list?

GnuPG is better ;-). There are some defaults defined, so that there
is not real need for this in PGP. And PGP is not OpenPGP compliant.

> I think GnuPG is the sexiest thing since sliced bread. TIA,

Oh, I still take a knife to cut the bread :-)

--
Werner Koch at guug.de www.gnupg.org keyid 621CC013

Boycott Amazon! - http://www.gnu.org/philosophy/amazon.html
RE: A few more GnuPG / NAI questions [ In reply to ]
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

> -----Original Message-----
> From: Werner Koch [mailto:wk@gnupg.org]
> Sent: 07 January 2000 16:47
> To: s.simpson@mia.co.uk
> Subject: Re: A few more GnuPG / NAI questions
>
>
> [I removed the crosspost to -devel]
>
> On Fri, 7 Jan 2000, Simpson, Sam wrote:
>
> > a) Will GnuPG support "as standard" RSA (with or without
> IDEA) from Sept
> > 2000?
>
> Sure. However I still think that DSA keys are better.

FWIW, I agree totally - but there are still many "die hard" RSA
users that refuse to change (DUH!).

<SNIP RE: getting the decrypted session key>

> If you need this, someone will probably be able to implement it
for
> you and Shashdot will get a long thread ...

Yes, I see your point. Someone will see that you can get at the
decrypted session key and (out of ignorance...) shout about how
this is insecure :(

> > e) Question on key prefs(gpg --edit-key x|
> pref)....NAI/PGP created keys
> > report "S2 S3 S1" whilst GnuPG keys (of either type) report
> "S10 S3 H3 H2 Z2
> > Z1". I guess Sx refers to symmetric cipher whilst Hx is
> Hash and Zx is
>
> This means: Get the cipher algorithm by doing an intersection
of your
> software algorithm list with Twofish, CAST5, 3DES

3DES is implicitly tagged onto the end...

> and use the first
> match. Same for hash algorithms and compression algorithm

This makes Twofish the default algorithm if two GnuPG
implementations (e.g. S10 first in the key options!) are used?

> > compression....My questions on this are: Is the first item
> in the list the
> > "prefered" option? Why isn't S3+S4 included in the GnuPG
> list? Why does
>
> You mean S3, S1? S1 is IDEA and not supported by GnuPG so it
does not
> make sense to create keys which advertise that IDEA may be
used.

(I'm probably confused on this option but...) I meant S3 + S4 -
S3 is CAST & S4 is Blowfish - these two are supported by GnuPG
and I'd expect them to be in this list.

> New GnuPG keys have Blowfish again in their list, as it has
turned out
> that it will take some more time to resolve some problems with
OpenPGP
> extensions.

Not a key I've just created (w/v1.0.0).

> > NAI/PGP list only symmetric ciphers whereas GnuPG offers a
> more complete
> > list?
>
> GnuPG is better ;-). There are some defaults defined, so that
there
> is not real need for this in PGP.

I see.

> And PGP is not OpenPGP compliant.

Yes, so it would appear ;)


Regards,

Sam Simpson
Communications Analyst
- -- http://www.scramdisk.clara.net/ for ScramDisk hard-drive
encryption & Delphi Crypto Components. PGP Keys available at the
same site.
-----BEGIN PGP SIGNATURE-----
Version: 6.0.2ckt http://members.tripod.com/IRFaiad/

iQA/AwUBOHYdi+0ty8FDP9tPEQL+swCgnPj8HiYorZagn3DTz5t0xfTesfAAn2PK
br/1Lf5kJAB80guc2/kzLhYl
=CI4W
-----END PGP SIGNATURE-----