December 2003 - cypherpunks-legacy

Re: BROKEN REMAILERS
by lcottrell＠popmail.ucsd.edu 17 Dec '03

17 Dec '03

>I have tested remailers and their PGP keys over some time now and this is >what I have found: > >MIXMASTER is a great, fast and reliable remailer but you can't PGP with it. > >REMAILER(a)NATELY.UCSD.EDU is just as fast and reliable (same guy) and you >can use PGP, but a few weeks ago it started acting really funny: It will >forward any ATTACHED files, but it will only read the first commandline >of a PGP block that has been pgp'ed with Nately's public key. In other >words, if you wrap layers of chained remailers inside each other, you >will now have to have NATELY as the *last* link in your chain as it will >otherwise discard the rest of the block. Alternatively you can have the >chain as a series of attached files. > I am not sure what is going on with remailer@nately. I just tested it, sending a PGP encrypted message chained through it 3 times, I also sent a message chained through nately to c2. They both worked fine. Are you using a program to build the chain, like "Chain"? If so, which? Mixmaster now has a PGP key (it has had one for a long time for testing purposes, this is a new one). The front end for using mixmaster remailer type 2 messages is on my WWW page. Right now it only runs on Suns. -----BEGIN PGP PUBLIC KEY BLOCK----- Version: 2.6 mQCNAy8y6PAAAAEEAJF05HUUQM4n9G5wHOZ+bHZbF6yFmeX5wBGElssHqUnbb4PQ S/f8jWvJyfiZkjlOkzOEJ47fZ++x+yYWgzKGUKp2ejIKu2MvnzgLfIkwKfxioPXq xgBpLcFaDrbM52Mi4MEpEYW4DJqxk9TrAnHZBwXmxpOZQ57UcgbmF1BLzawNAAUR tC5NaXhtYXN0ZXIgUmVtYWlsZXIgPG1peG1hc3RlckBuYXRlbHkudWNzZC5lZHU+ iQCVAwUQLzLpflVkk3dax7hlAQH2GQP9FGBU7uGako4YghrUNzyb3tcdwZwsGkBx Gy9MGojOSwWgXQ0CRLD9+IWgkZnRY902pU65C3IZhFB7erXBk5eyGOXhbahHv07o mhHRiE2v4Z9hNapWVHHeDo0F2AoL7B8kkiFjXI2fVYzofB6gmXd7X8Wzh/9AEQsz 63U8X9cbajyJAJUDBRAvMulXBuYXUEvNrA0BAQOkA/9QfgNGLp6MX1Qwk55uLiHG ajeT+Rtgruaj9DPOoZFVj9w0pImBkCQNw5eA6iZTgiAqkbSG7RaR9yKMU27Gf/O+ yz3cWAx4slFmCpVDlEl5qSa1Bd+2LuneN4EvnxfeI7TLUIFKlPcIOnx9zsu5Ny0I SCtHVHBaGw43OT1OYmLjgA== =Zr4D -----END PGP PUBLIC KEY BLOCK----- -------------------------------------------------- Lance Cottrell who does not speak for CASS/UCSD loki(a)nately.ucsd.edu PGP 2.6 key available by finger or server. Encrypted mail welcome. Home page http://nately.ucsd.edu/~loki/ Check out my essay on the next generation remailer Mixmaster on the WWW page. For anon remailer info, mail remailer(a)nately.ucsd.edu Subject: remailer-help "Love is a snowmobile racing across the tundra. Suddenly it flips over, pinning you underneath. At night the ice weasels come." --Nietzsche

1 0

Re: signature validation on secure edit messages
by joelm＠eskimo.com 17 Dec '03

17 Dec '03

Try turning Word Wrap off in Eudora. I was encountering the same thing in the Windows version of Eudora. With Word Wrap on, it seems to add CRs (or CR/LFs on a PC) to the end of each line upon sending. This obviously causes a message not to match the signature, since all of those CRs weren't there when you composed it. With Word Wrap off, your text still wraps on the screen when you're composing, but those unwanted CRs in the sent mail don't get added. Joel McNamara joelm(a)eskimo.com - finger for PGP key >Jeez. I really don't know why the heck this always happens to me, my >digsigs not validating. > >You can finger me at tjbryce(a)amherst.edu to retrieve a message with a valid >signature. > >I always write the message in eudora, copy it, sign it in MacPGP as TEXT >with attached signature, then go back and paste it over the old message. >Then I go back to MacPGP and check the signature on the clipboard to be >sure - no problems. But after getting mailed, it doesn't work. > >Damn. > >Tom >

1 0

Re: signature validation on secure edit messages
by dmandl＠panix.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Good point. But if you turn word wrap off, your message will go out as one long line, or one long line for each paragraph. Most smart mail programs will handle this cleanly, but I assume there are still many that won't. Have all the people who are using PGP with Eudora been sending mail without line breaks? Is this not a problem for 99.99% of the mail-reading software out there? --Dave. At 5:28 PM 2/5/95, Joel McNamara wrote: >Try turning Word Wrap off in Eudora. I was encountering the same thing in the Windows version of Eudora. With Word Wrap on, it seems to add CRs (or CR/LFs on a PC) to the end of each line upon sending. This obviously causes a message not to match the signature, since all of those CRs weren't there when you composed it. With Word Wrap off, your text still wraps on the screen when you're composing, but those unwanted CRs in the sent mail don't get added. > >Joel McNamara >joelm(a)eskimo.com - finger for PGP key > >>Jeez. I really don't know why the heck this always happens to me, my >>digsigs not validating. >> >>You can finger me at tjbryce(a)amherst.edu to retrieve a message with a valid >>signature. >> >>I always write the message in eudora, copy it, sign it in MacPGP as TEXT >>with attached signature, then go back and paste it over the old message. >>Then I go back to MacPGP and check the signature on the clipboard to be >>sure - no problems. But after getting mailed, it doesn't work. >> >>Damn. >> >>Tom >> -----BEGIN PGP SIGNATURE----- Version: 2.6 iQCVAwUBLzWQLZeBqjXzSLLFAQHBbwP+OwhP/d7Yim3/GTY0sI3s17QY4ve+dH6K UoxUboys/VuPFD60gYMdM8LrlxwQksRQm7JWxgp204crKucPZuyOO3ZpZyqiwAmx 23NMf5qtEmxJV5fCPQ8mh61vVeGXOuG7VUA1S+Q3yd7AmZw/PKOllF41VtfmPZvK weRwnYmb7Ho= =q+U5 -----END PGP SIGNATURE----- -- Dave Mandl dmandl(a)panix.com

1 0

Re: "encrypt tcp connections" hacks
by lrh＠PrimeNet.Com 17 Dec '03

17 Dec '03

I realize this may be heresy, but have you considered a hardware solution? Cisco is currently offering a version of their popular routers with a builtin Zylink crypto box... uses either DES or a "proprietary single bit CFB mode" system (maybe RC4?) which is exportable. The boxes are relatively inexpensive. Zylink offers just the crypto box for even less. Both boxes offer DH key management. Lyman To obtain my PGP key, finger lrh(a)primenet.com or see my Web page at http://www.primenet.com/~lrh/index.html

1 0

Zimmermann charges dropped?
by tcmay＠netcom.com 17 Dec '03

17 Dec '03

Apologies if this has been discussed and I somehow missed it. I just got back from Monte Carlo last night and have briefly scanned the 500+ accumulated messages, not seeing this item mentioned. Drew Taubman (sp?), the new Executive Director of the EFF, told me that in recent days the grand jury meeting in the PRZ case did in fact decide to go forward with the indictment. However, says Drew, the judge in the case dismissed the charges after looking at them, apparently in about an hour or so. He cited the absence of relevant law in the case (as I understand it, the argument being that there is no law against what PRZ did, i.e., "make code available" in such a way that _others_ might then access it and perhaps then illegally export it (assuming such export might be later proved to be illegal). Drew went on to tell me that the judge said there would need to be a relevant law on this, and that Senator Leahy and his staff (recall their involvement in the Digital Telephony Bill) are already at work drafting such legislation. I haven't seen mention of this, so I thought I'd mention it here. I could of course try to get confirmation first from Phil, but he is notoriously hard to reach in e-mail, ironically, and this seems sufficiently important to mention to the list. And if you can't trust the EFF, who _can_ you trust? (Don't answer that.) --Tim May -- .......................................................................... Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)netcom.com | anonymous networks, digital pseudonyms, zero | knowledge, reputations, information markets, W.A.S.T.E.: Aptos, CA | black markets, collapse of governments. Higher Power: 2^859433 | Public Key: PGP and MailSafe available. Cypherpunks list: majordomo(a)toad.com with body message of only: subscribe cypherpunks. FAQ available at ftp.netcom.com in pub/tc/tcmay

1 1

Here we go again...
by Johnathan Corgan 17 Dec '03

17 Dec '03

+---------------------------------------------------------+ ##### # # # # # ##### #### ##### # ## ## # # # # # # # # # ### # # # # # # # # # ### #### # # # # # ####### ####### # # # # # ##### # # # # # # ##### ##### # # # +---------------------------------------------------------+ -> EMA ALERT <- News For and About the Members of the ELECTRONIC MESSAGING ASSOCIATION ============================================================ February 3, 1995 -- Number 18 <----------------------------------------------------------> ***** SPECIAL ALERT ***** - Congress to consider making all system operators liable for messaging content. Bill would force employers to monitor message content. ACTION NEEDED NOW! <----------------------------------------------------------> UNREASONABLE NETWORK POLICING PROPOSED Yesterday, Senator Jim Exon (D-NE) introduced S.314, the Communications Decency Act of 1995, in the United States Senate. In an effort to stamp out digital pornography, it makes all telecommunications providers doing business in the United States (from the telephone companies all the way down to offices that use LANs) liable for the content of anything sent over their networks. To avoid the possibility of tens of thousands of dollars in fines and up to two years in jail, business owners would be forced to police their networks and monitor in advance all messages sent over them. WITHOUT ACTION - COULD BE LAW IN MONTHS This bill is substantially the same as the one he put forward last year. He will offer it as an amendment to the pending telecommunications deregulation legislation in the U.S. Senate, which is expected to be enacted by July. Last year, his amendment was adopted even though many thought it hastily drafted and poorly thought out. Fortunately, the telecommunications deregulation legislation died. This year, a more conservative U.S. Congress may be even more reluctant to challenge a "morality" amendment; and its legislative vehicle, the telecommunications deregulation legislation, stands a much better chance of passage this year. ACTION NEEDED NOW Action by the business community is needed now. Please notify your corporate government affairs office and/or your legal counsel. This measure could be adopted as an amendment to the telecommunications bill IN A MATTER OF WEEKS (or potentially added to any legislation pending on the U.S. Senate floor), if business does not mobilize against it. S.314 will not stop digital pornography, but it could devastate the messaging business. If you are interested in further information or are able to participate in lobbying efforts over the next few weeks, contact Sarah Reardon at EMA (see below). ------------------------------------------------------------ EMA ALERT is published and copyrighted (1995) by the Electronic Messaging Association. Permission to reproduce and/or redistribute with attribution is hereby given to all EMA members. For more information about anything in EMA ALERT, contact EMA via e-mail - use either X.400 (S=info; O=ema; A=mci; C=us) or Internet (info(a)ema.org) address, facsimile (1-703-524-5558), or telephone (1-703-524-5550). Any EMA staff member can be addressed directly via e-mail by using, for X.400, G=<firstname>; S=<lastname>; O=ema; A=mci; C=us, and, for Internet, <firstinitial><lastname>@ema.org. EMA's postal address is 1655 N. Fort Myer Dr. #850, Arlington, VA 22209 USA. == Johnathan Corgan "Violence is the last refuge of the incompetent." jcorgan(a)aeinet.com -Isaac Asimov

2 1

Re: Remailer encryption module
by Hal 17 Dec '03

17 Dec '03

From: eric(a)remailer.net (Eric Hughes) > -- The RSA-encrypted session key does not have a flat representation > over its multiword container. This yields a statistical traffic > analysis hole. (This point is irrelevant without fixing 4.) Hal and > I completely solved this problem last year. For reference, here is that old message with an algorithm that produces an encrypted session key with a flat distribution over a specified number of bytes, along with a proof that it works. The purpose of this is you could strip off the PGP header stuff and have a file which looked for all intents and purposes like totally random bytes, but if you knew the secret key then you could decrypt it just fine. (I recently took my CP archives and indexed them using Mark Zimmermann's (no relation to Phil, apparently) FreeText browser which lets me do keyword searches. Pretty nice.) > Date: Mon, 7 Mar 94 08:34:04 -0800 > From: hughes(a)ah.com (Eric Hughes) > Message-Id: <9403071634.AA10351(a)ah.com> > To: cypherpunks(a)toad.com > In-Reply-To: Hal's message of Sun, 6 Mar 1994 11:22:17 -0800 <199403061922.LAA26901(a)jobe.shell.portal.com> > Subject: Truly Stealthy PGP (algorithm) > Sender: owner-cypherpunks(a)toad.com > Precedence: bulk > Status: RO > > >If I understand Eric's general idea, we would keep trying session keys > >under a set of rules which would lead to the desired statistical > >distribution of the encrypted key. > > I actually said nothing about how to get the particular distribution > of keys specified, since that was another issue. I was more concerned > with just getting the one result across. > > >Here is an algorithm which would work. > > It does work, and I'll put down a proof sketch below. > > Notation alert: > >Let L be the next power of 256 above the modulus n. Let t be the integer > >part of L/n, so that L = n*t + s with s in [0,n). Call the PGP IDEA session > >key SK, and the encrypted version of that m = SK^e. Now do these steps: > > >1) Pick a random SK in [0,n). > > This random number in [0,n) is the wrong distribution, but that's OK, > since we'll be throwing some numbers away. > > >2) RSA-encrypt it to form m = SK^e mod n. > > RSA encryption is a bijection (an 1-1 map). If it were not, there > would be two or more possible decryptions for a given ciphertext. > Therefore RSA encryption is a permutation, and a permutation of > probabilities preserves expected values of functions of the > probability, such as entropy. Since we assume the entropy of the SK > is maximal (probabilistic entropy), therefore the entropy of the m's > is maximal. So the m's have a flat distribution. > > (As always, the above statements about bijection hold only if SK is > multiple of one of the divisors of the modulus. But then if you do > find one of those, you've also factored the modulus and thus broken > the key. We assume this doesn't happen, since if it does little of > this matters anyway.) > > >3) Choose a random k in [0,t]. > >4) Calculate the "stegged" encrypted key as M = m + k*n. > > Hal now observes that M is uniformly distributed. This is correct, > and happens because m is in [0,n) and we are adding a multiple of n to > m. This means that each M has a unique represenative as some pair > <m,k>. Since both m and k are independently random (max entropy, flat > distribution), so is M. > > >5) if M is not in [0,L) (i.e. if M >= L) then go back to step 1. > >The idea is that once we get M uniform in [0,(t+1)*n) we can make it > >uniform in [0,L) simply by rejecting those candidates which were too high. > > What we have here is a Markov chain. We have accepting states and > rejecting/retrying states. Since the probabilities in the chain are > independent of each other and are also time-invariant, the > distribution of final probabilities is the same as the distribution of > normalized accepting probabilities. > > In simple terms, you can just retry until you get it right. Since the > probabilities are all the same before, they will all be the same > after, only larger to account for the fact that some possibilities > didn't work. > > [re: rejection and retry] > >This will only happen if k=t and m>=s. > > That's right, and that means that for m < s you have valid k in > [0,t+1) and for m >= s only for [0,t). If you go back an look at the > entropy expression, you'll see exactly this difference in relative > probability for the two parts of [0,n). > > >Now, it seems to me that the worst case for rejection is when n=L-1, in > >which case t=1, s=1, and almost one-half of all initial SK choices will > >be rejected. > > Right, but the worst case for rejection is not the same as the worst > case for entropy loss, which occurs at n=L/2+1 and s=t-1, i.e. at the > other end of the spectrum entirely. > > >Following Eric's reasoning, this would be an effective loss > >of one bit of key length, from say 1024 to 1023, which is tolerable. > > Actually not. The loss of effective key length happens based on the > posterior distribution of the session keys, not on the number of > rejections that happen in the process. > > >Using this algorithm with the current Stealth PGP would produce a > >"truly stealthy" version which I think would be indistinguishable from > >random bytes without access to the receiver's private key. > > Indeed. Observe, though, that as far as deployment went, this would > require modification to PGP itself for it to be anything like > widespread. > > Eric

1 0

Re: finney's perl scripts
by Hal 17 Dec '03

17 Dec '03

From: anonymous-remailer(a)shell.portal.com > hal , gotta question regarding your perl scripts . > i ran em thru sh and d/l'd em , and here are the > results of my tests . first , i pgp'd a file in > binary format , then ran " perl pgppadt.pl > test.pgp 10 " . the error i got was " Couldn't > create test.pgp.pad " . so i renamed the file to > " test " and tried again with good results ! i got > " Input file test has size 732 bytes 10 bytes > pgppad returns code -3 " . then , iran " perl > pgppad.pl test 10 " and after the bit about perl > running under dos/4gw protected mode , i get dropped > to my command prompt . i took a look at the file , > and it's size wasn't any different , so i renamed the > file to test.pgp and ran it again and got the same > results . so i guess i'm wonderin' if it added the > padding , or what might be the problem ? for your > info , i'm using perl 4.0. Unfortunately, my PC's disk died several months ago so I don't have one right now. pgppadt.pl sets the output file name with: open (OUT, ">$ARGV[0].pad") || die ("Couldn't create $ARGV[0].pad\n"); This doesn't work on DOS since it appends .pad to the input file name so it doesn't fit the 8.3 character format. The other errors you are getting are probably due to the difference between binary and ascii I/O mode on DOS. I forgot about that in my test script. Try this revision of pgppadt.pl, and let me know if it works on DOS: ----------------------cut here------------------------------ # Test program for pgppad.pl, showing how to use it. # Revised 2/5/95 for DOS legality # Usage: perl pgppadt.pl infile <bytes-to-add> # Output file is infile, stripped of extensions, with .pad appended. require 'pgppad.pl'; open (IN, $ARGV[0]) || die ("Couldn't open $ARGV[0]\n"); $outfile = $ARGV[0]; $outfile =~ s/\..*//; open (OUT, ">$outfile.pad") || die ("Couldn't create $outfile.pad\n"); binmode IN; binmode OUT; $padding = $ARGV[1]; @stat = stat(IN); $size = $stat[7]; print "Input file $ARGV[0] has size $size bytes\n"; print "Output file $outfile.pad will have size ",$size+$padding," bytes\n"; if (($code = &pgppad (IN, OUT, $size+$padding)) < 0) { die ("pgppad returns code $code\n"); } close (IN); close (OUT); print ("Done\n"); ----------------------cut here------------------------------

1 0

Judge not...
by anonymous-remailer＠shell.portal.com 17 Dec '03

17 Dec '03

Here's something I though would probably be of interest to more than a few here that read Usenet news. First article is the "Cancel FAQ" put out by the "Judges List", which many may have seen on the new.admin.* groups. Following it is the "Judges List" FAQ. I doubt this is a major threat, but is a good example of the sort of thing that needs to be nipped in the bud. If the "Cancel FAQ" doesn't trip any alarms, be sure to read the following "Judges List" FAQ. Please pardon the length. Articles follow: Article xxxxx of news.admin.misc: From: judges[tm](a)arch.ping.dk (NetNews Judges[TM] List) Newsgroups: news.admin.policy,comp.admin.policy,comp.security.misc,misc.legal.computing,news.admin.misc,news.groups.questions,news.groups,news.newusers.questions,alt.censorship,alt.comp.acad-freedom.talk,alt.current-events.net-abuse Subject: Cancel Messages: Frequently Asked Questions (FAQ) Date: Sat, 4 Feb 95 16:29:13 +0100 (CET) Organization: NetNews Judges[TM] List Message-ID: <01050105.limupb(a)arch.ping.dk> Reply-To: judges[tm](a)arch.ping.dk (NetNews Judges[TM] List) X-Mailer: uAccess - Macintosh Release: 1.6v2 Lines: 382 Cancel Messages: Frequently Asked Questions (FAQ). Ver. 2.0 --------------------------------------------------------- Summary: You can protect your reputation as a information source by cancelling articles posted under your name as soon as you discover that they are erroneous. Cancelling other's articles, however, can expose you, your site, and the Net as a whole, to serious threats. The sender should be notified when articles need to be cancelled. Disputes or doubtful cases can be directed to the Judges' List for resolution. --------------------------------------------------------- List of Frequently Asked Questions *** Purpose of this document 1) What are cancel messages? 2) How have cancel messages been used? 3) When should I issue a cancel message? 4) When should I not issue a cancel message? 5) What should I do when in doubt about whether a cancel is appropriate? 6) What should I do if I receive a request to cancel one of my own messages? 7) How can I request that a message be cancelled when I can not do so myself? 8) What should I do if I suspect one of my posts has been improperly cancelled? 9) Why should I follow these guidelines? *** Cancellation request form *** Contributors to this FAQ --------------------------------------------------------- *** Purpose of this document This FAQ list serves as an introduction to cancel messages. More advanced information is presented in the document "Administration of Cancel Messages", which is directed to News system administrators. The objectives and operational procedures of Judges' List are specified and explained in the Judges-L FAQ, also available as the Welcome message for new subscribers to the List. 1) What are cancel messages? Cancel messages are special USENET messages from a class known as "Control" messages. Control messages don't result in postings for people to read. Instead, they give instructions to the USENET server software at each site that gets the message. A Cancel message is a control message that indicates that a particular message (named through its unique Message-ID) should be deleted. 2) How have cancel messages been used? Most commonly they are used when a person posts something they want to delete, correct, or retract. One can cancel an old message and optionally issue a new one. (There is a better way to cancel and re-issue called Superseding, but most user programs do not support it.) Cancels are sometimes used by moderators of moderated newsgroups to delete messages that should not have been posted. 3) When should I issue a cancel message? You should issue a cancel message if it becomes necessary to delete, correct, or retract one of your own posts. You can issue a cancel message to delete a forgery: A message posted by someone else, that appears to be from you. A forgery can often be cancelled by you as if it were your own article. Such cancels should be reported to the Judges' List (see below). Some victims of forgeries don't cancel them, but consider it sufficient to post disclaimers to the affected newsgroups alerting everyone to the forgery. This is wise, in any case, since the forged message is likely to be seen by many people, even if it is cancelled. The newsreaders rn, nn, and trn use "C" as the cancel command. In Rusnews type ":cancel" while reading the message to be deleted. 4) When should I not issue a cancel message? It is not appropriate to cancel an article posted by someone else. However, a forgery is an exception. A low-tech form of "forgery" is to insert fabricated quotes in an article. Such posts should be followed-up with a correction and the poster notified. Misattribution of quotations, as opposed to fabrication, is often a result of inferior news-reading software. Such posts can be followed-up with a correction and the poster should be notified. It is inappropriate to cancel someone else's writing simply because you find the opinions expressed offensive. You can, however, ask the author to cancel the offensive post. Your news-reading software should permit you to enter the names of offensive posters in a "kill" file. Then, you will not see articles from that person in the future. Ask your administrator to upgrade your software if your news-reading software does not have this capability. The poster should be asked to cancel a chain letter or a libelous article, or any other post, such as an advertisement or pyramid scheme, that has inappropriate content. A copy of the request should be directed to the postmaster at the originating site, if the content is unlawful. 5) What should I do when in doubt about whether a cancel is appropriate? Unless you are certain the poster is acting in bad faith, you should explain to the poster why you object to the post and ask that it be deleted. State that you are considering submitting a complaint to the Judges' List and explain how one can participate in the dispute-resolution process (see below). You may explain how the offending post, or a post that accomplishes the same objective, can be broadcast. One source of information is a list of Frequently Asked Questions, "How to find the right place to post (FAQ)" in the newsgroup "news.newusers.questions". If the post comes from your site, contact your postmaster or News administrator for an opinion about whether the article violates the site's usage agreement. Doubtful cases can be submitted to the Judges' List, for an opinion. A complaint to the Judges' List must include a: :complete copy of the offending post. :complete copy of your letter to the poster, requesting that the post be cancelled. : complete copy of any reply from the poster or a News administrator (only if permission to forward is not refused). The complaint must have a subject line that starts with the characters "COMPLAINT: ", followed by the subject of the offending post. If you receive additional information after submitting a complaint or wish to withdraw the complaint, the same subject line should be used. Mail the complaint to JUDGES-L(a)UBVM.cc.buffalo.edu or JUDGES-L(a)UBVM.BITNET. If you wish to participate in the discussion of the complaint, email to LISTSERV(a)UBVM.cc.buffalo.edu or LISTSERV(a)UBVM.BITNET with only the command: SUB JUDGES-L <your-name-here> in the message body. Replace the <your-name-here>, with your first and last name, in the above command. Before making comments on the List, please wait for someone to respond to your complaint. If there is no response within a week, it is most likely that you did not file a satisfactory complaint (see above). If you find no error in your complaint and the problem still has not been resolved, send an additional message asking for help. 6) What should I do if I receive a request to cancel one of my own messages? If you do not agree that the request is valid, reply to the request with your reasons for not cancelling. If the request does not mention the Judges' List, refer to this FAQ and ask that you be notified if a complaint is submitted. 7) How can I request that a message be cancelled when I can not do so myself? If your software does not permit cancellation, or you are not confident in using it, ask your administrator to issue a cancel message for you. 8) What should I do if I suspect one of my posts has been improperly cancelled? If you suspect that your post has been improperly cancelled, contact your News administrator to rule out the possibility of a technical failure. If there has been an improper cancel, complete documentation should be directed to the Judges' List. Follow the procedures for submitting a complaint, but replace the characters "COMPLAINT: " with "CANCEL: ". 9) Why should I follow these guidelines? If you do not cancel erroneous articles that you have posted, you waste readers' time and damage your own reputation as a reliable source. Consider an article posted to a typical newsgroup with 36,000 readers, a post that takes an average of 1 second for each reader to deal with (i. e., examining the subject line) uses a total of ten man-hours (36,000 seconds / 3,600 seconds/hour = 10 hours). If the article uses up an average of four seconds, then the total time expenditure is 40 hours, the equivalent of a work-week. This is probably the minimum time expenditure on any article that is even selected for scanning. Thus, a few minutes spent cancelling an article can save a lot of time for readers. Groups with a lot of erroneous articles tend to lose readers, and articles posted to those groups reach a limited audience. By cancelling erroneous posts, you also reduce the risk that your name will be entered in numerous kill files, thus limiting the audience you can reach with your posts. Posting of apologies for faulty articles, such as those with an empty body, is to be avoided, since this wastes even more time of the reader, without supplying useful information. If you abuse the cancel facility, by cancelling other's articles, you force people to take defensive actions. Many administrators have disabled cancels, because they have been abused. The disabling of cancels means that you will not be able to delete all copies of your own posts. Therefore, if you cancel an erroneous article and then post a corrected one, readers at certain sites will first see the erroneous one and then the corrected one. They may think that the corrected one is a duplicate and not read it. This situation is obviously of no benefit to posters or readers. If your site's administrator has disabled cancels, it is wise for you to notify your administrator when you issue a cancel, so at least it can be deleted at your site. Disabling of cancels does not prevent a cancel from working at all, since the cancel message can delete the cancelled article before it gets to a site where cancels are disabled. Therefore, a cancel will be most effective if it is issued immediately after the target article is posted. In some cases, robot posters have been used to immediately repost articles that have been cancelled. This can rapidly lead to a newsgroup being overloaded at a site where the cancel facility has been disabled. This type of conflict situation should be avoided, since it makes the newsgroup unusable for most readers. Cancel messages must be used with extreme care. If you cancel articles based upon their content, you expose your site to a legal risk, since it can then be argued that you are taking responsibility for all articles' content, some of which may violate copyright or pornography laws, for example. Originating sites already are exposed to this risk, therefore, they should be allowed to cancel such posts. They should also have a user agreement that permits them to discipline a user, thus ensuring that repeated postings of this type do not occur, and thereby reducing the legal risk at all News sites. This is not a hypothetical risk. Many sites have been closed down, had their equipment seized, and their operators prosecuted, and in some cases jailed. The effectiveness of a common-carrier or "bookstore" type of defense has yet to be established (But see Mike Godwin <mnemonic(a)eff.org>. INTERNET LIBEL: IS THE PROVIDER RESPONSIBLE? Internet World, Nov./Dec. 1993) [URL: ftp://ftp.eff.org/pub/CAF/lawlibel_1.IW] The use of non-originating site cancels, to control content of articles, reduces the chances that such a defense can ever be established. This could make the risk of operating a News site unacceptable. Some sites have already been prohibited from operating moderated newsgroups or mailing lists by their legal counsel, because of the legal risk associated with assuming responsibility for content of distributed messages. The damage to USENET caused by the most extreme case of abusive posting was much less than that caused by external threats, even when those threats did not directly result in disruption of a News site. Therefore, in doubtful cases, cancel messages should not be issued. Notification of authorities external to the Net should be avoided. Contacts to the Press should be left to those who have been trained in public relations or have equivalent experience. In order to minimize legal liability, you should in, first instance, try to stop abusive posting at the source. By objecting directly to a poster of offensive material and making the person aware of this FAQ, you reduce the chances of repeated abuse. In the case, that a poster refuses to comply with a request to cancel an article and you refer the matter to the Judges' List, the conflict can be resolved in a way which benefits the Net in the long term, and reduces the risk that you will be subject to retaliation, for example, by having your posts improperly cancelled. Complaints to the Judges' List also permits a database of repeat abusers to be built up, with much more effective resolution of such cases then becoming possible. *** Cancellation request form Sample reply to a poster requesting cancellation or explanation (not usable in cases where only the opinions expressed are offensive): --------------------------------------------------------------- I object to your post, because .... Please cancel this post and notify me that you have done so, or explain why you will not delete the post. Please note that any reply may be forwarded to the JUDGES-L list, unless explicit objection is given. If I get no response, I will submit a complaint to the Judges' List. If you wish to participate in the discussion of the complaint on that List, email to LISTSERV(a)UBVM.cc.buffalo.edu or LISTSERV(a)UBVM.BITNET with only the command: SUB JUDGES-L <your-name-here> in the message body. Replace the <your-name-here>, with your first and last name, in the above command. --------------------------------------------------------------- *** Contributors to this FAQ include: Bob MacDowell <bobmacd(a)NETCOM.COM> Tim Pierce <twpierce(a)midway.uchicago.edu> David Stodolsky <david(a)arch.ping.dk> Dimitri Vulis <dlv(a)dm.com> ================================================ Posted on behalf of the NetNews Judges(TM) List. Judges-L Registrar, dss David S. Stodolsky, PhD * Social * Internet: david(a)arch.ping.dk Tornskadestien 2, st. th. * Research * Tel.: + 45 38 33 03 30 DK-2400 Copenhagen NV, Denmark * Methods * Fax: + 45 38 33 88 80 And now, the "Judges List" FAQ:

3 2

re finney's perl scripts
by anonymous-remailer＠shell.portal.com 17 Dec '03

17 Dec '03

hal sez , : The other errors you are getting are probably due to the difference : between binary and ascii I/O mode on DOS. I forgot about that : in my test script. Try this revision of pgppadt.pl, and let me know if : it works on DOS: works like a charm , thanks !

1 0