|
Home > Archive > Anonymous Servers > September 2004 > QS 1.2.4: barring "last remailers"
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
QS 1.2.4: barring "last remailers"
|
|
| Anonymous 2004-09-23, 10:01 am |
| Apart from randseed, what remailers should I bar from beign last in the chain?
Freddie
| |
| emperor 2004-09-23, 10:01 am |
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
In article <a95b5b23bca84334e31c90e558925e56@ecn.org>
Anonymous <cripto@ecn.org> wrote:
>Apart from randseed, what remailers should I bar from beign last in the chain?
>
>Freddie
If you don't want to use HashCash - "panta".
*** emperor
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQA/AwUBQVKBojk6FTOF+YXVEQL/VwCgsKkZJ+gBGRUUyWMpk3p3RhQkMzEAoJqN
Z5GyFwQiQgXbTK96g2E4prCh
=dUr3
-----END PGP SIGNATURE-----
| |
| panta-admin 2004-09-23, 10:01 am |
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi emperor !
>If you don't want to use HashCash - "panta".
Could you explain why ?
Cheers,
panta-admin
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQA/ AwUBQVKS2B2e88Id2BOOEQIGTQCffepSxQn7R2VZ
vvj4/lajS5OLS6kAniMp
U72ub5jzGml08EGQPNAnfrhc
=5If9
-----END PGP SIGNATURE-----
~~~~~~~~~~~~~~~~~~~~~
This message was posted via one or more anonymous remailing services.
The original sender is unknown. Any address shown in the From header
is unverified. You need a valid hashcash token to post to groups other
than alt.test and alt.anonymous.messages. Visit www.panta-rhei.dyndns.org
for abuse and hashcash info.
| |
| emperor 2004-09-23, 10:01 am |
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
In article <WPWESSXI38253.4310416667@anonymous.poster>
panta-admin <anonymous@panta-rhei.dyndns.org> wrote:
>Hi emperor !
Hello, panta-admin!
>
>Could you explain why ?
Sure.
You require a valid HashCash-token for posting to the Usenet. If
such a token is not supplied, your remailer forwards the message on
to a randomly chosen remailer. This is a good idea to stop msgs from
being dropped just because some/many people don't know/understand
your requirement.
However, due to your remailer being middle that way, there is no
reason why it should be used as exit intentionally. I can't speak
for others, but I personally don't want you (or anyone else, for
that matter) to decide for me how long my chain is/should be. For
instance, if I specify "Chain: *,*,*; copies=2", I expect it to
choose exactly that, 3 remailers and the third one (whichever that
would be) as exit. Of course, the exit-remailer actually should be
an exit-remailer.
Furthermore, adding yet another remailer also adds more (unwanted?)
latency. Reliability goes down a bit, too.
Conclusion:
People wishing to post to Usenet, understanding what HashCash is,
being aware of your requirement and not intending to use HC, would
be better off barring "panta" from being chosen as the exit/last
remailer. JMHO.
*** emperor
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQA/ AwUBQVLD3Dk6FTOF+YXVEQLgTgCeOxV2agjkZQ89
AMNs03D0rJFNo4oAoOsp
zbRU08H4IxRCN332CJia6uSV
=dj8V
-----END PGP SIGNATURE-----
| |
| panta-admin 2004-09-23, 10:01 am |
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi !
>However, due to your remailer being middle that way, there is no
>reason why it should be used as exit intentionally. I can't speak
>for others, but I personally don't want you (or anyone else, for
>that matter) to decide for me how long my chain is/should be. For
>instance, if I specify "Chain: *,*,*; copies=2", I expect it to
>choose exactly that, 3 remailers and the third one (whichever that
>would be) as exit. Of course, the exit-remailer actually should be
>an exit-remailer.
As you said, hashcash only applies to usenet postings, but banning panta as
exit will also prevent it from being chosen as last remailer for email.
I am not sure if it enhances your anonymity set if you completely ban panta
as exit as you would lose one of only 19 exit remailers for email, which
happens to be one of only 5 with full from support.
>Furthermore, adding yet another remailer also adds more (unwanted?)
>latency. Reliability goes down a bit, too.
Using a random remailer you do not have too much knowledge about the
latency anyway, and as the random remailer at panta is chosen by mixmaster
it should not be much different from the one your client would have chosen.
Well, on second thought about the latency issue, I could make randhopped
news posts have zero latency at panta to minimize that effect.
>Conclusion:
>People wishing to post to Usenet, understanding what HashCash is,
>being aware of your requirement and not intending to use HC, would
>be better off barring "panta" from being chosen as the exit/last
>remailer. JMHO.
Well from my experience by far the most traffic is email, so I would not
generalize this statement.
I also think most people wishing to post to usenet should find a good exit
remailer which will post to the groups they want to post to and then
manually fix it as last remailer to enhance reliability.
Alas, you are fully entitled to your opinion, just wanted to make sure
newbies do understand that the hashcash issue is only applying to news
postings.
Two more questions though if you have time:
How is your posting success rate with a *,*,* chain ?
I thought most people would choose the exit remailer manually for news
postings ?
And: Do you happen to know why he wants to ban randseed ?
Cheers,
panta-admin
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQA/AwUBQVLMfB2e88Id2BOOEQKAtwCg/HMUf0YoalTjj9Te7llJGn2X6T0An1Vo
jHa1Mm6mXoFggBefW0qomCyA
=xsEB
-----END PGP SIGNATURE-----
~~~~~~~~~~~~~~~~~~~~~
This message was posted via one or more anonymous remailing services.
The original sender is unknown. Any address shown in the From header
is unverified. You need a valid hashcash token to post to groups other
than alt.test and alt.anonymous.messages. Visit www.panta-rhei.dyndns.org
for abuse and hashcash info.
| |
| Max Mustermann 2004-09-23, 5:45 pm |
| On 23 Sep 2004, panta-admin <anonymous@panta-rhei.dyndns.org> wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA1
>
>Hi emperor !
>
>
>Could you explain why ?
If no valid hashcash token is supplied, you choose a *random*
remailer.
That is *better* than plain dumping the post (100% loss), but with
~50% remailers barring any given Usenet newsgroup (some bar *whole*
Usenet), your policy still results in ~50% posts lost.
I strongly suggest that you change your policy, and start using a
*static* subrogate remailer instead of a *random* one.
If you don't want to put the burden of *all* your (non-HC) posts on a
single remailer, build a short list of Usenet-friendly remailers and
randomly choose among those.
Randomly choosing the exit remailer was always strongly discouraged,
and only ignorant idiots (or idiots with an agenda to weaken the
remailer network's reliability) can insist that "random choice for
exit remailers is *the* way"
| |
| emperor 2004-09-23, 5:45 pm |
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
[References restored]
In article <5FG07R7Z38253.5999768519@anonymous.poster>
panta-admin <anonymous@panta-rhei.dyndns.org> wrote:
>Hi !
Hello!
[snip]
>As you said, hashcash only applies to usenet postings, but banning panta as
>exit will also prevent it from being chosen as last remailer for email.
I totally understand your points. The problems are your mixed config
and HashCash not being a standard.
>I am not sure if it enhances your anonymity set if you completely ban panta
>as exit as you would lose one of only 19 exit remailers for email, which
>happens to be one of only 5 with full from support.
Well, according to the "AARG! Pinger", there are 22 exit remailers,
6 with full and 1 with partial "From:" support. Then there is "dingo"
operating as exit with full "From:" support for Usenet-postings.
Besides, even if a remailer is marked as middle, one can still
specify it directly in the chain.
[snip]
>Using a random remailer you do not have too much knowledge about the
>latency anyway, and as the random remailer at panta is chosen by mixmaster
>it should not be much different from the one your client would have chosen.
That may be so, but it's not changing the fact that another
(unwanted?) hop is being added. The bigger the chain, the less the
reliability.
>Well, on second thought about the latency issue, I could make randhopped
>news posts have zero latency at panta to minimize that effect.
Not sure, but wouldn't that make traffic analysis easier?
[snip]
>Well from my experience by far the most traffic is email, so I would not
>generalize this statement.
I wrote: "People wishing to post to Usenet, ...". :o)
>I also think most people wishing to post to usenet should find a good exit
>remailer which will post to the groups they want to post to and then
>manually fix it as last remailer to enhance reliability.
Agreed.
>Alas, you are fully entitled to your opinion, just wanted to make sure
>newbies do understand that the hashcash issue is only applying to news
>postings.
Sorry for not being more clear. Yes, your remailer does _not_
require HashCash for mere e-mails.
>Two more questions though if you have time:
>How is your posting success rate with a *,*,* chain ?
I don't use such a chain, thus I can't tell. Sorry.
>I thought most people would choose the exit remailer manually for news
>postings ?
Ah, yes. My referenced "Chain: *,*,*; copies=2" was just an example. ;o)
>And: Do you happen to know why he wants to ban randseed ?
Ummm, no, I don't _know_. ;o) But I _suspect_ it's because...
http://www.noreply.org/echolot/thesaurus/1.conf
-----------------------------------------------------------------
The following newsgroups/domains are blocked:
alt.privacy.anon-server
[...]
-----------------------------------------------------------------
*** emperor
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQA/ AwUBQVL3Wzk6FTOF+YXVEQINxQCeMDP8v9NtkXiO
/PRtlPv2n/TxddYAnR84
C2UX+immZWgD01kR4MkybEYD
=lmUK
-----END PGP SIGNATURE-----
| |
| QuickSilver 2004-09-23, 8:45 pm |
| -----BEGIN PGP SIGNED MESSAGE-----
In article <5FG07R7Z38253.5999768519@anonymous.poster>
panta-admin <anonymous@panta-rhei.dyndns.org> wrote:
>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi !
>
>
> As you said, hashcash only applies to usenet postings, but banning panta as
> exit will also prevent it from being chosen as last remailer for email.
> I am not sure if it enhances your anonymity set if you completely ban panta
> as exit as you would lose one of only 19 exit remailers for email, which
> happens to be one of only 5 with full from support.
>
>
> Using a random remailer you do not have too much knowledge about the
> latency anyway, and as the random remailer at panta is chosen by mixmaster
> it should not be much different from the one your client would have chosen.
> Well, on second thought about the latency issue, I could make randhopped
> news posts have zero latency at panta to minimize that effect.
>
>
> Well from my experience by far the most traffic is email, so I would not
> generalize this statement.
> I also think most people wishing to post to usenet should find a good exit
> remailer which will post to the groups they want to post to and then
> manually fix it as last remailer to enhance reliability.
Using a specific remailer for exit is certainly an option, and I'm
quite sure a good many QS users do just that. No doubt there are QS
users with hashcash using your remailer as exit. However, I believe
allowing mixmaster to choose the final remailer is the better way.
There have been reasons that allowing the final remailer to be chosen
randomly could and would reduce reliability. Your own remailer was an
excellent example. It was good of you to reconsider and change that
policy.
Now, I have implemented a feature that users can bar *any* remailer
from last in the chain. You are not alone. Any remailer that blocks a
newsgroup a user wishs to use will drop the message if that remailer is
chosen for last. Astute users can determine the remailers that might
cause problems and bar those from last.
I have done what I can to allow QS to use your remailer without
hashcash and, if they wish, without penalty of an extra hop. If people
bar your remailer, either last or completely, it is your doing. You
have decided to require hashcash. The remailer network, as it stands,
is not ready for this and I'm pretty certain we will never see it in
mixmaster. Possibly mixminion. Anyway, I've provided an option that
allows users to work with you or without you.
> Alas, you are fully entitled to your opinion, just wanted to make sure
> newbies do understand that the hashcash issue is only applying to news
> postings.
>
> Two more questions though if you have time:
> How is your posting success rate with a *,*,* chain ?
For myself, maybe 8 or 9 out of 10. With the new 'bar from last'
option, users can remove known impediments to reliable delivery. I'm
sure reliability will improve.
> I thought most people would choose the exit remailer manually for news
> postings ?
> And: Do you happen to know why he wants to ban randseed ?
The original poster here brought up randseed because I used it as an
example in a message to my support mailing list. The example was that
there are other remailers users might wish to bar from last. I noted
that if a post is headed for a.p.a-s then randseed is a problem because
it blocks a.p.a-s. I'll be posting a similar message in this group in a
day or 2 announcing this new capability. Just want to make sure all is
well first.
I hope you see that what I have done is fair and reasonable. The fact
is, you have improved QS by spurring on this new 'bar from last'
capability. I see this sould have been done long ago.
The user has many choices:
1) Users can choose any remailer they wish for exit.
2) Users can provide hashcash and use your remailer for exit if they
wish.
3) Users can allow the final remailer to be chosen randomly, without
hashcash, and if yours is selected, they get an extra hop. That was
your wish and it is available to QS users.
4) Users who dislike what you are doing and want no part of your
remailer can bar you completely.
5) Users who wish to use you as middleman but not last have that option.
You cannot have it all. If you want hashcash then some are going to bar
you. It is as simple as that. Full from headers...email
unaffected...each user will make the decision themselves. I have not
barred any remailers from last by default. If users want to do that
themselves that is their option.
Richard
- --
R.Christman
quicksilver <at> skuz.net
http://www.quicksilvermail.net
How many people download QS? http://quicksilvermail.net/ftpusage.html
QS mailing list http://www.quicksilvermail.net/mailman/listinfo/qslist
Superb remailer statistical analysis http://www.noreply.org
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQBVAwUBQVOD2/ kiDnOECmYBAQF1fgIA1MZRF0pCRWB+b7JpKT5UNK
FRIREf+w0Z
0ygeut1nMpLJogVhVfKGM2CPdqZmk2ZlnRFZBgYH
ZICeAGuSDKGTPQ==
=PKZh
-----END PGP SIGNATURE-----
| |
| starwars 2004-09-24, 2:45 am |
| In article <5FG07R7Z38253.5999768519@anonymous.poster>
panta-admin <anonymous@panta-rhei.dyndns.org> wrote:
>
> And: Do you happen to know why he wants to ban randseed ?
>
> Cheers,
> panta-admin
Yes, I was wondering why certain factions are having a downer on Randseed.
I've always found it very reliable, but of course it depends where you update
your keys from. I guess having got rid of Frog, who too was reliable, they are
now moving on to Randseed. Who will be next after Randseed?
| |
| Max Mustermann 2004-09-24, 8:45 pm |
| On Thu, 23 Sep 2004, emperor <emperor@the.empire> wrote:
>However, due to your remailer being middle that way, there is no
>reason why it should be used as exit intentionally. I can't speak
>for others, but I personally don't want you (or anyone else, for
>that matter) to decide for me how long my chain is/should be. For
>instance, if I specify "Chain: *,*,*; copies=2", I expect it to
>choose exactly that, 3 remailers and the third one (whichever that
>would be) as exit. Of course, the exit-remailer actually should be
>an exit-remailer.
Then you better quit using remailers completely, because your messages are
routinely remixed and randhopped on their way between those randomly
selected remailers.
I thought everyone knew this?!?
| |
| Tarapia Tapioco 2004-09-25, 2:45 am |
| On Sat, 25 Sep 2004, Max Mustermann <anonymous@remail.amessage.info> wrote:
>On Thu, 23 Sep 2004, emperor <emperor@the.empire> wrote:
>
>
>Then you better quit using remailers completely, because your messages are
>routinely remixed and randhopped on their way between those randomly
>selected remailers.
>
>I thought everyone knew this?!?
Some of us did. ;) And mixminion will probably do even more of this, FWIW.
| |
| Anonymous 2004-09-27, 5:45 pm |
| In article <b71a148a914089f1937745cc09096528@remail.amessage.info>
Max Mustermann <anonymous@remail.amessage.info> wrote:
>
> On Thu, 23 Sep 2004, emperor <emperor@the.empire> wrote:
>
>
> Then you better quit using remailers completely, because your messages are
> routinely remixed and randhopped on their way between those randomly
> selected remailers.
>
> I thought everyone knew this?!?
No, at least I didn't. Please explain.
Are you saying that my message sent out as rot26,cripto,randseed,italy with
no directives will probably end up being something like
rot26,*,*,cripto,*,randseed,*,*,italy?
| |
|
| -----BEGIN PGP SIGNED MESSAGE-----
On Sat, 25 Sep 2004, Max Mustermann <anonymous@remail.amessage.info> wrote:
>On Thu, 23 Sep 2004, emperor <emperor@the.empire> wrote:
>
>
>Then you better quit using remailers completely, because your messages are
>routinely remixed and randhopped on their way between those randomly
>selected remailers.
>
>I thought everyone knew this?!?
You are incorrect about one point. Unless you specify randhop as a
directive, or select a middleman remailer for your exit, you message will
not be randhopped. Remixed, possibly, but not randhopped.
IB
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQA/ AwUBQViQoXaabJ0OMl2fEQIXmwCcCO2hHWdktxDe
mIr4alXMmTRhKTQAoLMl
neMzhpA1gSh1IcHK665B024C
=AYwh
-----END PGP SIGNATURE-----
| |
| Anonymous 2004-09-27, 8:45 pm |
| In article <200409272231.i8RMVAiN007031@marco.aarg.net>
IB <IB@hod.aarg.net> wrote:
>
> On Sat, 25 Sep 2004, Max Mustermann <anonymous@remail.amessage.info> wrote:
<snip>
>
> You are incorrect about one point. Unless you specify randhop as a
> directive, or select a middleman remailer for your exit, you message will
> not be randhopped. Remixed, possibly, but not randhopped.
>
> IB
Thank you. I thought I was going nutzo.
| |
| emperor 2004-09-28, 7:53 am |
| -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
In article < 44f98f53975e2554303f82e3a5bebcbb@paranoi
ci.org>
Anonymous <nobody@paranoici.org> wrote:
[snip]
>Thank you. I thought I was going nutzo.
"Max Mustermann" was just trying to troll me (and possible other
readers as well)... :o)
*** emperor
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQA/AwUBQVlWFDk6FTOF+YXVEQK5qQCg/OTYYmTQDkMm1tNTdSIBraAKeOQAoLkh
Hp78dQWyZu5g7U2YDYsY7aq+
=13gz
-----END PGP SIGNATURE-----
| |
| Thomas J. Boschloo 2004-09-28, 5:45 pm |
| -----BEGIN PGP SIGNED MESSAGE-----
emperor wrote:
| In article < 44f98f53975e2554303f82e3a5bebcbb@paranoi
ci.org>
| Anonymous <nobody@paranoici.org> wrote:
|
| [snip]
|
|
|>>Thank you. I thought I was going nutzo.
|
|
| "Max Mustermann" was just trying to troll me (and possible other
| readers as well)... :o)
I thought he meant that remailers are randomly selected, now that I read
his post again I see that this is not the case. Only middleman remailers
randhop for mix and cpunk nodes they have no knowledge off (so if a
remailer is unmaintained and middle, in addition to having old
(incorrect) keys, it might also randhop to another (new) remailer that
is not white-listed in their config!).
Regards,
Thomas
- --
The Thraddash: "So, what's this? SNORT! An unknown alien species?"
"How wonderful! Someone new to fight!"
Full Game Win/Mac/Linux: <http://sc2.sourceforge.net>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iQB5AwUBQVmN4wEP2l8iXKAJAQFNpAMeJDKyM2zS
wFHMBrviVI3S9qfCByRmpGEz
Jvdo5MIrMkMMX0Hjuw4ZE+6D4kB6VOAmEsLojQJE
4kloxU7NGGteNIu4xbty1j9T
adW4gm5bHEKHAttqmg6gmiUPEOR8Kd9hTCQEWA==
=RBIL
-----END PGP SIGNATURE-----
|
|
|
|
|