goes in here, comes out there
Elemental
http://frogadmin.yi.org/MISC/FromHead.html
I noticed discepancies. As an example, a pgp Anon-Post msg sent through
Antani really gets posted through Freedom. It's not due to anything that
Frog is doing in gathering statistics - it's what the remailer is actually
doing.
I did a few tests of my own and found that it (antani) might come out
through Freedom, Abditum, Frog, etc. Yet Antani is not marked as middleman
in its capstring.
There are many more of similar nature. Here are some notes. Most involve
pgp, but some are mix and a few are plain (all are Post, from the 2nd
table):
freedom pgp (test comes out from holodeck)
bunker mix (comes out from randseed)
italy pgp + mix (etc., etc.)
jmbcv pgp
metacolo pgp + plain
paranoia pgp
randseed pgp + mix
riot pgp + mix
vger plain
antani pgp
I didn't check exhaustively, but 4 from above that I did check (randseed,
metacolo, antani, bunker) are all running Mixmaster remailers.
What is happening?
Frog-Admin
Thanks for your feedback: I was wondering too.
But, because I must still keep my VDU activity minimal,
I did not investigate properly beyond the AARG/ORANGE/CF case
First, I was suspicious of my own code.
My current hypothesis is that,
- -when asked to perform "POST: ",
- -some brands of remailers actually perform
a rand-hop before "TO: m2news" actually takes place.
I was also wondering if, instead of linking my page to google
I should not rather link to my own server (or both):
- -google is 18-24hours behind
- -the view from google and from my server may vary, due to propagation
issues
If somebody wants to check my VB code before I clean it...
(I may still have to wear sun-glasses for the next 4 weeks)
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQEVAwUBPlVSUoDgT488d3zFAQFg8wgAgcQxa6LhyUFBCx31UC/otw7SIIgjyht6
8YyQp2g0kCU0pZtgTZxi38MPTzuU4m4y/CCumIqEzsUI2pGuy2qdFJ+zERyoItyv
aVoF8UBsKkHnHrQ20vl9m/HD5JtHSK3cCXkaKyayLhMVJnt2PepfISnm73qcqDNb
b7aOEv5CKExFOF2nGWNcjyupzdTlhQo1VNP7WWD3jk0pFjwq7a3sVd+Z1T4WlCQA
3SGIMwi4xwqqCnFWskRgo9WQGnYA7vQiEe4iQcd+EGSK+8rgM4HGukg0Z4CNaeyi
gQWf29JcZv0O+UP6+dbHV83/Nd7/Ao/Zrdv3gF1gPsX839/WJB/nag==
=o/6s
-----END PGP SIGNATURE-----
Nomen Nescio
wrote:
How come you don't use freedom mail2news for your tests?
Elemental
That seems likely. Add this supporting evidence: casual inspection via my
newsreader seems to show that the posts (which you test with) that show
this anomaly (the suspected extra hop) consistently get posted later than
one would expect. For instance, in today's series, antani Post + pgp shows
up on news >7 hrs after antani Post + mix and also To + pgp.
One curious question would be: why did the original programmer(s) design it
to add the extra hop?
>
>I was also wondering if, instead of linking my page to google
>I should not rather link to my own server (or both):
>- -google is 18-24hours behind
>- -the view from google and from my server may vary, due to propagation
>issues
I'd say that would only let the problem show up quicker on your stats page,
and this isn't a time-dependent problem anyways. IMO, the next logical step
would instead be to identify which remailers/versions exhibit the behavior,
then have a notation on your stats page for them to indicate which rows in
your table are not-applicable, because the true exit remailer is unknown.
One method to identify which remailers exhibit the behavior would be just
to make n identical posts and observe whether or not the end result varies.
The value for n would have to be large enough (3? 5?) so as to reasonably
preclude one of those remailers adding the extra hop *to itself* (which I
did observe one time). Kindly pardon me if I state the obvious.
>
>If somebody wants to check my VB code before I clean it...
>(I may still have to wear sun-glasses for the next 4 weeks)
Sorry, can't help there - not my language. Good luck with the eyes, though.
Also let me take the oportunity to say thanks for providing the From: stats
page in the first place, without which this might not have surfaced.
Elmer Fud
> Also let me take the oportunity to say thanks for providing the From: stats
> page in the first place, without which this might not have surfaced.
This already surfaced at FarOut's From: statistics (whatever spheres
Farout admin might roam now).
Mixmaster remailers introduce a random-hop when posting with Post:
directive instead of Anon-To: m2n or mixmaster post facilities.
Actually, this has been known for years and nobody ever wondered about
it. Maybe it's a feature.
Frog-Admin
On Fri, 21 Feb 2003, Nomen Nescio <nob...@dizum.com> wrote:
>How come you don't use freedom mail2news for your tests?
- -Putting the 4 mail2news names would break a 80c limit
and potentially cause the whole message to be lost
- -Putting the 4 mail2news might exceed a
"[Addresses Per Header Maximum]=3"
which is enforced at *some* remailers
(3, 4, 5, 10 are populat figures, Frog's maximum is 5
- -freedom mail2news adds a disclaimer at the Top
it would blur the information I ptrovide in "FromHead"
(unless I perform extra work to discriminate Feedom's part)
My VB sources are meant to be available and I welcome "competitors"
The names of mail2news used in the tests are one of the parameters
from some \MyParams.ini, easy-to-modify text file:
[MiscPath]="Y:\MISC"
[Misc127]="http://127.0.0.1/MISC"
[MiscM2NewsStd]="mail...@frogadmin.yi.org,mail...@anon.lcs.mit.edu,mail...@dizum.com"
[MiscTestNG]="alt.test"
[MiscFHAppName]="FromHead"
[MiscAppName]="Probe"
[MiscFHSubjectChztn]="Frog-Probe [FromHeader] _ "
[MiscFHSuppliedFrom]="Jack Ass <Jac...@jackass.invalid>"
[MD5_Seed1]="xxxPwd1"
[MD5_Seed2]="yyyPwd2"
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQEVAwUBPljeBoDgT488d3zFAQEbgggA0TObezOw8Eg1M2qt7IXQ0RfJCyCpJfix
FQYsYi01+xBtN4G+XWm+XQuOL81rt8PrnOtmVjRAhavKxVMzQpEkoBkLfl4CGHvT
HNQrYQD9CqxBGYFZLvCIpgkD04L1P2Hy6jiCQpcRfah0JBqyuaDwe7PB7L+3/bjC
iTZipk4x8ZYqA+1DbodDzbAEcW9waQOtliaCf9ogo3x/aHqZyVfuoZaGe5R0378t
jDD6NayTpOSiMx4CyiowhwpXIzFlSPjWBp+E2alUzWayelKBXKueXWskR9CNOnE5
7tRNLn5ANMPVWknTp1GvU2+415XxJkXS/90jbPi9HxCts9DYJGUCjQ==
=CPqs
-----END PGP SIGNATURE-----
Frog-Admin
On 21 Feb 2003, Anonymous...@See.Comment.Header (Elmer Fud)
wrote:
>Elemental wrote:
>
>> Also let me take the oportunity to say thanks for providing the From: stats
>> page in the first place, without which this might not have surfaced.
>
>This already surfaced at FarOut's From: statistics (whatever spheres
>Farout admin might roam now).
That is absolutely correct, and I still link to Farout's goodies,
because I did not yet adapt all what he did
CC had tried to introduce something about "From" in capstrings,
but his suggestion was not adopted.
Farout started his programming after CC's suggestion appeared to be
dead
BTW
I am not the inventor of "accurate stats and keyrings", not even of
"MetaStats"
Drule and Gretchen made the first steps in that direction.
Drule is the one whose activity marked me the most because,
even after he had to stop the remailer
and went on with "stats and keyrings only"
it was obvious that he remained an essential part of the community
ALSO
No idea is "copyrighted" here (not for me, at least)
If I do something and somebody can make even better (webpage,
program...)
that is fine for me and I don't ask for credit.
Even if there is no improvement on what I am doing, I welcome
competition:
the more "Frog" overs *exclusive_and_useful_services*
the more nefarious consequences with its potential demise
the more it becomes a valuable target for anybody willing to attack
the network
(besides focusing the envy and hatred of all do-nothing)
Healthy "competitors" protect me from such risk,
and I expressed that idea since Frog's services gained some popularity
Competition about "From Headers" and exploitation of "Remailer
Description" data is where it is more needed, on the "application"
side.
Another M2news (5th, 6th, ...) is probably the most important thing
(Frog's m2news thoughput is typically in the 400-600 posts/day now,
after spammers tested it and got barred)
Browse into Frog Remailer Load Graphs
http://www.privacyresources.org/frogadmin/Graphs/Browse.html
>Mixmaster remailers introduce a random-hop when posting with Post:
>directive instead of Anon-To: m2n or mixmaster post facilities.
>
>Actually, this has been known for years and nobody ever wondered about
>it. Maybe it's a feature.
I don't know the "history".
The behaviour was a surprise to me, because personnally I always use
mail2news,
(which offer the only way I know to cross-post an E-mail @ and Usenet)
-----BEGIN PGP SIGNATURE-----
Version: N/A
iQEVAwUBPljnFYDgT488d3zFAQEJPQgAgXgKHeljbw6nMC5j1aWsDeNHev0xpbAS
k6iOsQ/EHz0lWiDRKBzQWX5oc72F/Bk5fsyrK3vPkdlasm0zgKbk+Ck55Tx5zikX
53114bIv/wIiaKzzUnhtGBMc6Il+4ELh/Ktr1T7qxvM9XE9qZ3KE/XIWO2DAs4fH
pGXuUu8reJMhR4acD6fe/bB4OqzmIVZMJO7s30Rs/jr1mi8AjKqcVPijR4eekM43
5uoxD464SycqToXCV8nb6Nqgx1P14ug5S+UYDzns7V2RP7mHgsCJzAOv8DWG6V1l
0xXSdJTtuj1BsqSx6/AhxcEkTk5L8i5xRRPK43Ix3ZHiYWUYvWfuBQ==
=K0kf
-----END PGP SIGNATURE-----