May 1993 - cypherpunks-legacy

Dolphin Encrypt
by Stanton McCandlish 19 May '93

19 May '93

<tried to email this, but it bounced> I too would like more info on this subject! Guess all us yokels at UNM are clueless or something. My question is, how does the recipient get the key, and how do they (she, whatever) know to use that long de command? What would happend if they didn't, just get gibberish? > > The recipient captures the entire message as, say, G.ENC, then runs: > > DE D G.ENC G.DEC /t > > (Of course, she has to know the encryption key.) Dolphin Encrypt > skips over P1 to get at C2 and writes G.DEC containing P2. Voila! > > For further info send me a snailmail address. See below for snail address... -- Testes saxi solidi! ********************** Podex opacus gravedinosus est! Stanton McCandlish, SysOp: Noise in the Void Data Center BBS IndraNet: 369:1/1 FidoNet: 1:301/2 Internet: anton(a)hydra.unm.edu Snail: 8020 Central SE #405, Albuquerque, New Mexico 87108 USA Data phone: +1-505-246-8515 (24hr, 1200-14400 v32bis, N-8-1) Vox phone: +1-505-247-3402 (bps rate varies, depends on if you woke me up...:)

1 0

No FTP? You can still get PGP *UTILITIES*!
by Stanton McCandlish 19 May '93

19 May '93

Due to legal threats from RSA, I cannot distribute PGP itself. I am not sure about the legalities of distributing source code for it. Anyone know? ANYWAY: if you need the latest PGP/RSA utils, likely useful for any number of encryption programs, and can't get them via FTP, try calling the BBS listed in the .sig below. Look in the CRYPTO file area. You may logon as Anonymous, password GUEST, if you wish. I have the latest versions of the utils distributed with PGP22, much improved over that release, and direct from the author. I also have Fido-tech mail utils, public key-rings, etc. Due to a corrupt user data file, the system will be down for a few days, but try calling AFTER May 20th, and get the goodies you need! Please keep in mind that after 11pm, and on weekends, tends to be the cheapest rate periods for most phone extortionists I mean companies. PS: there is no charge, other than standard phone extortion of course, for utilizing NitV BBS. Full access on first call. -- Testes saxi solidi! ********************** Podex opacus gravedinosus est! Stanton McCandlish, SysOp: Noise in the Void Data Center BBS IndraNet: 369:1/1 FidoNet: 1:301/2 Internet: anton(a)hydra.unm.edu Snail: 8020 Central SE #405, Albuquerque, New Mexico 87108 USA Data phone: +1-505-246-8515 (24hr, 1200-14400 v32bis, N-8-1) Vox phone: +1-505-247-3402 (bps rate varies, depends on if you woke me up...:)

1 0

REF ... GCHQ
by Clive 19 May '93

19 May '93

Ref the GCHQ question, I have posted to this group previously, ie yesterday, details on GCHQ, and I wish to take issue with >>>Duncan Frissell <76630.3577(a)CompuServe.COM> To: <cypherpunks(a)toad.com> Original Question >>>What does GCHQ stand for, and what does it do? Is it similar to the NSA? Duncan replied >> The Royal Signals Establishment at Cheltenham, Gloucestershire, >> England is known as General Communications Head Quarters (GCHQ). >> It is also called 'Cheltenham' after its location. >> It performs many of the same sigint jobs as NSA. THIS IS WRONG ..... GCHQ stands for the Government Communications HeadQuarters. They are part of the Foreign and Commonwealth Office. They have NOTHING to do with the Ministry of Defence. The Royal Signals Establishment, when it existed, it closed in about 1970 was located at Christchurch. It amalgamated with the Royal Radar Establishment at Malvern to form the Royal Signals & Radar Establishment (RSRE) with the prime site being at Malvern. They were once more reformed 2 years ago to become the Defence Research Agency, which includes the prime non-nuclear research establishments in the UK. FYI these are RSRE at Malvern, at Fort Halstead the Royal Armaments Research & Develeopment Establishment, at Farnborough the Royal Aircraft Establishment and at Portland, the Admiralty Research Establishment (They upset the monarchy about the turn of the century and as a result have never received the Royal Charter) Let me say again that GCHQ or it's groups have nothing at all to do with the MoD and that the Royal Signals Establishment (or even the Royal Signals & Radar Establishment) are not located at Cheltenham and are not associated with GCHQ!!!! Clive Walmsley Walmsley(a)ccint1.rsre.mod.uk

1 0

PGP and offline-readers
by anon03e2＠nyx.cs.du.edu 19 May '93

19 May '93

I am getting involved in networking some local BBS' and message bases. One of the primary functions of this network will be to serve as a library for PGP keys and to use PGP in sending messages. Now, if people read their messages offline, is there any offline readers that exist that integrate PGP (all this is IBM based) or is there an easy way to do this. Ideally, I want to avoid having to decrypt each message individually and would like PGP interfaced so that it could recognize a message header, apply th eright key, and output in the reader. I am probablky asking for too much. Does anyone have any experience/advice w/ this tpye of large scale implementation of PGP? I know the need to encrypt most of the messages will be small, but I think it is important to get these people so familar w/ encryption, they won't be scared and will defend it when the need arises.

3 2

Re: Question (GCHQ)
by sjw＠liberty.demon.co.uk 19 May '93

19 May '93

I live not too far away from Cheltenham, and I can confirm that the UK Government would have rather a hard job ever denying the existence of GCHQ. As you leave the M5 motorway at junction 11, taking the dual carriageway into Cheltenham, the signposts for GCHQ can't be missed. (Left at the second roundabout.) The building is visible to anyone driving into Cheltenham. (If you carry on towards the town centre for another mile or so, the Police HQ building has a rather impressive aerial on its roof, albeit comparable to other Police HQs.) It seems reasonable to believe that the various secret service agencies have reciprocal arrangements to tap each other's citizens. E.g. the American agencies tap the Royals' phones, then repeatedly rebroadcast the tapes using powerful transmitters, so they're almost certain to be picked up by some radio ham, or retired bank manager with a scanner, within a day or two. Steve Whitrow sjw(a)liberty.demon.co.uk

1 0

Mixing ciphertext and plaintext
by meyer 19 May '93

19 May '93

>Return-Path: <mdiehl(a)triton.unm.edu> >Subject: mixed plaintext and crypted text. >Date: Sat, 15 May 93 1:26:55 MDT > >Hey again, all. I've RTFM'd, and I don't even know if it can be done, but >does anyone know how I can have pgp display the un-encrypted AND the >decrypted parts of a message? Lets say my message goes like this: > >This is a plaintext message to tell you stuff that needs not be private. >BTW, can you find the subliminal message? ;^)j >-------- begin pgp block ----- >asdfkasjd;fThEjsflajslfjaslfjadfajs;flasjdaslfjasldfjalsj >asjlgfiuituqoNsAj;slkjasdifuoejrqlwfiasuokjel;rqwiuasdsii >asdfasljdflasjfdSuXsldfjalsjljeqljerwljrlejqlwjerlqwjerlq >elqrkjlasjl >134as >-------- end pgp block ------- > >Thanx a lot > >Yours virtually, >+-----------------------+-----------------------------+---------+ >| J. Michael Diehl ;-) | I thought I was wrong once. | PGP KEY | >| | But, I was mistaken. |available| If you wish to mix plaintext and ciphertext in an email message then you can use Dolphin Encrypt. Let's say your message is to consist of P1/C2/P3, plaintext/ciphertext/plaintext, where C2 is ciphertext derived from plaintext P2. What you do is put P2 into a file F.TXT, then run Dolphin Encrypt with a command such as: DE E F.TXT F.ENC /t After you give Dolphin Encrypt the encryption key F.ENC will be a file which consists of pseudo-uuencoded ciphertext (that's what the "/t" is for). Run your mailer program. Write your P1. ASCII-upload F.ENC, which is your C2. (No need to mark the beginning/end of the ciphertext block.) Finish off with your P3 and send your message. The recipient captures the entire message as, say, G.ENC, then runs: DE D G.ENC G.DEC /t (Of course, she has to know the encryption key.) Dolphin Encrypt skips over P1 to get at C2 and writes G.DEC containing P2. Voila! For further info send me a snailmail address. -- Peter Meyer

2 1

It was I
by TWEETYBIRD＠U.WASHINGTON.EDU 19 May '93

19 May '93

who was the source of the anonymous Mykotronix postings. Through certain leads I have been able to find more info and will let cypherpunks know as soon as is feasible. Please don't ask how I am doing it, I don't think I should reveal my sources. If anyone feels the need to verify my postings, please go ahead and post the results to this list. thank you, Tweetybird (Whistleblower

1 0

macpgp question
by Rusty Hoover 18 May '93

18 May '93

Hi. Can anyone answer a question about MacPGP? My MaccPGP 2.2 seems to have developed a curious problem. I'm using it on a Mac Classic with system software 6.07. Both yesterday and today, I received a message from a friend, encrypted with my public key. When I downloaded this message from my Unix account to my Mac, it appeared on my screen as an MS Word icon. (MS Word is the word processing program I use.) I launched MacPGP, and selected "Open/Decrypt" from the file menu, and up came the box with the list of files to choose from. I chose the just-downloaded message, and then, in the PGP message area, PGP gave me the following message: "File is encrypted. Secret key is required to read it. This message can only be read by: keyID: xxx You do not have the secret key needed to decrypt this file." But the key ID number listed **is** in fact MY valid, working key ID !! Would anyone have any suggestions about why this is happening? Thanks, enormously, in advance. =============================================================================== Rusty Hoover | PGP 2.2 public key | Too much time, so little to do ... rusty(a)panix.com | by finger & e-mail | no, reverse that ... ===============================================================================

1 0

Re: Neural Nets to decrypt?
by peb＠PROCASE.COM 18 May '93

18 May '93

In my experience, neural nets are good at generalizing across sparse data for recognizing patterns not seen before. GAs are more useful for converging (at an exponential rate giving Holland's schema theorem) on a solution to a problem. A GA is easier to train if the score is a continuous real number while most neural network implementations expect actual examples of what is in the set of things to be recognized. For a GA cryptoanalysis tool, a vector representing an experiment would be used as a genotype and the result could be the output of a specialized message detector (==1 if the text looks like plain English, ==.0001 if only a few words are seen, etc. (and of course it would need to detect file formats like that of compress)). Given this, a GA could find a solution. However, in learning theory, there are problems considered to be unlearnable and the standard example is encrypted information! The solution space could be like a plane with a single "needle" in it that is the solution with no hills in the general direction of the needle. This kind of solution space requires exhaustive search, unfortunately. It is difficult to characterize a solution space, but it is the key part--the mapping of a gene vector to a fraction representing the completeness of the solution is critical--and if it is completely flat with a needle, then it is not worth it. Alternatively, if it is completely random, then it also is not worth it. The solution space must be somewhere in these two extremes to be useful for a GA. Based on my limited experience with cbw (crypt breakers workbench), it is possible to get partial results (e.g., ex*lo*e -> explore and other words are then filled in) and zoom in the full solution, so based on that, a GA would be helpful. cbw is for an Enigma type machine and newer algorithms are much more sophisticated, so I don't know if the same kind of partial knowledge applies for RSA, DES3, IDEA cracking. Paul E. Baclace peb(a)procase.com

1 0

Neural Nets to decrypt?
by anon03e2＠nyx.cs.du.edu 18 May '93

18 May '93

I am naive on a couple of fronts, so apologies if this is off-base. As computing evolves, would it be very possible to use a neural net or like system to design an ideal code-breaking engine? From my u understanding, code-breakers are only as good as their designer's knowledge of the algorithim. Such a hypothetical system would develop its own cracking alogorithim which could be more efficient by several factors. I think another variation would the use of GA's to create a more efficient code-breaker. I know that GAs have been used to achieve near-ideal results in least-step sorting algorithims. Would code-breaking be the same in principle?

1 0