[Dovecot] On-delivery deduplication?
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
Hi,
A feature of Cyrus-IMAPd I really missed after migrated to Dovecot is their optional "duplicate suppression", which eliminates duplicate message at deliver time, if their envelope sender, recipient and message-id matches. For example, if one subscribes to a mailing list, and someone hit "Reply All" to reply to him, there will be two different SMTP deliveries, one from the replier's server and one from the mailing list.
Is it possible to do the same on latest Dovecot? (Looking at the documentation it seems not?) Or is there some better ways to implement the same functionality?
Thanks in advance!
Cheers,
Xin LI <delphij@delphij.net> http://www.delphij.net/ FreeBSD - The Power to Serve! Live free or die -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (FreeBSD)
iQEcBAEBCAAGBQJN79XjAAoJEATO+BI/yjfBCO0IAMLNZ9Rkargb6bQqILAcbobt 2ZFb6cO8f4vJu/BPO7Dq3TZoIEmfR+/pbruwgV+FSslTL+5TjkB5Rd5Yb+HYgvZT Qdbk8Jshgq1XBj2MvbLQugHKhNcN37oDI9eGCDPttYe/C4Jb2hh1W+RtqGpfS6ud jXUTUzS4yr7HWRRYcyDMvL4HuG/VnKEuFYmX2T/mT8oHfz5FckIPM+k+0prbgMSx IGTdkKpSMnqYnyUjxQklSmMJ5iJMM7Lr6FMS2c6ORKwG8rD0w1GKrU77saHqJJLQ YJqqIwJcr5753US6Q1Rk4gK+/L4JmJ+2BUXDdNMYBIfPOFjH9nOLlE5HIfSskBg= =fo5J -----END PGP SIGNATURE-----
On 8.6.2011, at 23.04, Xin LI wrote:
A feature of Cyrus-IMAPd I really missed after migrated to Dovecot is their optional "duplicate suppression", which eliminates duplicate message at deliver time, if their envelope sender, recipient and message-id matches. For example, if one subscribes to a mailing list, and someone hit "Reply All" to reply to him, there will be two different SMTP deliveries, one from the replier's server and one from the mailing list.
IMHO that's a good thing. I know some people in this list think they want deduplication, but sometimes when I'm later replying to a message (I've messages over a month old I know need replying) they really would prefer for it to be shown in they inbox when I finally reply to them rather than dovecot list mailbox (which they never end up reading then).
Is it possible to do the same on latest Dovecot? (Looking at the documentation it seems not?) Or is there some better ways to implement the same functionality?
I'm not planning on implementing anything like that.
Xin LI <delphij@delphij.net> (Wed Jun 8 22:04:51 2011): (…)
A feature of Cyrus-IMAPd I really missed after migrated to Dovecot is their optional "duplicate suppression", which eliminates duplicate message at deliver time, if their envelope sender, recipient and message-id matches.
Not sure, but I *think*, cyrus does dedup based on the message id and recipient only.
For example, if one subscribes to a mailing list, and someone hit "Reply All" to reply to him, there will be two different SMTP deliveries, one from the replier's server and one from the mailing list.
If your above statement about the dup detection is true, these two messages will be delivered both, since the envelope sender is likely to differ (most if not all list servers replace the envelope sender for several reasons (bounce detection, SPF passing, …))
Is it possible to do the same on latest Dovecot? (Looking at the documentation it seems not?) Or is there some better ways to implement the same functionality?
And these two messages are no duplicates. They differ in the headers, probably they differ in parts of the body (signature of the list server), EVEN they carry the same message id!
A duplicate is a duplicate if the message and its copies are identical. And from MPOV this is very unlikly to happen in common environments.
OTOH, if you need such feature, it shouln't be too challenging to write a MDA replacement, that decides about duplicity and finally passes the remaining messages to the Dovecot MDA.
-- Heiko :: dresden : linux : SCHLITTERMANN.de GPG Key 48D0359B : 3061 CFBF 2D88 F034 E8D2 7E92 EE4E AC98 48D0 359B
On 08/06/11 22:48, Heiko Schlittermann wrote:
Xin LI <delphij@delphij.net> (Wed Jun 8 22:04:51 2011): (…)
A feature of Cyrus-IMAPd I really missed after migrated to Dovecot is their optional "duplicate suppression", which eliminates duplicate message at deliver time, if their envelope sender, recipient and message-id matches.
Not sure, but I *think*, cyrus does dedup based on the message id and recipient only.
For example, if one subscribes to a mailing list, and someone hit "Reply All" to reply to him, there will be two different SMTP deliveries, one from the replier's server and one from the mailing list.
If your above statement about the dup detection is true, these two messages will be delivered both, since the envelope sender is likely to differ (most if not all list servers replace the envelope sender for several reasons (bounce detection, SPF passing, …))
Is it possible to do the same on latest Dovecot? (Looking at the documentation it seems not?) Or is there some better ways to implement the same functionality?
And these two messages are no duplicates. They differ in the headers, probably they differ in parts of the body (signature of the list server), EVEN they carry the same message id!
A duplicate is a duplicate if the message and its copies are identical. And from MPOV this is very unlikly to happen in common environments.
OTOH, if you need such feature, it shouln't be too challenging to write a MDA replacement, that decides about duplicity and finally passes the remaining messages to the Dovecot MDA.
This happens because the person replying uses either the reply, or the reply-to-all button in his MUA. The MUA should honour the headers of the message it replies to, and most of the time MUAs do that.
People who want replies on their messages sent back only to the list, and not to a personal address, should set the reply-to header of their message to the list address:
From: me@example.com To: list@example.org Reply-to: list@example.org
This should always make a reply going only to list@example.org, whether the person replying uses reply, reply-all or reply-to-list. This can be done by hand, or with a mailclient plugins that does this for all messages that are recognized as mailing list messages. Unfortunately I did not find such a plugin for thunderbird yet...
NB I just set the headers in this message per my example above, for you to test. If you hit any reply* button to react on to this message, a decent MUA should not include my personal address as a recipient of your message.
-- Regards, Tom
Hello Tom,
Tom Hendrikx <tom@whyscream.net> (Wed Jun 8 23:17:29 2011): (…)
OTOH, if you need such feature, it shouln't be too challenging to write a MDA replacement, that decides about duplicity and finally passes the remaining messages to the Dovecot MDA.
This happens because the person replying uses either the reply, or the reply-to-all button in his MUA. The MUA should honour the headers of the message it replies to, and most of the time MUAs do that.
I understand what you're explaining. But I do not understand how it is related to duplicated messages.
BTW and OT: the MUA agent I'm using (mutt) knows "list reply" feature, it works independent on From/Reply-To header fields. Using just "reply" in presence of a Reply-To header field, it asks if it's my intention to send the reply to somebody else than the originator. And it knows "group reply", that's the same as "reply all" in other MUA. And finally it knows a "bounce", sometimes known as "resend" - not to be confused with forward. It just lacks a shiny *G*ui.
Greetings from Dresden,
Heiko :: dresden : linux : SCHLITTERMANN.de GPG Key 48D0359B : 3061 CFBF 2D88 F034 E8D2 7E92 EE4E AC98 48D0 359B
On 08/06/11 23:46, Heiko Schlittermann wrote:
Hello Tom,
Tom Hendrikx <tom@whyscream.net> (Wed Jun 8 23:17:29 2011): (…)
OTOH, if you need such feature, it shouln't be too challenging to write a MDA replacement, that decides about duplicity and finally passes the remaining messages to the Dovecot MDA.
This happens because the person replying uses either the reply, or the reply-to-all button in his MUA. The MUA should honour the headers of the message it replies to, and most of the time MUAs do that.
I understand what you're explaining. But I do not understand how it is related to duplicated messages.
The point is that when you set the headers correctly on your message, a reply from someone on your message will not generate a duplicate in the first place, thereby eliminating your problem even before it exists :)
-- Regards,
TomOn 06/08/2011 05:58 PM, Tom Hendrikx wrote: [...]
The point is that when you set the headers correctly on your message, a reply from someone on your message will not generate a duplicate in the first place, thereby eliminating your problem even before it exists :)
To add a data point, this message was a "reply-to-all" in Thunderbird 3.1.10. It included Tom's address, ignoring the reply-to: header.
Considering Thunderbird's popularity, just using a reply-to: header won't solve the duplicate message "problem". This is especially true on open mail lists, like those at kernel.org, where "reply-to-all" is expected of participants.
Phil
On 09/06/11 00:47, Phil Turmel wrote:
On 06/08/2011 05:58 PM, Tom Hendrikx wrote: [...]
The point is that when you set the headers correctly on your message, a reply from someone on your message will not generate a duplicate in the first place, thereby eliminating your problem even before it exists :)
To add a data point, this message was a "reply-to-all" in Thunderbird 3.1.10. It included Tom's address, ignoring the reply-to: header.
Actually I only set the reply-to header (by hand) on the message in which I said that I did that, and not on the second one, because I am lazy and there is no tb plugin to make my life easier. Please try again on the correct message.
Considering Thunderbird's popularity, just using a reply-to: header won't solve the duplicate message "problem". This is especially true on open mail lists, like those at kernel.org, where "reply-to-all" is expected of participants.
I use Thunderbird too, and I did test what I documented. To be sure, please check the headers of both messages.
I don't use this practice often (depends on the ml, and the ppl on it), but some time ago the duplicate issue irritated me enough to spend a good thought at what the real problem was. Just shared my results, especially since they are apparently non-obvious :)
But maybe we're getting a bit off-topic. After all this is a list about dovecot and IMAP-related stuff.
-- Regards, Tom
On 06/08/2011 07:05 PM, Tom Hendrikx wrote:
On 09/06/11 00:47, Phil Turmel wrote:
On 06/08/2011 05:58 PM, Tom Hendrikx wrote: [...]
The point is that when you set the headers correctly on your message, a reply from someone on your message will not generate a duplicate in the first place, thereby eliminating your problem even before it exists :)
To add a data point, this message was a "reply-to-all" in Thunderbird 3.1.10. It included Tom's address, ignoring the reply-to: header.
Actually I only set the reply-to header (by hand) on the message in which I said that I did that, and not on the second one, because I am lazy and there is no tb plugin to make my life easier. Please try again on the correct message.
I checked before I sent, and sure enough, it's there. Maybe Timo has set the list to add it. This one of yours has it, too.
Considering Thunderbird's popularity, just using a reply-to: header won't solve the duplicate message "problem". This is especially true on open mail lists, like those at kernel.org, where "reply-to-all" is expected of participants.
I use Thunderbird too, and I did test what I documented. To be sure, please check the headers of both messages.
Checked.
I don't use this practice often (depends on the ml, and the ppl on it), but some time ago the duplicate issue irritated me enough to spend a good thought at what the real problem was. Just shared my results, especially since they are apparently non-obvious :)
But maybe we're getting a bit off-topic. After all this is a list about dovecot and IMAP-related stuff.
True enough. I'll stop here.
Phil
Only kind of relevant to this thread, but:
I hate this whole idea of being able to tell where messages should be replied to. Basically the rule seems to be "don't give me a separate notification if the email reply was sent to me 0-2 weeks after I sent the originating message, but do send it if it was sent after that". Of course no one can know how long you actually bother to follow the mailing list, usually not even you. The only exception are people like me and [2-10] other people who really do follow it year after year. Perhaps I should remove the feature from my patched mailman, although then maybe I'd have to figure out how to implement it for myself to the clients I use.. Oh well.
(And while I'm complaining .. I hate just as much the bottom posts with everything quoted than top posts. At least with top posts I see what people are saying, but with bottom posts I have to scroll pagefuls of old text to see it. Try to put something new into the first pageful of the screen..)
On 9.6.2011, at 0.46, Heiko Schlittermann wrote:
Hello Tom,
Tom Hendrikx <tom@whyscream.net> (Wed Jun 8 23:17:29 2011): (…)
OTOH, if you need such feature, it shouln't be too challenging to write a MDA replacement, that decides about duplicity and finally passes the remaining messages to the Dovecot MDA.
This happens because the person replying uses either the reply, or the reply-to-all button in his MUA. The MUA should honour the headers of the message it replies to, and most of the time MUAs do that.
I understand what you're explaining. But I do not understand how it is related to duplicated messages.
BTW and OT: the MUA agent I'm using (mutt) knows "list reply" feature, it works independent on From/Reply-To header fields. Using just "reply" in presence of a Reply-To header field, it asks if it's my intention to send the reply to somebody else than the originator. And it knows "group reply", that's the same as "reply all" in other MUA. And finally it knows a "bounce", sometimes known as "resend" - not to be confused with forward. It just lacks a shiny *G*ui.
Greetings from Dresden,
Heiko :: dresden : linux : SCHLITTERMANN.de GPG Key 48D0359B : 3061 CFBF 2D88 F034 E8D2 7E92 EE4E AC98 48D0 359B
On 08/06/2011 23:13, Timo Sirainen wrote:
(And while I'm complaining .. I hate just as much the bottom posts with everything quoted than top posts. At least with top posts I see what people are saying, but with bottom posts I have to scroll pagefuls of old text to see it. Try to put something new into the first pageful of the screen..)
:-)
For some reason this always seems to be a massive bone of contention to a whole load of people! Largely those who prefer one way over the other don't seem to recognise that the other bunch also have a valid reason for their preference and the religious wars continue...
However! I found a useful plugin for Thunderbird (which I see you don't use, but still). It's called Quote Collapse and basically it does what you might imagine, the indented bits of messages get collapsed to just their first line (for context) and you get a +/- button to expand if you need to read the rest of the context. Personally for me this makes bottom posting readable in general and top posting is fine, and my preference then becomes "interleaved" because it looks quite pretty with the collapsed quoting...
Actually, email clients also get religious, but despite it's other deficiencies I find Thunderbird with the following plugins helpful:
- Country Lookup - shows which country the email was sent from (great for customer support)
- Display Mail User Agent - shows the mail client used (great for customer support)
- Quote Collapse (as above)
- Coloured Diffs - pretty printing for code diffs found in emails
- Sieve - Minimal sieve rule editing
Perhaps this is enough to tempt you from Mac Mail?
Cheers
Ed W
On 2011-06-09 6:14 AM, Ed W wrote:
However! I found a useful plugin for Thunderbird (which I see you don't use, but still). It's called Quote Collapse and basically it does what you might imagine, the indented bits of messages get collapsed to just their first line (for context) and you get a ± button to expand if you need to read the rest of the context. Personally for me this makes bottom posting readable in general and top posting is fine, and my preference then becomes "interleaved" because it looks quite pretty with the collapsed quoting...
Been using it since forever, and I agree, it is quite useful, although only for plain text messages that use proper quote characters...
--
Best regards,
Charles
On Thu, 09 Jun 2011 08:31:26 -0400 Charles Marcus <CMarcus@Media-Brokers.com> articulated:
On 2011-06-09 6:14 AM, Ed W wrote:
However! I found a useful plugin for Thunderbird (which I see you don't use, but still). It's called Quote Collapse and basically it does what you might imagine, the indented bits of messages get collapsed to just their first line (for context) and you get a ± button to expand if you need to read the rest of the context. Personally for me this makes bottom posting readable in general and top posting is fine, and my preference then becomes "interleaved" because it looks quite pretty with the collapsed quoting...
Been using it since forever, and I agree, it is quite useful, although only for plain text messages that use proper quote characters...
Many MUAs such as "claws-mail" have that function build it. I use it extensively on some list. I still find it hard to believe that some moron will actually reply to a message with quadruple quotes or more and not have the common sense to edit them out. I basically drop anything over two levels when replying.
-- Jerry ✌ Dovecot.user@seibercom.net
Disclaimer: off-list followups get on-list replies or get ignored. Please do not ignore the Reply-To header.
Health nuts are going to feel stupid someday, lying in hospitals dying of nothing.
Redd FoxxOn Thu, June 9, 2011 12:14, Ed W wrote:
However! I found a useful plugin for Thunderbird (which I see you don't use, but still). It's called Quote Collapse and basically it does what you might imagine, the indented bits of messages get collapsed to just their first line (for context) and you get a +/- button to expand if you need to read the rest of the context. Personally for me this makes bottom posting readable in general and top posting is fine, and my preference then becomes "interleaved" because it looks quite pretty with the collapsed quoting...
I found this plugin (v0.8), but it is not compatible with Thunderbird v3.1.10 apparently.
On Thu, June 9, 2011 15:17, Charles Marcus wrote:
On 2011-06-09 9:14 AM, Giles Coochey wrote:
I found this plugin (v0.8), but it is not compatible with Thunderbird v3.1.10 apparently.
Just edit the rdf file and bump the version and it works fine...
Humm, I'm not well versed on plug-ins. I have an XPI file... 7-Zip can extract it and I can see a install.rdf, but I have no idea how to re-package the edited plug-in so that thunderbird can read it again.
Think I'll wait for the author to update...
On 2011-06-09 9:23 AM, Giles Coochey wrote:
On Thu, June 9, 2011 15:17, Charles Marcus wrote:
On 2011-06-09 9:14 AM, Giles Coochey wrote:
I found this plugin (v0.8), but it is not compatible with Thunderbird v3.1.10 apparently.
Just edit the rdf file and bump the version and it works fine...
Humm, I'm not well versed on plug-ins. I have an XPI file... 7-Zip can extract it and I can see a install.rdf, but I have no idea how to re-package the edited plug-in so that thunderbird can read it again.
Think I'll wait for the author to update...
You'll be waiting forever...
Just open the .xpi file, then open the install.rdf file, change the version to 3.1.*. then save it... done... now you can install it...
Or better... I've attached a fixed version...
--
Best regards,
Charles
On 09/06/2011 14:14, Giles Coochey wrote:
On Thu, June 9, 2011 12:14, Ed W wrote:
However! I found a useful plugin for Thunderbird (which I see you don't use, but still). It's called Quote Collapse and basically it does what you might imagine, the indented bits of messages get collapsed to just their first line (for context) and you get a +/- button to expand if you need to read the rest of the context. Personally for me this makes bottom posting readable in general and top posting is fine, and my preference then becomes "interleaved" because it looks quite pretty with the collapsed quoting...
I found this plugin (v0.8), but it is not compatible with Thunderbird v3.1.10 apparently.
Baffling - that's the version I appear to be using for both...
Ed W
This has been quite an informative topic, although it veered from the initial question quite early on.
I know Dovecot does not do deduplication on its own, but perhaps someone can point me to a good Maildir deduplication program. In my case it has nothing to do with mailing lists. We have a shared namespace for certain groups where multiple people can put email regarding certain projects.
Naturally, sometimes two people receive the same email and they will both go put their copy in the same folder. It'd be nice (not critical) if I could remove the duplicates somehow. Obviously they are not byte-for-byte duplicates because their headers are going to be /slightly/ different near the end of the Received sequence.
Willie
On Wed, Jun 8, 2011 at 5:04 PM, Xin LI <delphij@delphij.net> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
Hi,
A feature of Cyrus-IMAPd I really missed after migrated to Dovecot is their optional "duplicate suppression", which eliminates duplicate message at deliver time, if their envelope sender, recipient and message-id matches. For example, if one subscribes to a mailing list, and someone hit "Reply All" to reply to him, there will be two different SMTP deliveries, one from the replier's server and one from the mailing list.
Is it possible to do the same on latest Dovecot? (Looking at the documentation it seems not?) Or is there some better ways to implement the same functionality?
Thanks in advance!
Cheers,
Xin LI <delphij@delphij.net> http://www.delphij.net/ FreeBSD - The Power to Serve! Live free or die -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (FreeBSD)
iQEcBAEBCAAGBQJN79XjAAoJEATO+BI/yjfBCO0IAMLNZ9Rkargb6bQqILAcbobt 2ZFb6cO8f4vJu/BPO7Dq3TZoIEmfR+/pbruwgV+FSslTL+5TjkB5Rd5Yb+HYgvZT Qdbk8Jshgq1XBj2MvbLQugHKhNcN37oDI9eGCDPttYe/C4Jb2hh1W+RtqGpfS6ud jXUTUzS4yr7HWRRYcyDMvL4HuG/VnKEuFYmX2T/mT8oHfz5FckIPM+k+0prbgMSx IGTdkKpSMnqYnyUjxQklSmMJ5iJMM7Lr6FMS2c6ORKwG8rD0w1GKrU77saHqJJLQ YJqqIwJcr5753US6Q1Rk4gK+/L4JmJ+2BUXDdNMYBIfPOFjH9nOLlE5HIfSskBg= =fo5J -----END PGP SIGNATURE-----
Hello Mr. Xin Li,
would you try the OpenDedup File System? (http://www.opendedup.org) If your problem is (only) the message size, this will works like a charm to you.
BTW, this is a great project to use with any maildir compatible server... it reduces a LOT of disk usage space...
Best regards.
-- Use cópia oculta (BCC ou CCO) e apague dados pessoais no campo da mensagem ao encaminhar qualquer e-mail.
On 09/06/2011 13:36, Allan Cassaro wrote:
would you try the OpenDedup File System? (http://www.opendedup.org) If your problem is (only) the message size, this will works like a charm to you.
BTW, this is a great project to use with any maildir compatible server... it reduces a LOT of disk usage space...
Just theoretically I wouldn't expect dedup to make an interesting difference, especially not a naive dedup algorithm? Would you mind sharing some statistics from your installation?
Cheers
Ed W
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
On 06/09/11 05:36, Allan Cassaro wrote: [...]
would you try the OpenDedup File System? (http://www.opendedup.org) If your problem is (only) the message size, this will works like a charm to you.
Well, for me it's not a matter of size (I use FreeBSD ZFS which can do dedup by the way), but simply same message showing up on same inbox for twice, for IMAP, this can be quite annoying.
Also I don't think filesystem level dedup would work well for mail servers even using a filesystem that is capable for doing dedup (not to mention it's sometimes slow), because we store the message header together with the bodies, and they are likely to differ slightly (someone@ vs someoneelse@domainname), causing an offset in the message and thus make block-level dedup useless, so in order to efficiently use dedup (for speed and space), one have to migrate to a new storage format.
I'll take a look at the code and see if I can do that myself when I have some spare time, and thanks for everyone who replied.
Cheers,
Xin LI <delphij@delphij.net> http://www.delphij.net/ FreeBSD - The Power to Serve! Live free or die -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (FreeBSD)
iQEcBAEBCAAGBQJN8Tv/AAoJEATO+BI/yjfBz6AH/3FWAMNuqTWTvaT8lnSfDl99 7KclQamlczrKvJ3XKjfFCaAzf73PQffwfJbz/RtMfwCgKZhiZG86U5uo15qb8eKU pUxDVpukxarrb9r/13x5vxrSJz33QVebwdzvdBckZk2sbzUodW1MoMIH/66jmuDG gojKzn4gdTW1jH5OUlWN3NZSmib44xvTvSjRDX/6OV6DBvAmpHRFOm1f6iEk5BHE sPnbKEzwNoqAvDlqzPynuhOc+acTuaVL0/+dyEmVcI5neZf6eYRljgoP+NlzF4yk R9IpghAnRe5Us4ZIXvla1fT345ckAhPtvP0j5FxTjXq4IBChkV4k7PXkwlQSocE= =PyS4 -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
On 06/09/11 05:36, Allan Cassaro wrote: [...]
Well, for me it's not a matter of size (I use FreeBSD ZFS which can do dedup by the way), but simply same message showing up on same inbox for twice, for IMAP, this can be quite annoying.
Also I don't think filesystem level dedup would work well for mail servers even using a filesystem that is capable for doing dedup (not to mention it's sometimes slow), because we store the message header together with the bodies, and they are likely to differ slightly (someone@ vs someoneelse@domainname), causing an offset in the message and thus make block-level dedup useless, so in order to efficiently use dedup (for speed and space), one have to migrate to a new storage format.
I'll take a look at the code and see if I can do that myself when I have some spare time, and thanks for everyone who replied.
Cheers,
Xin LI <delphij@delphij.net> http://www.delphij.net/ FreeBSD - The Power to Serve! Live free or die -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (FreeBSD)
iQEcBAEBCAAGBQJN8Tv/AAoJEATO+BI/yjfBz6AH/3FWAMNuqTWTvaT8lnSfDl99 7KclQamlczrKvJ3XKjfFCaAzf73PQffwfJbz/RtMfwCgKZhiZG86U5uo15qb8eKU pUxDVpukxarrb9r/13x5vxrSJz33QVebwdzvdBckZk2sbzUodW1MoMIH/66jmuDG gojKzn4gdTW1jH5OUlWN3NZSmib44xvTvSjRDX/6OV6DBvAmpHRFOm1f6iEk5BHE sPnbKEzwNoqAvDlqzPynuhOc+acTuaVL0/+dyEmVcI5neZf6eYRljgoP+NlzF4yk R9IpghAnRe5Us4ZIXvla1fT345ckAhPtvP0j5FxTjXq4IBChkV4k7PXkwlQSocE= =PyS4 -----END PGP SIGNATURE-----
participants (11)
-
Allan Cassaro
-
Charles Marcus
-
Ed W
-
Giles Coochey
-
Heiko Schlittermann
-
Jerry
-
Phil Turmel
-
Timo Sirainen
-
Tom Hendrikx
-
Willie Gillespie
-
Xin LI