* PGP support @ 2001-11-12 21:48 Simon Josefsson 2001-11-13 0:03 ` Matt Armstrong 2001-11-13 15:53 ` Per Abrahamsen 0 siblings, 2 replies; 67+ messages in thread From: Simon Josefsson @ 2001-11-12 21:48 UTC (permalink / raw) -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Thanks to Sascha Lüdecke ROT13(<fnfpun@zrgn-k.qr>) we now have PGP (as in RFC 1991, not 2015/3156 which we already have) support in CVS. As you can see, the PGP armors are QPified. I suspect this isn't perfect, but I can't find a good document that says how this should work. RFC 2440 section 7 talks a little about it, but I'm not sure if that is what is used out there? -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE78EOZ8U/viFyYAJcRAm7lAKDFx/Mg3DDaEX51Ua5vJxc+UScOiwCgjdum Y6fpjEXoEa8nqCAwb5qgsGk= =JPtk -----END PGP SIGNATURE----- ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-12 21:48 PGP support Simon Josefsson @ 2001-11-13 0:03 ` Matt Armstrong 2001-11-13 1:06 ` Josh Huber ` (2 more replies) 2001-11-13 15:53 ` Per Abrahamsen 1 sibling, 3 replies; 67+ messages in thread From: Matt Armstrong @ 2001-11-13 0:03 UTC (permalink / raw) Cc: ding Simon Josefsson <jas@extundo.com> writes: > Thanks to Sascha Lüdecke ROT13(<fnfpun@zrgn-k.qr>) we now have PGP (as > in RFC 1991, not 2015/3156 which we already have) support in CVS. For the RFC ignorant, what is the difference? I think we had PGP/MIME. What is RFC 1991? > As you can see, the PGP armors are QPified. For the lingo ignorant like myself, what does QPified mean? :-) -- matt ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 0:03 ` Matt Armstrong @ 2001-11-13 1:06 ` Josh Huber 2001-11-13 9:53 ` Simon Josefsson 2001-11-17 11:12 ` Florian Weimer 2 siblings, 0 replies; 67+ messages in thread From: Josh Huber @ 2001-11-13 1:06 UTC (permalink / raw) "Matt Armstrong" <matt+dated+1008201839.c20230@lickey.com> writes: > I think we had PGP/MIME. What is RFC 1991? Apperently that's the original RFC for clearsigning messages. Although I've never seen QP encoded non-pgp-mime messages before... > For the lingo ignorant like myself, what does QPified mean? :-) Quoted printable encoding, converts those characters that MTAs like to munge, like "^From" at the start of a line when delivering into an mbox folder. ttyl, -- Josh Huber ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 0:03 ` Matt Armstrong 2001-11-13 1:06 ` Josh Huber @ 2001-11-13 9:53 ` Simon Josefsson 2001-11-13 10:47 ` Fabien Penso 2001-11-17 11:12 ` Florian Weimer 2 siblings, 1 reply; 67+ messages in thread From: Simon Josefsson @ 2001-11-13 9:53 UTC (permalink / raw) Cc: ding On Mon, 12 Nov 2001, Matt Armstrong wrote: > Simon Josefsson <jas@extundo.com> writes: > > > Thanks to Sascha Lüdecke ROT13(<fnfpun@zrgn-k.qr>) we now have PGP (as > > in RFC 1991, not 2015/3156 which we already have) support in CVS. > > For the RFC ignorant, what is the difference? > > I think we had PGP/MIME. What is RFC 1991? PGP (RFC 1991/2440) means plain (Open)PGP blobs inserted into mail without any MIME awareness at all. It is what Mailcrypt natively supports. PGP/MIME (RFC 2015/3156) is PGP with MIME awareness. Oort has supported PGP/MIME for some time, using gpg.el or Mailcrypt as the backend. With a few more menu items and commands added to Gnus, such as "Insert Public Key", I don't have to use the Mailcrypt menu hooks at all. Also, it would be nice if we could interop with those Outlook PGP plugins. It seems as if they support a MIME-aware PGP mode that isn't 2015/3156, but rather having all MIME headers etc look as if it was a unencrypted mail, and before sending, wrapping each MIME body in a PGP blob. Everyone I communicate with seem to use the stuff, so I'll probably add it. Maybe we should add a FSF-owned PGP-backend as well. The FLIM/SEMI people seem to have one, but it requires a few files that look old and non-FSF-owned. Hm.. > > As you can see, the PGP armors are QPified. > > For the lingo ignorant like myself, what does QPified mean? :-) QP is a horrible escape language for non-ASCII stuff in mail. "To QPify" means to add QP encoding in a piece of data. That's about all you want to know about QP, I think. :-) ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 9:53 ` Simon Josefsson @ 2001-11-13 10:47 ` Fabien Penso 2001-11-13 11:32 ` Simon Josefsson 0 siblings, 1 reply; 67+ messages in thread From: Fabien Penso @ 2001-11-13 10:47 UTC (permalink / raw) [-- Attachment #1: Type: text/plain, Size: 1169 bytes --] Simon on Tue, 13 Nov 2001 10:53:53 +0100 (CET) wrote: >> > Thanks to Sascha Lüdecke ROT13(<fnfpun@zrgn-k.qr>) we now have PGP (as >> > in RFC 1991, not 2015/3156 which we already have) support in CVS. >> >> For the RFC ignorant, what is the difference? >> >> I think we had PGP/MIME. What is RFC 1991? > PGP (RFC 1991/2440) means plain (Open)PGP blobs inserted into mail without > any MIME awareness at all. It is what Mailcrypt natively supports. > PGP/MIME (RFC 2015/3156) is PGP with MIME awareness. Oort has supported > PGP/MIME for some time, using gpg.el or Mailcrypt as the backend. > With a few more menu items and commands added to Gnus, such as "Insert > Public Key", I don't have to use the Mailcrypt menu hooks at all. Good ! Mailcrypt doesn't support sig with international characters (which sux). If Gnus now does, I'll be able to use plain text sign for my usenet posts. > Also, it would be nice if we could interop with those Outlook PGP plugins. Yeap. I agree. -- Fabien Penso <penso@linuxfr.org> | LinuxFr a toujours besoin de : http://perso.LinuxFR.org/penso/ | http://linuxfr.org/dons/ [-- Attachment #2: Type: application/pgp-signature, Size: 239 bytes --] ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 10:47 ` Fabien Penso @ 2001-11-13 11:32 ` Simon Josefsson 0 siblings, 0 replies; 67+ messages in thread From: Simon Josefsson @ 2001-11-13 11:32 UTC (permalink / raw) Cc: ding On Tue, 13 Nov 2001, Fabien Penso wrote: > > PGP (RFC 1991/2440) means plain (Open)PGP blobs inserted into mail without > > any MIME awareness at all. It is what Mailcrypt natively supports. > > PGP/MIME (RFC 2015/3156) is PGP with MIME awareness. Oort has supported > > PGP/MIME for some time, using gpg.el or Mailcrypt as the backend. > > > With a few more menu items and commands added to Gnus, such as "Insert > > Public Key", I don't have to use the Mailcrypt menu hooks at all. > > Good ! Mailcrypt doesn't support sig with international characters > (which sux). If Gnus now does, I'll be able to use plain text sign for > my usenet posts. The PGP/MIME standard allows non-ASCII, so this should already work. If people like PGP and/or MIME on the UseNet is a different question though, and the answer seems to depend on what hierarchy you are in. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 0:03 ` Matt Armstrong 2001-11-13 1:06 ` Josh Huber 2001-11-13 9:53 ` Simon Josefsson @ 2001-11-17 11:12 ` Florian Weimer 2001-11-17 11:27 ` Simon Josefsson 2 siblings, 1 reply; 67+ messages in thread From: Florian Weimer @ 2001-11-17 11:12 UTC (permalink / raw) Cc: ding "Matt Armstrong" <matt+dated+1008201839.c20230@lickey.com> writes: > Simon Josefsson <jas@extundo.com> writes: > >> Thanks to Sascha Lüdecke ROT13(<fnfpun@zrgn-k.qr>) we now have PGP (as >> in RFC 1991, not 2015/3156 which we already have) support in CVS. > > For the RFC ignorant, what is the difference? > > I think we had PGP/MIME. What is RFC 1991? Is the predecessor of RFC 2440 (OpenPGP). It hasn't got much to do with Internet mail or news, so it's a red herring. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-17 11:12 ` Florian Weimer @ 2001-11-17 11:27 ` Simon Josefsson 0 siblings, 0 replies; 67+ messages in thread From: Simon Josefsson @ 2001-11-17 11:27 UTC (permalink / raw) Cc: ding Florian Weimer <fw@deneb.enyo.de> writes: > "Matt Armstrong" <matt+dated+1008201839.c20230@lickey.com> writes: > >> Simon Josefsson <jas@extundo.com> writes: >> >>> Thanks to Sascha Lüdecke ROT13(<fnfpun@zrgn-k.qr>) we now have PGP (as >>> in RFC 1991, not 2015/3156 which we already have) support in CVS. >> >> For the RFC ignorant, what is the difference? >> >> I think we had PGP/MIME. What is RFC 1991? > > Is the predecessor of RFC 2440 (OpenPGP). It hasn't got much to do > with Internet mail or news, so it's a red herring. Unfortunately it is the only thing that Outlook PGP plugins seem to grok... (At least if you regard 1991 and 2440 as almost the same thing, which, from Gnus' point of view, they are.) ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-12 21:48 PGP support Simon Josefsson 2001-11-13 0:03 ` Matt Armstrong @ 2001-11-13 15:53 ` Per Abrahamsen 2001-11-13 18:38 ` Simon Josefsson ` (2 more replies) 1 sibling, 3 replies; 67+ messages in thread From: Per Abrahamsen @ 2001-11-13 15:53 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > As you can see, the PGP armors are QPified. Actually, I can't. Probably because of X-MIME-Autoconverted: from quoted-printable to 8bit by sheridan.dina.kvl.dk id WAA23581 Would such convertions affect PGP? ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 15:53 ` Per Abrahamsen @ 2001-11-13 18:38 ` Simon Josefsson 2001-11-14 10:03 ` Per Abrahamsen 2001-11-13 18:52 ` Josh Huber 2001-11-17 11:14 ` Florian Weimer 2 siblings, 1 reply; 67+ messages in thread From: Simon Josefsson @ 2001-11-13 18:38 UTC (permalink / raw) Cc: ding Per Abrahamsen <abraham@dina.kvl.dk> writes: > Simon Josefsson <jas@extundo.com> writes: > >> As you can see, the PGP armors are QPified. > > Actually, I can't. Probably because of > > X-MIME-Autoconverted: from quoted-printable to 8bit by sheridan.dina.kvl.dk id WAA23581 > > Would such convertions affect PGP? It shouldn't, but it might. Could you verify the PGP signature? ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 18:38 ` Simon Josefsson @ 2001-11-14 10:03 ` Per Abrahamsen 0 siblings, 0 replies; 67+ messages in thread From: Per Abrahamsen @ 2001-11-14 10:03 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > It shouldn't, but it might. Could you verify the PGP signature? I have no idea. I have never used PGP on Unix, and neither "W g" nor "W s" does anything visible. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 15:53 ` Per Abrahamsen 2001-11-13 18:38 ` Simon Josefsson @ 2001-11-13 18:52 ` Josh Huber 2001-11-13 20:43 ` Matt Armstrong 2001-11-15 0:52 ` PGP support Sascha Lüdecke 2001-11-17 11:14 ` Florian Weimer 2 siblings, 2 replies; 67+ messages in thread From: Josh Huber @ 2001-11-13 18:52 UTC (permalink / raw) Per Abrahamsen <abraham@dina.kvl.dk> writes: > Actually, I can't. Probably because of > > X-MIME-Autoconverted: from quoted-printable to 8bit by > sheridan.dina.kvl.dk id WAA23581 > > Would such convertions affect PGP? Well, did the message verify for you? I suspect that the message is signed first, and then encoded QP, which would explain the QP encoding of the PGP headers. FWIW mutt does not do this. Does Outhouse Express support PGP signed messages that are QP encoded like this? ttyl, -- Josh Huber ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 18:52 ` Josh Huber @ 2001-11-13 20:43 ` Matt Armstrong 2001-11-13 21:18 ` Simon Josefsson 2001-11-15 0:52 ` PGP support Sascha Lüdecke 1 sibling, 1 reply; 67+ messages in thread From: Matt Armstrong @ 2001-11-13 20:43 UTC (permalink / raw) Josh Huber <huber@alum.wpi.edu> writes: > Well, did the message verify for you? > > I suspect that the message is signed first, and then encoded QP, which > would explain the QP encoding of the PGP headers. > > FWIW mutt does not do this. Does Outhouse Express support PGP > signed messages that are QP encoded like this? I didn't even realize that Simon's original post was signed. Is Gnus supposed give some kind of visual cue that the message was signed? I have this setup in .gnus, where my-save-require catches any require errors and just returns nil -- but in this case 'gpg was successfully loaded. (cond ((my-safe-require 'gpg) (setq mml2015-use 'gpg) (setq gpg-temp-directory (expand-file-name "~/.gnupg/tmp")) (setq gpg-command-default-alist (quote ((gpg . "gpg") (gpg-2comp . "gpg")))))) -- matt ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 20:43 ` Matt Armstrong @ 2001-11-13 21:18 ` Simon Josefsson 2001-11-13 21:33 ` Matt Armstrong ` (3 more replies) 0 siblings, 4 replies; 67+ messages in thread From: Simon Josefsson @ 2001-11-13 21:18 UTC (permalink / raw) "Matt Armstrong" <matt+dated+1008276238.4e3562@lickey.com> writes: > Josh Huber <huber@alum.wpi.edu> writes: > >> Well, did the message verify for you? >> >> I suspect that the message is signed first, and then encoded QP, which >> would explain the QP encoding of the PGP headers. >> >> FWIW mutt does not do this. Does Outhouse Express support PGP >> signed messages that are QP encoded like this? > > I didn't even realize that Simon's original post was signed. Is Gnus > supposed give some kind of visual cue that the message was signed? There is a `p' in the Article buffer's modeline. Using the fringe for this is one idea. You can also get big buttons if you like to clutter up your article buffer: (setq gnus-buttonized-mime-types '("multipart/encrypted" "multipart/signed")) ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 21:18 ` Simon Josefsson @ 2001-11-13 21:33 ` Matt Armstrong 2001-11-13 21:52 ` Simon Josefsson 2001-11-14 10:50 ` Per Abrahamsen ` (2 subsequent siblings) 3 siblings, 1 reply; 67+ messages in thread From: Matt Armstrong @ 2001-11-13 21:33 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > "Matt Armstrong" <matt+dated+1008276238.4e3562@lickey.com> writes: > >> I didn't even realize that Simon's original post was signed. Is Gnus >> supposed give some kind of visual cue that the message was signed? > > There is a `p' in the Article buffer's modeline. Using the fringe for > this is one idea. Yes, a single 'p' is not very obvious. ;-) Also, without buttons, how are you supposed to verify the signatures? I saw no obvious way -- W p didn't do much. > You can also get big buttons if you like to clutter up your article > buffer: > > (setq gnus-buttonized-mime-types > '("multipart/encrypted" "multipart/signed")) Ahh, cool, there is the visual cue I was looking for, as well as the UI necessary to see gpg's output, etc. Perhaps this should be the default? -- matt ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 21:33 ` Matt Armstrong @ 2001-11-13 21:52 ` Simon Josefsson 2001-11-14 10:07 ` Per Abrahamsen ` (3 more replies) 0 siblings, 4 replies; 67+ messages in thread From: Simon Josefsson @ 2001-11-13 21:52 UTC (permalink / raw) "Matt Armstrong" <matt+dated+1008279199.7e6a5d@lickey.com> writes: > Yes, a single 'p' is not very obvious. ;-) > > Also, without buttons, how are you supposed to verify the signatures? > I saw no obvious way -- W p didn't do much. Try `W s'. Oops, this wasn't documented. Fixed. >> You can also get big buttons if you like to clutter up your article >> buffer: >> >> (setq gnus-buttonized-mime-types >> '("multipart/encrypted" "multipart/signed")) > > Ahh, cool, there is the visual cue I was looking for, as well as the > UI necessary to see gpg's output, etc. Perhaps this should be the > default? It was the default, but perhaps it interfered with reading. If we can find a non-obtrusive mechanism to let the user know the this information, it would be good. Also, the buttons create false security. I posted a mail to this list that showed PGP buttons that said the message verified OK, but I created them using ^H which Gnus bolidifies. Whatever the mechanism will be, the article buffer is the wrong place for it. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 21:52 ` Simon Josefsson @ 2001-11-14 10:07 ` Per Abrahamsen 2001-11-14 10:59 ` dme ` (2 subsequent siblings) 3 siblings, 0 replies; 67+ messages in thread From: Per Abrahamsen @ 2001-11-14 10:07 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > It was the default, but perhaps it interfered with reading. Yes, one requirement for signing to be popular is that it doesn't annoy people who don't care about it. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 21:52 ` Simon Josefsson 2001-11-14 10:07 ` Per Abrahamsen @ 2001-11-14 10:59 ` dme 2001-11-14 11:52 ` Simon Josefsson 2001-11-16 18:26 ` news 2001-11-17 11:15 ` Florian Weimer 3 siblings, 1 reply; 67+ messages in thread From: dme @ 2001-11-14 10:59 UTC (permalink / raw) * jas@extundo.com [2001-11-13 21:52:16] > Also, the buttons create false security. I posted a mail to this > list that showed PGP buttons that said the message verified OK, but > I created them using ^H which Gnus bolidifies. Whatever the > mechanism will be, the article buffer is the wrong place for it. How about a background pixmap which indicates the verification state of text ? dme. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 10:59 ` dme @ 2001-11-14 11:52 ` Simon Josefsson 2001-11-14 16:02 ` Andreas Fuchs ` (2 more replies) 0 siblings, 3 replies; 67+ messages in thread From: Simon Josefsson @ 2001-11-14 11:52 UTC (permalink / raw) Cc: ding On Wed, 14 Nov 2001 dme@dme.org wrote: > * jas@extundo.com [2001-11-13 21:52:16] > > Also, the buttons create false security. I posted a mail to this > > list that showed PGP buttons that said the message verified OK, but > > I created them using ^H which Gnus bolidifies. Whatever the > > mechanism will be, the article buffer is the wrong place for it. > > How about a background pixmap which indicates the verification state > of text ? Then I'll just send a text/html mail with the pixmap as background. The article buffer can't be used securely for this purpose. The modeline is perhaps a good place, if we can get a nice icon that says "Signature Good", "Signature Bad", "Signature OK but sender unknown/untrusted". Any takers? ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 11:52 ` Simon Josefsson @ 2001-11-14 16:02 ` Andreas Fuchs 2001-11-14 17:11 ` Simon Josefsson 2001-11-17 11:18 ` Florian Weimer 2001-11-17 11:17 ` Florian Weimer 2001-12-29 5:14 ` Lars Magne Ingebrigtsen 2 siblings, 2 replies; 67+ messages in thread From: Andreas Fuchs @ 2001-11-14 16:02 UTC (permalink / raw) [-- Attachment #1.1: Type: text/plain, Size: 608 bytes --] Today, Simon Josefsson <jas@extundo.com> wrote: > The modeline is perhaps a good place, if we can get a nice icon that > says "Signature Good", "Signature Bad", "Signature OK but sender > unknown/untrusted". Any takers? (playing devil's advocate:) What about more than one part? Also, an evil one could just go ahead and sign an empty line of text (with an untrusted key). Then, I presume, the user will see the "OK but untrusted" message. I would just die to see the fringe be used for this (or an overlay or whatever, for xemacs). -- Andreas Fuchs, <asf@acm.org>, asf@jabber.at, antifuchs [-- Attachment #2: Type: application/pgp-signature, Size: 231 bytes --] ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 16:02 ` Andreas Fuchs @ 2001-11-14 17:11 ` Simon Josefsson 2001-11-14 17:34 ` Nevin Kapur 2001-11-17 11:18 ` Florian Weimer 1 sibling, 1 reply; 67+ messages in thread From: Simon Josefsson @ 2001-11-14 17:11 UTC (permalink / raw) Andreas Fuchs <asf@void.at> writes: > Today, Simon Josefsson <jas@extundo.com> wrote: >> The modeline is perhaps a good place, if we can get a nice icon that >> says "Signature Good", "Signature Bad", "Signature OK but sender >> unknown/untrusted". Any takers? > > (playing devil's advocate:) What about more than one part? Also, an evil > one could just go ahead and sign an empty line of text (with an > untrusted key). Then, I presume, the user will see the "OK but > untrusted" message. Right. Not good. I haven't seen any attempt at solving this problem though. Partially signed MIME bodies are too complex to comprehend for users that aren't familiar with how things are implemented. > I would just die to see the fringe be used for this (or an overlay > or whatever, for xemacs). Yes. However, the more I think about it, it is not perfect either. I could send a (signed) text/html part that includes a cid: URL pointing to another (unsigned) MIME part containing a gif picture of a Gnus article buffer with some text in. The fringe would then most likely be highlighted for the lines where the image is (because they could contain signed text as well), which would be incorrect. It becomes even more problematic if you consider charset-splitted MIME bodies. How could you usefully highlight that some characters on a line are digitally signed and the rest of the line wasn't? We also have the problem of giving a false impression of what is digitally signed or encrypted -- the RFC 2822 headers never are. It seems as if there is something fundamentally flawed here (perhaps my reasoning :)), and we should give up. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 17:11 ` Simon Josefsson @ 2001-11-14 17:34 ` Nevin Kapur 2001-11-14 17:57 ` Matt Armstrong ` (2 more replies) 0 siblings, 3 replies; 67+ messages in thread From: Nevin Kapur @ 2001-11-14 17:34 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > It seems as if there is something fundamentally flawed here (perhaps > my reasoning :)), and we should give up. How about adding something that changes each time the article is displayed? What if the current time was added to the button? *[[PGP Signed Part: XYZ <xyz@abc.xom> "Wed Nov 14 12:34:04 2001"]]* -- Nevin ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 17:34 ` Nevin Kapur @ 2001-11-14 17:57 ` Matt Armstrong 2001-11-17 11:19 ` Florian Weimer 2001-11-18 7:23 ` Paul Jarc 2 siblings, 0 replies; 67+ messages in thread From: Matt Armstrong @ 2001-11-14 17:57 UTC (permalink / raw) Nevin Kapur <nevin@jhu.edu> writes: > Simon Josefsson <jas@extundo.com> writes: > >> It seems as if there is something fundamentally flawed here >> (perhaps my reasoning :)), and we should give up. > > How about adding something that changes each time the article is > displayed? What if the current time was added to the button? > > *[[PGP Signed Part: XYZ <xyz@abc.xom> "Wed Nov 14 12:34:04 2001"]]* I like that idea. I think using buttons is the way to go. I'm thinking: - Inserting a "verified on TIME/DATE" into the button is a good idea, and greatly hinders people's ability to spoof it. - However, I'd want to be able to turn this off to save some screen real estate. I am not personally worried about button spoofing. In the cases where I truly care I'll verify the button is real by either moving the mouse over it (when under a GUI) or selecting the button to see the raw gpg output. - Buttons work both under a GUI and when emacs is running in a terminal. -- matt ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 17:34 ` Nevin Kapur 2001-11-14 17:57 ` Matt Armstrong @ 2001-11-17 11:19 ` Florian Weimer 2001-11-18 7:23 ` Paul Jarc 2 siblings, 0 replies; 67+ messages in thread From: Florian Weimer @ 2001-11-17 11:19 UTC (permalink / raw) Nevin Kapur <nevin@jhu.edu> writes: > How about adding something that changes each time the article is > displayed? What if the current time was added to the button? > > *[[PGP Signed Part: XYZ <xyz@abc.xom> "Wed Nov 14 12:34:04 2001"]]* Ah, yes, that's the way to go! I should have read the entire thread first. (mutt does the same thing, BTW, and it includes the complete GnuPG output by default.) ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 17:34 ` Nevin Kapur 2001-11-14 17:57 ` Matt Armstrong 2001-11-17 11:19 ` Florian Weimer @ 2001-11-18 7:23 ` Paul Jarc 2001-11-18 10:53 ` Andreas Fuchs 2 siblings, 1 reply; 67+ messages in thread From: Paul Jarc @ 2001-11-18 7:23 UTC (permalink / raw) Nevin Kapur <nevin@jhu.edu> wrote: > How about adding something that changes each time the article is > displayed? What if the current time was added to the button? > > *[[PGP Signed Part: XYZ <xyz@abc.xom> "Wed Nov 14 12:34:04 2001"]]* Another idea: when the signature is checked, modify the summary line - add a mark or something. That can't be faked even if articles can dynamically include content from a URL embedded in the article, which I imagine will probably happen at some point, if it's not already possible. paul ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-18 7:23 ` Paul Jarc @ 2001-11-18 10:53 ` Andreas Fuchs 2001-11-18 20:09 ` Raymond Scholz 0 siblings, 1 reply; 67+ messages in thread From: Andreas Fuchs @ 2001-11-18 10:53 UTC (permalink / raw) [-- Attachment #1.1: Type: text/plain, Size: 993 bytes --] Today, Paul Jarc <prj@po.cwru.edu> wrote: > Another idea: when the signature is checked, modify the summary line - > add a mark or something. That can't be faked even if articles can > dynamically include content from a URL embedded in the article, which > I imagine will probably happen at some point, if it's not already > possible. Another idea I just got would be some kind of MIME part tree, which shows signature state. That way, you could even show the tree-like structure of multipart-message attachments which themselves contain multipart messages. Something looking like this (inspired by speedbar): [-] Part 1 [Message] (signed) [+] Part 1 [Message] (signed, failed) [ ] foobar.gif [Image] (signed) [+] Part 2 [Message] (signed, untrusted) I can imagine that this could only be compromised by a w3 page with frames. Maybe I'll hack something together, but don't hold your breath yet. (-: -- Andreas Fuchs, <asf@acm.org>, asf@jabber.at, antifuchs [-- Attachment #2: Type: application/pgp-signature, Size: 231 bytes --] ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-18 10:53 ` Andreas Fuchs @ 2001-11-18 20:09 ` Raymond Scholz 2001-11-18 23:38 ` Andreas Fuchs 0 siblings, 1 reply; 67+ messages in thread From: Raymond Scholz @ 2001-11-18 20:09 UTC (permalink / raw) Andreas Fuchs <asf@void.at> writes: > Something looking like this (inspired by speedbar): > > [-] Part 1 [Message] (signed) > [+] Part 1 [Message] (signed, failed) > [ ] foobar.gif [Image] (signed) > [+] Part 2 [Message] (signed, untrusted) > > I can imagine that this could only be compromised by a w3 page with > frames. Maybe I'll hack something together, but don't hold your breath > yet. (-: Hint: looks very similar to the output you get if pressing `C-d' on a MIME message. Cheers, Ray -- WYSIWYG is a step backwards. Human labor is used to do that which the computer can do better. (Andrew S. Tanenbaum) ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-18 20:09 ` Raymond Scholz @ 2001-11-18 23:38 ` Andreas Fuchs 0 siblings, 0 replies; 67+ messages in thread From: Andreas Fuchs @ 2001-11-18 23:38 UTC (permalink / raw) [-- Attachment #1: Type: text/plain, Size: 613 bytes --] On 2001-11-18, Raymond Scholz <ray-2001@zonix.de> wrote: >> I can imagine that this could only be compromised by a w3 page with >> frames. Maybe I'll hack something together, but don't hold your >> breath yet. (-: > > Hint: looks very similar to the output you get if pressing `C-d' on a > MIME message. You're right. Been too long since I used that particular feature. Now, if I could just display this buffer beside the message buffer (and still have it convey the information I pointed out above), and we'd be half done (-: Thanks, -- Andreas Fuchs, <asf@acm.org>, asf@jabber.at, antifuchs [-- Attachment #2: Type: application/pgp-signature, Size: 231 bytes --] ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 16:02 ` Andreas Fuchs 2001-11-14 17:11 ` Simon Josefsson @ 2001-11-17 11:18 ` Florian Weimer 1 sibling, 0 replies; 67+ messages in thread From: Florian Weimer @ 2001-11-17 11:18 UTC (permalink / raw) Andreas Fuchs <asf@void.at> writes: > (playing devil's advocate:) What about more than one part? Also, an evil > one could just go ahead and sign an empty line of text (with an > untrusted key). Then, I presume, the user will see the "OK but > untrusted" message. I would just die to see the fringe be used for this > (or an overlay or whatever, for xemacs). Works only on X11, I think. We have to include the necessary information in the message body itself, with the current time. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 11:52 ` Simon Josefsson 2001-11-14 16:02 ` Andreas Fuchs @ 2001-11-17 11:17 ` Florian Weimer 2001-12-29 5:14 ` Lars Magne Ingebrigtsen 2 siblings, 0 replies; 67+ messages in thread From: Florian Weimer @ 2001-11-17 11:17 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > The modeline is perhaps a good place, if we can get a nice icon that says > "Signature Good", "Signature Bad", "Signature OK but sender > unknown/untrusted". Any takers? Some people run GNU Emacs inside GNU screen: I don't think they have got icons in the modeline. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 11:52 ` Simon Josefsson 2001-11-14 16:02 ` Andreas Fuchs 2001-11-17 11:17 ` Florian Weimer @ 2001-12-29 5:14 ` Lars Magne Ingebrigtsen 2 siblings, 0 replies; 67+ messages in thread From: Lars Magne Ingebrigtsen @ 2001-12-29 5:14 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > The modeline is perhaps a good place, if we can get a nice icon that says > "Signature Good", "Signature Bad", "Signature OK but sender > unknown/untrusted". Any takers? For Emacs 21, the fringe could display a symbol to the left (or right) of each part that gives you the status of the part. For instance, use the same symbol Mozilla does -- that lock thingie. Broken when bad, etc. For fringe-less environments, other methods might be used. -- (domestic pets only, the antidote for overdose, milk.) larsi@gnus.org * Lars Magne Ingebrigtsen ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 21:52 ` Simon Josefsson 2001-11-14 10:07 ` Per Abrahamsen 2001-11-14 10:59 ` dme @ 2001-11-16 18:26 ` news 2001-11-16 18:43 ` Werner Koch 2001-11-17 4:28 ` Andreas Fuchs 2001-11-17 11:15 ` Florian Weimer 3 siblings, 2 replies; 67+ messages in thread From: news @ 2001-11-16 18:26 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > "Matt Armstrong" <matt+dated+1008279199.7e6a5d@lickey.com> writes: > >> Yes, a single 'p' is not very obvious. ;-) > > It was the default, but perhaps it interfered with reading. If we can > find a non-obtrusive mechanism to let the user know the this > information, it would be good. > > Also, the buttons create false security. I posted a mail to this list > that showed PGP buttons that said the message verified OK, but I > created them using ^H which Gnus bolidifies. Whatever the mechanism > will be, the article buffer is the wrong place for it. How about a header field: X-Gnus: pgp-verified signature with key 0x454545 valid untrusted Chris ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 18:26 ` news @ 2001-11-16 18:43 ` Werner Koch 2001-11-16 19:20 ` Simon Josefsson 2001-11-17 4:28 ` Andreas Fuchs 1 sibling, 1 reply; 67+ messages in thread From: Werner Koch @ 2001-11-16 18:43 UTC (permalink / raw) On Fri, 16 Nov 2001 13:26:20 -0500, Chris Beggy said: > How about a header field: > X-Gnus: pgp-verified signature with key 0x454545 valid untrusted BTW, I might have missed it but an important feature would be to add a line like: X-Gnus-Orig-Encrypted-To: 0x12345678, 0x34567890 So that one can see that the message was originally encrypted and even more important to automagically suggest to encrypt any reply. Ciao, Werner ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 18:43 ` Werner Koch @ 2001-11-16 19:20 ` Simon Josefsson 2001-11-16 19:37 ` Josh Huber ` (3 more replies) 0 siblings, 4 replies; 67+ messages in thread From: Simon Josefsson @ 2001-11-16 19:20 UTC (permalink / raw) Cc: ding Werner Koch <wk@gnupg.org> writes: > On Fri, 16 Nov 2001 13:26:20 -0500, Chris Beggy said: > >> How about a header field: > >> X-Gnus: pgp-verified signature with key 0x454545 valid untrusted > > BTW, I might have missed it but an important feature would be to add a > line like: > > X-Gnus-Orig-Encrypted-To: 0x12345678, 0x34567890 > > So that one can see that the message was originally encrypted and even > more important to automagically suggest to encrypt any reply. A header to say that replies should be encrypted seems useful, but it isn't really Gnus specific, and I would also feel better if requests for security are signed (i.e. in S/MIME SignerInfo, I don't know if OpenPGP has something similar). Hm. I don't understand the value of having a header line that says the message was originally encrypted, the client already knows this? And users should probably not trust such headers without the client saying it is OK, and if the client can do that, the client could use some other (better) way of conveying this information anyway. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 19:20 ` Simon Josefsson @ 2001-11-16 19:37 ` Josh Huber 2001-11-16 20:40 ` Simon Josefsson 2001-11-17 9:47 ` Fabien Penso 2001-11-16 20:18 ` news ` (2 subsequent siblings) 3 siblings, 2 replies; 67+ messages in thread From: Josh Huber @ 2001-11-16 19:37 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > I don't understand the value of having a header line that says the > message was originally encrypted, the client already knows this? Right...is there an easy way to access this information (if the message was signed or encrypted) from a reply buffer? I'm thinking about a way to implement the reply_sign, reply_encrypt, and reply_signencrypt options of mutt...If you get a message which is encrypted, and reply to it, Gnus would automatically encrypt the reply. -- Josh Huber ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 19:37 ` Josh Huber @ 2001-11-16 20:40 ` Simon Josefsson 2001-11-16 20:58 ` Jack Twilley ` (2 more replies) 2001-11-17 9:47 ` Fabien Penso 1 sibling, 3 replies; 67+ messages in thread From: Simon Josefsson @ 2001-11-16 20:40 UTC (permalink / raw) Josh Huber <huber@alum.wpi.edu> writes: > Simon Josefsson <jas@extundo.com> writes: > >> I don't understand the value of having a header line that says the >> message was originally encrypted, the client already knows this? > > Right...is there an easy way to access this information (if the > message was signed or encrypted) from a reply buffer? I don't think so... someone please contribute it. :-) > I'm thinking about a way to implement the reply_sign, reply_encrypt, > and reply_signencrypt options of mutt...If you get a message which is > encrypted, and reply to it, Gnus would automatically encrypt the > reply. That would be cool. Perhaps the idea could be integrated with BBDB as well (or maybe it already is, I think I remember something similar). ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 20:40 ` Simon Josefsson @ 2001-11-16 20:58 ` Jack Twilley 2001-11-16 21:21 ` Josh Huber 2001-11-17 9:50 ` Fabien Penso 2 siblings, 0 replies; 67+ messages in thread From: Jack Twilley @ 2001-11-16 20:58 UTC (permalink / raw) >>>>> "Josh" == Josh Huber <huber@alum.wpi.edu> writes: Josh> I'm thinking about a way to implement the reply_sign, Josh> reply_encrypt, and reply_signencrypt options of mutt...If you Josh> get a message which is encrypted, and reply to it, Gnus would Josh> automatically encrypt the reply. >>>>> "Simon" == Simon Josefsson <jas@extundo.com> writes: Simon> That would be cool. Perhaps the idea could be integrated with Simon> BBDB as well (or maybe it already is, I think I remember Simon> something similar). bbdb-pgp is an addon. When you C-c C-c, it checks the BBDB record of the addressee. If their record has the field 'pgp-mail' set to 'sign', it asks you if you want to sign the message. If their record has the field 'pgp-mail' set to 'encrypt', it asks you if you want to encrypt the message. As distributed, it's mailcrypt-specific. The changes to work with mml-secure are very very obvious one-liners. Jack. (I use it all the time) -- Jack Twilley jmt at twilley dot org http colon slash slash www dot twilley dot org slash tilde jmt slash ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 20:40 ` Simon Josefsson 2001-11-16 20:58 ` Jack Twilley @ 2001-11-16 21:21 ` Josh Huber 2001-11-16 21:38 ` Jack Twilley 2001-11-17 9:50 ` Fabien Penso 2 siblings, 1 reply; 67+ messages in thread From: Josh Huber @ 2001-11-16 21:21 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > That would be cool. Perhaps the idea could be integrated with BBDB > as well (or maybe it already is, I think I remember something > similar). Yeah, I use a partially hacked version of bbdb-pgp.el to do this. But, this just lets you say whether or not to /always/ sign or encrypt to a certain person. This works well, but if someone (say, not in your bbdb) sends you encrypted mail, it's nice to return the favor :) ttyl, -- Josh Huber ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 21:21 ` Josh Huber @ 2001-11-16 21:38 ` Jack Twilley 2001-11-17 9:52 ` Fabien Penso 2001-11-17 11:23 ` Florian Weimer 0 siblings, 2 replies; 67+ messages in thread From: Jack Twilley @ 2001-11-16 21:38 UTC (permalink / raw) >>>>> "Josh" == Josh Huber <huber@alum.wpi.edu> writes: [...] Josh> But, this just lets you say whether or not to /always/ sign or Josh> encrypt to a certain person. This works well, but if someone Josh> (say, not in your bbdb) sends you encrypted mail, it's nice to Josh> return the favor :) What *I*'d like, personally, is to get that gpg-ring.el software working, and then every week traverse my gpg key ring, and create bbdb entries with pgp-mail settings for each key I've got. Josh> ttyl, Josh> -- Josh Huber Jack. (so whenever someone owns and fixes gpg-ring, I'll be ecstatic) -- Jack Twilley jmt at twilley dot org http colon slash slash www dot twilley dot org slash tilde jmt slash ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 21:38 ` Jack Twilley @ 2001-11-17 9:52 ` Fabien Penso 2001-11-17 10:08 ` Jack Twilley 2001-11-17 11:23 ` Florian Weimer 1 sibling, 1 reply; 67+ messages in thread From: Fabien Penso @ 2001-11-17 9:52 UTC (permalink / raw) Cc: ding [-- Attachment #1: Type: text/plain, Size: 800 bytes --] Jack on Fri, 16 Nov 2001 13:38:54 -0800 wrote: >>>>>> "Josh" == Josh Huber <huber@alum.wpi.edu> writes: > [...] Josh> But, this just lets you say whether or not to /always/ sign or Josh> encrypt to a certain person. This works well, but if someone Josh> (say, not in your bbdb) sends you encrypted mail, it's nice to Josh> return the favor :) > What *I*'d like, personally, is to get that gpg-ring.el software > working, and then every week traverse my gpg key ring, and create bbdb > entries with pgp-mail settings for each key I've got. Is this the solution ? You would hen get lots of bbdb entries when everyone will use GPG :p -- Fabien Penso <penso@linuxfr.org> | LinuxFr a toujours besoin de : http://perso.LinuxFR.org/penso/ | http://linuxfr.org/dons/ [-- Attachment #2: Type: application/pgp-signature, Size: 239 bytes --] ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-17 9:52 ` Fabien Penso @ 2001-11-17 10:08 ` Jack Twilley 0 siblings, 0 replies; 67+ messages in thread From: Jack Twilley @ 2001-11-17 10:08 UTC (permalink / raw) Jack> What *I*'d like, personally, is to get that gpg-ring.el software Jack> working, and then every week traverse my gpg key ring, and Jack> create bbdb entries with pgp-mail settings for each key I've Jack> got. >>>>> "Fabien" == Fabien Penso <penso@linuxfr.org> writes: Fabien> Is this the solution ? You would hen get lots of bbdb entries Fabien> when everyone will use GPG :p I already have nearly six thousand entries. It's all good. :-) Jack. (bbdb good, gpg good, gnus good, all three triple good) -- Jack Twilley jmt at twilley dot org http colon slash slash www dot twilley dot org slash tilde jmt slash ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 21:38 ` Jack Twilley 2001-11-17 9:52 ` Fabien Penso @ 2001-11-17 11:23 ` Florian Weimer 2001-11-17 11:10 ` Jack Twilley 1 sibling, 1 reply; 67+ messages in thread From: Florian Weimer @ 2001-11-17 11:23 UTC (permalink / raw) Jack Twilley <jmt+usenet@twilley.org> writes: > What *I*'d like, personally, is to get that gpg-ring.el software > working, and then every week traverse my gpg key ring, and create bbdb > entries with pgp-mail settings for each key I've got. I'm sorry that I don't finish gpg-ring.el, but there's the risk that RMS decides to throw the code away and tell someone who has signed papers to start from scratch. That's why I'm not working on this anymore. (Yes, this has happened to me before.) ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-17 11:23 ` Florian Weimer @ 2001-11-17 11:10 ` Jack Twilley 2001-11-17 11:54 ` Per Abrahamsen 2001-11-17 14:39 ` Florian Weimer 0 siblings, 2 replies; 67+ messages in thread From: Jack Twilley @ 2001-11-17 11:10 UTC (permalink / raw) >>>>> "Florian" == Florian Weimer <fw@deneb.enyo.de> writes: Jack> What *I*'d like, personally, is to get that gpg-ring.el software Jack> working, and then every week traverse my gpg key ring, and Jack> create bbdb entries with pgp-mail settings for each key I've Jack> got. Florian> I'm sorry that I don't finish gpg-ring.el, but there's the Florian> risk that RMS decides to throw the code away and tell someone Florian> who has signed papers to start from scratch. That's why I'm Florian> not working on this anymore. RMS can't throw away stuff that's part of Gnus, can he? I thought that was up to the maintainers of Gnus -- the only thing RMS could do AFAIK is just not allow Gnus to be part of the GNU Emacs distribution until Lars bends to his whim. Am I misunderstanding? Besides, if you finish working on it, then dozens of people can actually use it, copyleft notwithstanding. Isn't that good? Florian> (Yes, this has happened to me before.) I'm sorry that happened. Bullshit politics suck all around. Jack. -- Jack Twilley jmt at twilley dot org http colon slash slash www dot twilley dot org slash tilde jmt slash ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-17 11:10 ` Jack Twilley @ 2001-11-17 11:54 ` Per Abrahamsen 2001-11-17 14:39 ` Florian Weimer 1 sibling, 0 replies; 67+ messages in thread From: Per Abrahamsen @ 2001-11-17 11:54 UTC (permalink / raw) Jack Twilley <jmt+usenet@twilley.org> writes: > RMS can't throw away stuff that's part of Gnus, can he? Nobody can "throw away" stuff that has been released as free software. However, most Gnus users are using the version bundled with Emacs, and at least some Gnus devlopers (like me) for that reason among several really don't see the point of working on a Gnus version that cannot be bundled with Emacs. So code for which there is no signed papers live in "contrib", and Gnus proper cannot depend on it, and can't be enabled by default. Which mean it become a lot less useful than code with papers. Which again means that it will be a huge benefit for everyone if someone who will and can sign papers creates a replacement. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-17 11:10 ` Jack Twilley 2001-11-17 11:54 ` Per Abrahamsen @ 2001-11-17 14:39 ` Florian Weimer 2001-11-18 13:31 ` Kai Großjohann 1 sibling, 1 reply; 67+ messages in thread From: Florian Weimer @ 2001-11-17 14:39 UTC (permalink / raw) Jack Twilley <jmt+usenet@twilley.org> writes: > Besides, if you finish working on it, then dozens of people can > actually use it, copyleft notwithstanding. Isn't that good? Working on gpg.el and gpg-ring.el means changing Gnus, too. I've signed a copyright assignment for Gnus (probably a mistake in retrospect, removing my contributions wouldn't have been too hard, and contrary to my expectations, there was plenty of time before the Emacs 21 release), but gpg.el and gpg-ring.el are copyrighted by my employer (well, they have the entire usufruct, I only own the moral rights). This conflict troubles me because I'm not sure whether or not my employer can claim copyright on my Gnus work if it is related to gpg.el or gpg-ring.el. I might obtain the usufruct from my employer, but I think I would need legal advice to examine the required contract. (The probability that the FSF sues me based on a copyright assignment is extremely low, but with my employer, it is existent---fairness is not an issue for some decision makers.) ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-17 14:39 ` Florian Weimer @ 2001-11-18 13:31 ` Kai Großjohann 2001-11-18 19:39 ` Norbert Koch 2001-11-19 21:53 ` Florian Weimer 0 siblings, 2 replies; 67+ messages in thread From: Kai Großjohann @ 2001-11-18 13:31 UTC (permalink / raw) What's a usufruct? kai -- I like BOTH kinds of music. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-18 13:31 ` Kai Großjohann @ 2001-11-18 19:39 ` Norbert Koch 2001-11-19 21:53 ` Florian Weimer 1 sibling, 0 replies; 67+ messages in thread From: Norbert Koch @ 2001-11-18 19:39 UTC (permalink / raw) Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes: > What's a usufruct? 1 definition found >From Webster's Revised Unabridged Dictionary (1913) : Usufruct \U"su*fruct\ (?; 277), n. [L. usufructus, ususfructus, usus et fructus; usus use + fructus fruit.] (Law) The right of using and enjoying the profits of an estate or other thing belonging to another, without impairing the substance. --Burrill. norbert. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-18 13:31 ` Kai Großjohann 2001-11-18 19:39 ` Norbert Koch @ 2001-11-19 21:53 ` Florian Weimer 1 sibling, 0 replies; 67+ messages in thread From: Florian Weimer @ 2001-11-19 21:53 UTC (permalink / raw) Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes: > What's a usufruct? It's called 'Nießbrauch' in German (but I'm sure this doesn't help much ;-). A usufruct is the right to use a thing for gaining profit of some kind, but without diminishing the value of the thing (for example, growing crops on a field in a responsible manner). I was told that 'usufruct of a copyright' is the correct term, but some people prefer 'exploitation rights', I think. German copyright law is a complex field. Copyright does not only include exploitation rights, but also moral rights of the author. In particular, copyright can only be transferred among heirs, and not during lifetime of the author. For computer programs, there are additional twists. For example, you can run COTS software legally without accepting the enclosed EULA in Germany, but you must accept most parts of the GPL before distributing GPLed code. In this area, even professionals make mistakes: Recently, the German branch of IPFI suggested to assign back the exploitation rights for online transmission to the original authors (invalidating all previous contracts), and to legalize Napster in its first incarnation (with songs stored on a central server). All this was done by accident, I suppose. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 20:40 ` Simon Josefsson 2001-11-16 20:58 ` Jack Twilley 2001-11-16 21:21 ` Josh Huber @ 2001-11-17 9:50 ` Fabien Penso 2 siblings, 0 replies; 67+ messages in thread From: Fabien Penso @ 2001-11-17 9:50 UTC (permalink / raw) [-- Attachment #1: Type: text/plain, Size: 1324 bytes --] Simon on Fri, 16 Nov 2001 21:40:54 +0100 wrote: > Josh Huber <huber@alum.wpi.edu> writes: >> Simon Josefsson <jas@extundo.com> writes: >> >>> I don't understand the value of having a header line that says the >>> message was originally encrypted, the client already knows this? >> >> Right...is there an easy way to access this information (if the >> message was signed or encrypted) from a reply buffer? > I don't think so... someone please contribute it. :-) >> I'm thinking about a way to implement the reply_sign, reply_encrypt, >> and reply_signencrypt options of mutt...If you get a message which is >> encrypted, and reply to it, Gnus would automatically encrypt the >> reply. > That would be cool. Perhaps the idea could be integrated with BBDB as > well (or maybe it already is, I think I remember something similar). There is a bbdb-pgp.el which can does something similar, you can tell in a bbdb field if you want to crypt or sign the mail, but it doesn't do what does Mutt as said previously. I modified bbdb-pgp.el so you can use plain sig, MIME sig, plain crypt or MIMe crypt. See http://perso.linuxfr.org/penso/lisp/ -- Fabien Penso <penso@linuxfr.org> | LinuxFr a toujours besoin de : http://perso.LinuxFR.org/penso/ | http://linuxfr.org/dons/ [-- Attachment #2: Type: application/pgp-signature, Size: 239 bytes --] ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 19:37 ` Josh Huber 2001-11-16 20:40 ` Simon Josefsson @ 2001-11-17 9:47 ` Fabien Penso 2001-11-17 11:25 ` Florian Weimer 1 sibling, 1 reply; 67+ messages in thread From: Fabien Penso @ 2001-11-17 9:47 UTC (permalink / raw) [-- Attachment #1: Type: text/plain, Size: 868 bytes --] Josh on Fri, 16 Nov 2001 14:37:14 -0500 wrote: >> I don't understand the value of having a header line that says the >> message was originally encrypted, the client already knows this? > Right...is there an easy way to access this information (if the > message was signed or encrypted) from a reply buffer? > I'm thinking about a way to implement the reply_sign, reply_encrypt, > and reply_signencrypt options of mutt...If you get a message which is > encrypted, and reply to it, Gnus would automatically encrypt the > reply. This would be _really_ cool. What about setting a var to t when you verify sign or cryptn and nil if not. Then when you send the reply, it would look for it. Is this stupid ? -- Fabien Penso <penso@linuxfr.org> | LinuxFr a toujours besoin de : http://perso.LinuxFR.org/penso/ | http://linuxfr.org/dons/ [-- Attachment #2: Type: application/pgp-signature, Size: 239 bytes --] ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-17 9:47 ` Fabien Penso @ 2001-11-17 11:25 ` Florian Weimer 0 siblings, 0 replies; 67+ messages in thread From: Florian Weimer @ 2001-11-17 11:25 UTC (permalink / raw) Fabien Penso <penso@linuxfr.org> writes: > This would be _really_ cool. What about setting a var to t when you > verify sign or cryptn and nil if not. Then when you send the reply, it > would look for it. Is this stupid ? Text properties would be even better. They survive cut and paste. we're slowly approaching the B2 Orange Book level here. ;-) ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 19:20 ` Simon Josefsson 2001-11-16 19:37 ` Josh Huber @ 2001-11-16 20:18 ` news 2001-11-16 21:04 ` Werner Koch 2001-11-16 21:20 ` Matt Armstrong 3 siblings, 0 replies; 67+ messages in thread From: news @ 2001-11-16 20:18 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > Werner Koch <wk@gnupg.org> writes: > >> On Fri, 16 Nov 2001 13:26:20 -0500, Chris Beggy said: >> >>> How about a header field: >> >>> X-Gnus: pgp-verified signature with key 0x454545 valid untrusted >> >> BTW, I might have missed it but an important feature would be to add a >> line like: >> >> X-Gnus-Orig-Encrypted-To: 0x12345678, 0x34567890 >> >> So that one can see that the message was originally encrypted and even >> more important to automagically suggest to encrypt any reply. > I don't understand the value of having a header line that says the > message was originally encrypted, the client already knows this? And > users should probably not trust such headers without the client saying > it is OK, and if the client can do that, the client could use some > other (better) way of conveying this information anyway. I thought you'd like this idea :-) I know you don't like the results placed in the message body, where they can be spoofed, as you showed. Previous posts in this thread have been discussing the shortcomings of: 1. poor visual cue from [hp e] in the modeline to convey signature/encryption status (signed,encrypted,valid,trusted?) 2. poor security of placing encryption status in the message body (you demonstrated this...) 3. introducing Orig-Encrypted-To info, presumably to confirm Cc and To fields, and to promote key exchange and web-of-trust scoring systems Using header fields addresses these points because it is a good place for visual cues when reading mail/news, it can be turned off if the reader doesn't want to be bothered, and it is a good place for ephemeral, timestamped information. Chris ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 19:20 ` Simon Josefsson 2001-11-16 19:37 ` Josh Huber 2001-11-16 20:18 ` news @ 2001-11-16 21:04 ` Werner Koch 2001-11-16 21:20 ` Matt Armstrong 3 siblings, 0 replies; 67+ messages in thread From: Werner Koch @ 2001-11-16 21:04 UTC (permalink / raw) On Fri, 16 Nov 2001 20:20:03 +0100, Simon Josefsson said: > A header to say that replies should be encrypted seems useful, but it > isn't really Gnus specific, and I would also feel better if requests No, that is not the point. Actually I send most mails in the clear and usually reply in clear. However some messages I receive are encrypted and the whole encryption exercise would be rendered meaningless if I would accidently forget to encrypt - actually it happens from time to time and the only chance I have is to kill the uucp job when my fingers were faster than my brain. So the default for an reply to an encrypted message must be 'encrypt'. > I don't understand the value of having a header line that says the > message was originally encrypted, the client already knows this? And > users should probably not trust such headers without the client saying Header lines have the advantage that you can store the mail unencrypted for easier grepping but Gnus is still aware that this message is kind of confidential and will suggest to encrypt a reply. Ciao, Werner ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 19:20 ` Simon Josefsson ` (2 preceding siblings ...) 2001-11-16 21:04 ` Werner Koch @ 2001-11-16 21:20 ` Matt Armstrong 2001-11-16 21:55 ` Simon Josefsson 3 siblings, 1 reply; 67+ messages in thread From: Matt Armstrong @ 2001-11-16 21:20 UTC (permalink / raw) Cc: ding Simon Josefsson <jas@extundo.com> writes: > Werner Koch <wk@gnupg.org> writes: > >> On Fri, 16 Nov 2001 13:26:20 -0500, Chris Beggy said: >> >>> How about a header field: >> >>> X-Gnus: pgp-verified signature with key 0x454545 valid untrusted > > I don't understand the value of having a header line that says the > message was originally encrypted, the client already knows this? > And users should probably not trust such headers without the client > saying it is OK, and if the client can do that, the client could use > some other (better) way of conveying this information anyway. It would be a display-time only field that Gnus generates -- it wouldn't reside on disk. -- matt Piki, you handsome cat, rest in peace http://www.lickey.com/piki/ ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 21:20 ` Matt Armstrong @ 2001-11-16 21:55 ` Simon Josefsson 0 siblings, 0 replies; 67+ messages in thread From: Simon Josefsson @ 2001-11-16 21:55 UTC (permalink / raw) Cc: ding "Matt Armstrong" <matt+dated+1008537652.ebf9f0@lickey.com> writes: >>>> How about a header field: >>> >>>> X-Gnus: pgp-verified signature with key 0x454545 valid untrusted >> >> I don't understand the value of having a header line that says the >> message was originally encrypted, the client already knows this? >> And users should probably not trust such headers without the client >> saying it is OK, and if the client can do that, the client could use >> some other (better) way of conveying this information anyway. > > It would be a display-time only field that Gnus generates -- it > wouldn't reside on disk. Ah. I get it now. Yes, it would be useful (just as long as Gnus doesn't care about the header in incoming articles). I'm not sure it should be displayed by default, but having more ways of conveying information about security status would be nice. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-16 18:26 ` news 2001-11-16 18:43 ` Werner Koch @ 2001-11-17 4:28 ` Andreas Fuchs 1 sibling, 0 replies; 67+ messages in thread From: Andreas Fuchs @ 2001-11-17 4:28 UTC (permalink / raw) [-- Attachment #1.1: Type: text/plain, Size: 795 bytes --] On 2001-11-16, Chris Beggy <news@kippona.com> wrote: >> Also, the buttons create false security. I posted a mail to this >> list that showed PGP buttons that said the message verified OK, but I >> created them using ^H which Gnus bolidifies. Whatever the mechanism >> will be, the article buffer is the wrong place for it. > > How about a header field: > > X-Gnus: pgp-verified signature with key 0x454545 valid untrusted <advocate whose="devil"> Again, what about more than one signed part? And the case where only one (maybe empty) line is signed even with an untrusted key and the rest isn't? There should be a simple solution (like banning all html messages and not inlining images <-;)... </advocate> -- Andreas Fuchs, <asf@acm.org>, asf@jabber.at, antifuchs [-- Attachment #2: Type: application/pgp-signature, Size: 231 bytes --] ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 21:52 ` Simon Josefsson ` (2 preceding siblings ...) 2001-11-16 18:26 ` news @ 2001-11-17 11:15 ` Florian Weimer 2001-11-17 11:41 ` Simon Josefsson 3 siblings, 1 reply; 67+ messages in thread From: Florian Weimer @ 2001-11-17 11:15 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > Also, the buttons create false security. I posted a mail to this list > that showed PGP buttons that said the message verified OK, but I > created them using ^H which Gnus bolidifies. Whatever the mechanism > will be, the article buffer is the wrong place for it. mutt includes the current time when presenting output from the OpenPGP implementation. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-17 11:15 ` Florian Weimer @ 2001-11-17 11:41 ` Simon Josefsson 2001-11-17 14:29 ` Florian Weimer 0 siblings, 1 reply; 67+ messages in thread From: Simon Josefsson @ 2001-11-17 11:41 UTC (permalink / raw) Cc: ding Florian Weimer <fw@deneb.enyo.de> writes: > Simon Josefsson <jas@extundo.com> writes: > >> Also, the buttons create false security. I posted a mail to this list >> that showed PGP buttons that said the message verified OK, but I >> created them using ^H which Gnus bolidifies. Whatever the mechanism >> will be, the article buffer is the wrong place for it. > > mutt includes the current time when presenting output from the OpenPGP > implementation. I can write a JavaScript that extracts current time and places it in the article buffer via W3. I can include a image/gif external MIME part with the current time. Or simply guess when the receiver will see the mail and add that time to the mail. I don't think we'll find The One Secure Method of conveying this information to the user, so multiple methods that each increases the awareness of the user are nice: Indicators in fringe MIME buttons (including time?) Header line created by Gnus (including time?) Modeline icon Summary Buffer icons ... other ideas Obtrusive things (MIME buttons, automatic display of the header line, etc) should probably be turned off by default, Gnus is about reading what people say not about reading digital signatures. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-17 11:41 ` Simon Josefsson @ 2001-11-17 14:29 ` Florian Weimer 0 siblings, 0 replies; 67+ messages in thread From: Florian Weimer @ 2001-11-17 14:29 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: >> mutt includes the current time when presenting output from the OpenPGP >> implementation. > > I can write a JavaScript that extracts current time and places it in > the article buffer via W3. So disable W3 in conjunction with crypto. That's the best idea anyway (W3 can compromise anonymity). > I can include a image/gif external MIME part with the current > time. Same problem, same solution. > Or simply guess when the receiver will see the mail and add that > time to the mail. 'g' doesn't change the timestamp, then. ;-) Well, I forgot that Gnus isn't just a simple mail or news reader. For mutt, doing these things is much easier. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 21:18 ` Simon Josefsson 2001-11-13 21:33 ` Matt Armstrong @ 2001-11-14 10:50 ` Per Abrahamsen 2001-11-15 9:49 ` Per Abrahamsen 2001-11-14 21:56 ` Jack Twilley 2001-11-15 21:23 ` How to minimize buttons? (was Re: PGP support) Jack Twilley 3 siblings, 1 reply; 67+ messages in thread From: Per Abrahamsen @ 2001-11-14 10:50 UTC (permalink / raw) Simon Josefsson <jas@extundo.com> writes: > There is a `p' in the Article buffer's modeline. Apropos, we should make these have tool tips and be mouse sensitive. The following hack does the first, but not the last. Anyone who can see what is wrong? PS: The code below creates the string and properties each time `gnus-article-wash-status' is called, they should probably only be created at start. (defconst gnus-article-wash-status-strings '((cite "c" "Citation toggles" "No citation toggles" gnus-article-hide-citation))) (defun gnus-gnus-article-wash-status-entry (key value) (let* ((entry (assoc key gnus-article-wash-status-strings)) (string (if value (nth 1 entry) (copy-seq " "))) (map (make-sparse-keymap))) (define-key map [ mouse-2 ] (nth 4 entry)) (put-text-property 0 1 'help-echo (if value (nth 2 entry) (nth 3 entry)) string) (put-text-property 0 1 'local-map map string) string)) (defun gnus-article-wash-status () "Return a string which display status of article washing." (save-excursion (set-buffer gnus-article-buffer) (let ((cite (memq 'cite gnus-article-wash-types)) (headers (memq 'headers gnus-article-wash-types)) (boring (memq 'boring-headers gnus-article-wash-types)) (pgp (memq 'pgp gnus-article-wash-types)) (pem (memq 'pem gnus-article-wash-types)) (signed (memq 'signed gnus-article-wash-types)) (encrypted (memq 'encrypted gnus-article-wash-types)) (signature (memq 'signature gnus-article-wash-types)) (overstrike (memq 'overstrike gnus-article-wash-types)) (emphasis (memq 'emphasis gnus-article-wash-types))) (concat (gnus-gnus-article-wash-status-entry 'cite cite) (format "%c%c%c%c%c" (if (or headers boring) ?h ? ) (if (or pgp pem signed encrypted) ?p ? ) (if signature ?s ? ) (if overstrike ?o ? ) (if emphasis ?e ? )))))) ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-14 10:50 ` Per Abrahamsen @ 2001-11-15 9:49 ` Per Abrahamsen 0 siblings, 0 replies; 67+ messages in thread From: Per Abrahamsen @ 2001-11-15 9:49 UTC (permalink / raw) Per Abrahamsen <abraham@dina.kvl.dk> writes: > Apropos, we should make these have tool tips and be mouse sensitive. > The following hack does the first, but not the last. I have commited the tool tips code for the washing status string, but not the mouse sensivity code. ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 21:18 ` Simon Josefsson 2001-11-13 21:33 ` Matt Armstrong 2001-11-14 10:50 ` Per Abrahamsen @ 2001-11-14 21:56 ` Jack Twilley 2001-11-15 21:23 ` How to minimize buttons? (was Re: PGP support) Jack Twilley 3 siblings, 0 replies; 67+ messages in thread From: Jack Twilley @ 2001-11-14 21:56 UTC (permalink / raw) I'm reading your original message with CVS up-to-the-minute Oort, and I don't see any 'p' in the Article buffer's modeline. The buttons do kinda sorta work, though. Jack. -- Jack Twilley jmt at twilley dot org http colon slash slash www dot twilley dot org slash tilde jmt slash ^ permalink raw reply [flat|nested] 67+ messages in thread
* How to minimize buttons? (was Re: PGP support) 2001-11-13 21:18 ` Simon Josefsson ` (2 preceding siblings ...) 2001-11-14 21:56 ` Jack Twilley @ 2001-11-15 21:23 ` Jack Twilley 2001-11-15 21:37 ` Andreas Fuchs 3 siblings, 1 reply; 67+ messages in thread From: Jack Twilley @ 2001-11-15 21:23 UTC (permalink / raw) >>>>> "Simon" == Simon Josefsson <jas@extundo.com> writes: [...] Simon> There is a `p' in the Article buffer's modeline. Using the Simon> fringe for this is one idea. You can also get big buttons if Simon> you like to clutter up your article buffer: Simon> (setq gnus-buttonized-mime-types '("multipart/encrypted" Simon> "multipart/signed")) One side question: how do I minimize the button after clicking on it? Clicking again doesn't do anything useful. Suggestions? Jack. (who likes buttons, yes he does.) -- Jack Twilley jmt at twilley dot org http colon slash slash www dot twilley dot org slash tilde jmt slash ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: How to minimize buttons? (was Re: PGP support) 2001-11-15 21:23 ` How to minimize buttons? (was Re: PGP support) Jack Twilley @ 2001-11-15 21:37 ` Andreas Fuchs 2001-11-16 11:39 ` Jack Twilley 0 siblings, 1 reply; 67+ messages in thread From: Andreas Fuchs @ 2001-11-15 21:37 UTC (permalink / raw) [-- Attachment #1.1: Type: text/plain, Size: 514 bytes --] Today, Jack Twilley <jmt+usenet@twilley.org> wrote: > Simon> (setq gnus-buttonized-mime-types '("multipart/encrypted" > Simon> "multipart/signed")) > > One side question: how do I minimize the button after clicking on it? > Clicking again doesn't do anything useful. Suggestions? Hitting the "g" key (gnus-summary-show-article) works fine, although it minimizes all mime content, and it takes some time to re-evaluate the pgp content. -- Andreas Fuchs, <asf@acm.org>, asf@jabber.at, antifuchs [-- Attachment #2: Type: application/pgp-signature, Size: 231 bytes --] ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: How to minimize buttons? (was Re: PGP support) 2001-11-15 21:37 ` Andreas Fuchs @ 2001-11-16 11:39 ` Jack Twilley 0 siblings, 0 replies; 67+ messages in thread From: Jack Twilley @ 2001-11-16 11:39 UTC (permalink / raw) >>>>> "Andreas" == Andreas Fuchs <asf@void.at> writes: Jack> One side question: how do I minimize the button after clicking Jack> on it? Clicking again doesn't do anything useful. Suggestions? Andreas> Hitting the "g" key (gnus-summary-show-article) works fine, Andreas> although it minimizes all mime content, and it takes some Andreas> time to re-evaluate the pgp content. Okay, that's not a good answer. How tough would it be to make the button a toggle, such that clicking the button again would return it to its previous state? Andreas> -- Andreas Fuchs, <asf@acm.org>, asf@jabber.at, antifuchs Jack. (wishing he were just a tad more skilled at this elisp stuff.) -- Jack Twilley jmt at twilley dot org http colon slash slash www dot twilley dot org slash tilde jmt slash ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 18:52 ` Josh Huber 2001-11-13 20:43 ` Matt Armstrong @ 2001-11-15 0:52 ` Sascha Lüdecke 1 sibling, 0 replies; 67+ messages in thread From: Sascha Lüdecke @ 2001-11-15 0:52 UTC (permalink / raw) -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Josh Huber <huber@alum.wpi.edu> writes: > Per Abrahamsen <abraham@dina.kvl.dk> writes: > >> Actually, I can't. Probably because of >> >> X-MIME-Autoconverted: from quoted-printable to 8bit by >> sheridan.dina.kvl.dk id WAA23581 >> >> Would such convertions affect PGP? > > Well, did the message verify for you? This message is in the 'old'format. You should be able to verify it. > > I suspect that the message is signed first, and then encoded QP, > which would explain the QP encoding of the PGP headers. Thats exacly what mml1991 does. Since the message body can contain non 7-bit characters, one has to protect them from certain MTAs. mml2015, which I took the code from, does that by QP encoding the message body and signing afterwards. This works fine, since the body is a separate MIME part. As far as I can guess/see there are two options with old PGP style: - using ASCII armouring as protection, which renders the messages unreadable without PGP - protecting the 8bit characters after signing. RFC 2045 (Multipurpose Internet Mail Extensions, Part one) states on page four (ASCII version), talking about MIME headers: ... (3) A Content-Transfer-Encoding header field, which can be used to specify both the encoding transformation that was applied to the body and the domain of the result. Encoding transformations other than the identity transformation are usually applied to data in order to allow it to pass through mail transport mechanisms which may have data or character set limitations. ... RFC 2049 (Multipurpose Internet Mail Extensions, Part five) states: A mail user agent that is MIME-conformant MUST: (1) Always generate a "MIME-Version: 1.0" header field in any message it creates. (2) Recognize the Content-Transfer-Encoding header field and decode all received data encoded by either quoted- printable or base64 implementations. The identity transformations 7bit, 8bit, and binary must also be recognized. ... Thus I assumed the QP transfer encoding is safe. Unless your recipient doesn't have a MIME capable mail tool. > FWIW mutt does not do this. Does Outhouse Express support PGP > signed messages that are QP encoded like this? I haven't tested how it works with other mail tools beside Gnus yet. Unfortunately the reason for hacking mml1991 together, an OutOfLuck user who coudln't make use of PGP/MIME, refused to further test it -- installation of plugins seemed to be too compilcated. Best regards, Sascha - -- Nothing is free in this world, so go get some nothing before they charge for it! -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iEYEARECAAYFAjvzEdYACgkQqI/9z8xhHua0LQCeMSLMQ3di9Y55F5RXhCYR1cZr PmkAn3Izr5mt9Mu/EEBbhTNR2/mAwjxt =dHDI -----END PGP SIGNATURE----- ^ permalink raw reply [flat|nested] 67+ messages in thread
* Re: PGP support 2001-11-13 15:53 ` Per Abrahamsen 2001-11-13 18:38 ` Simon Josefsson 2001-11-13 18:52 ` Josh Huber @ 2001-11-17 11:14 ` Florian Weimer 2 siblings, 0 replies; 67+ messages in thread From: Florian Weimer @ 2001-11-17 11:14 UTC (permalink / raw) Per Abrahamsen <abraham@dina.kvl.dk> writes: > X-MIME-Autoconverted: from quoted-printable to 8bit by sheridan.dina.kvl.dk id WAA23581 > > Would such convertions affect PGP? Yes, they can invalidate the signture. That's why you should use the RFC 3156 method. ^ permalink raw reply [flat|nested] 67+ messages in thread
end of thread, other threads:[~2001-12-29 5:14 UTC | newest] Thread overview: 67+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2001-11-12 21:48 PGP support Simon Josefsson 2001-11-13 0:03 ` Matt Armstrong 2001-11-13 1:06 ` Josh Huber 2001-11-13 9:53 ` Simon Josefsson 2001-11-13 10:47 ` Fabien Penso 2001-11-13 11:32 ` Simon Josefsson 2001-11-17 11:12 ` Florian Weimer 2001-11-17 11:27 ` Simon Josefsson 2001-11-13 15:53 ` Per Abrahamsen 2001-11-13 18:38 ` Simon Josefsson 2001-11-14 10:03 ` Per Abrahamsen 2001-11-13 18:52 ` Josh Huber 2001-11-13 20:43 ` Matt Armstrong 2001-11-13 21:18 ` Simon Josefsson 2001-11-13 21:33 ` Matt Armstrong 2001-11-13 21:52 ` Simon Josefsson 2001-11-14 10:07 ` Per Abrahamsen 2001-11-14 10:59 ` dme 2001-11-14 11:52 ` Simon Josefsson 2001-11-14 16:02 ` Andreas Fuchs 2001-11-14 17:11 ` Simon Josefsson 2001-11-14 17:34 ` Nevin Kapur 2001-11-14 17:57 ` Matt Armstrong 2001-11-17 11:19 ` Florian Weimer 2001-11-18 7:23 ` Paul Jarc 2001-11-18 10:53 ` Andreas Fuchs 2001-11-18 20:09 ` Raymond Scholz 2001-11-18 23:38 ` Andreas Fuchs 2001-11-17 11:18 ` Florian Weimer 2001-11-17 11:17 ` Florian Weimer 2001-12-29 5:14 ` Lars Magne Ingebrigtsen 2001-11-16 18:26 ` news 2001-11-16 18:43 ` Werner Koch 2001-11-16 19:20 ` Simon Josefsson 2001-11-16 19:37 ` Josh Huber 2001-11-16 20:40 ` Simon Josefsson 2001-11-16 20:58 ` Jack Twilley 2001-11-16 21:21 ` Josh Huber 2001-11-16 21:38 ` Jack Twilley 2001-11-17 9:52 ` Fabien Penso 2001-11-17 10:08 ` Jack Twilley 2001-11-17 11:23 ` Florian Weimer 2001-11-17 11:10 ` Jack Twilley 2001-11-17 11:54 ` Per Abrahamsen 2001-11-17 14:39 ` Florian Weimer 2001-11-18 13:31 ` Kai Großjohann 2001-11-18 19:39 ` Norbert Koch 2001-11-19 21:53 ` Florian Weimer 2001-11-17 9:50 ` Fabien Penso 2001-11-17 9:47 ` Fabien Penso 2001-11-17 11:25 ` Florian Weimer 2001-11-16 20:18 ` news 2001-11-16 21:04 ` Werner Koch 2001-11-16 21:20 ` Matt Armstrong 2001-11-16 21:55 ` Simon Josefsson 2001-11-17 4:28 ` Andreas Fuchs 2001-11-17 11:15 ` Florian Weimer 2001-11-17 11:41 ` Simon Josefsson 2001-11-17 14:29 ` Florian Weimer 2001-11-14 10:50 ` Per Abrahamsen 2001-11-15 9:49 ` Per Abrahamsen 2001-11-14 21:56 ` Jack Twilley 2001-11-15 21:23 ` How to minimize buttons? (was Re: PGP support) Jack Twilley 2001-11-15 21:37 ` Andreas Fuchs 2001-11-16 11:39 ` Jack Twilley 2001-11-15 0:52 ` PGP support Sascha Lüdecke 2001-11-17 11:14 ` Florian Weimer
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).