No sure what you mean with that. GnuPG does not care about
the data it processes. If you need to filter out certain
parts of a document, you need to do this with different
software but there will never be content filtering/plugins
in GnuPG. One tool - one task.

gpgsplit is a debbugging tools and not intended for message mangling. I understand what you want to achieve and recall that we already had such a discussion a couple of years ago.
IIRC, the outcome was to suggest sending individual messageses instead.

Solved. 1.41rc2 has been released and 1.4.1 is not far away

Old links are screwed up and thus not entirely shown (only
quarter directory).

Link changed. Thanks.

Sorry, this isn't the right place. Please, refer to http://
www.gnupg.org/documentation/

We now make use of SHA.
(Sorry for the badly late reply.)

Updated to 1.0.2.
(Sorry for the delay).

Ah well, please change the link from sf.net to

WinPT is still live and up to trhe latest gpg versions. It also includes card support. The problem is that the winpt at sf.net is separate from Timo's orginal project. Tehre will be a winpt directory at gnupg.org RSN.

Do you think gpg-agent and its passphrase-caching is adequate? You can even pre-store the passphrase on startup and never expire it. I think this is a much better solution. Please let me know if you still see the need for this feature given the superiority and completeness of gpg-agent.

Yes. I plan to do that. In fact that is the reason why
gpg-agent is protocol neutral and uses the so-called keygrip.

Solved in a different way

taken. however won't happen for 1.4.1

Already answered under a different entry. GnuPG uses an
hybrid approach and encodes a random session key. This
standard practice. Please read the literature or ask on
gnupg-users@gnupg.org.

Makes sense but not easy to implement. What we are thinking
of is to use a card as last resort in case a public key is
available but the secret one not.

I have been thinking about this problem and something should
be done. Either an export or a an implicit creation of a
stub subkey. Will be added to one of the next releases.

Meanwhile I added protection stuff but I am not sure whether entering the passphrase is a sufficient protection. It might be better to have a different gpg version for exporting keys.

Done

Done. I used 30 seconds by default.

This is what I was thinking as well (the alarm solution). It is different for each keyserver type since LDAP does in fact have a notion of a timeout already. In the other cases, just calling alarm() will do nicely. Do you know how portable this is to windows?

Just waiting for code

Its indeed possible but I am not sure whether we will
implement it for 1.4. It is more likely that it will be
done with the help of libgcrypt and thus sometime appear in
gpg 1.9

Anyone can do it ;-)

We know hot to solve it.

I don't think that there is a need to fix it in 1.2.

We know what to do

The code has changed and does not anymore takeup too much core.

Meanwhile keyservers are getting better (SKS) and thus it
won't be necessary anymore to upload to more than one server.

Did what we can reasonably.

Installation issues for Windows are out of scope for GnuPG.
Eventually we will provide a proper installer, though. In
the meantime the included registry settings should be seens
as documentation.

This is not a bug. Please ask at one of the mailing lists
or see www.gnupg.org for plugins we know about.

No we won't support this for various reasons. Please search
the mail archives to see why and feel free to change it if
you still can't live without.

This fix is part of 1.4.

Suggest asking for operational help on gnupg-users@gnupg.org.

Added feature for 1.4

GnuPG is not a VPN product

Yes, it is. See
http://lists.gnupg.org/pipermail/gnupg-users/2003-May/018034.html

Use the --allow-freeform-uid option when generating the user
ID. That allows you to use any characters you like.

Added to Mac OS and Windows list of supported MUAs.

In the *host* portion? Can you give an example?

Bug promptly fixed by Alex.

Waiting for PR fix report from howto author.

Link fixed.