February 1993 - cypherpunks-legacy

Re: Tagging copyrighted text
by tcmay＠netcom.com 11 Feb '93

11 Feb '93

ld231782(a)longs.lance.colostate.edu (I'd use his real name, but this is all I can find here....why don't people sign their messages?) writes, commenting on some stuff by Marc Ringuette: >>Tim May and many of us argue that copyrights will become unenforceable >>as copying bits becomes cheaper and crypto privacy and anonymity becomes >>more widely available. This will mean that any static collection of >>bits will just be FREE. Musicians will have to make money on live >>performances, because they couldn't sell many recordings. Etc. > >Er, I'd say this is another instance of cypherpunk extremism: ``The Er, Marc R. didn't say the world as we know it is about to collapse, nor did I. (BTW, for you cypherpunks, most of this debate about intellectual property rights took place recently over on the Extropians list.) But I certainly think some big changes are coming. >world as we know it is about to collapse.'' Yes, information will >probably be much freer over the nets, and copyright violations probably >more difficult to pursue and punish. But think about this: with the >speed and ubiquity of networks, it now becomes possible for every >author or creative artist to keep the only copies of his work. Then, he >could post "links" to it anywhere in the world. When people pass around >the work, they wouldn't pass around the work itself, they'd pass the >"link". Whenever someone wants to view the work, the link points to >the unchanging address for instantaneous downloading. A very small With music, art, text, or executable code, once it is available for listening, viewing, reading, or running on the customer's system, it is also available for copying. This is the whole point--if you can read the text, you can keep a copy of it. You mention this, but dismiss it as inconvenient. Well, a popular song may be simultaneously requested by tens of thousands of listeners....it is hard to imagine the 14 Gigabits per second (10,000 x CD-quality sound) coming out of someone's home or business to supply this need! The situation is vastly worse with video, naturally. (I suppose some "servers," like ftp sites, could be set up to deal with especially popular movies and music...cumbersome, though. And, yes, fiber optics can in theory support very high bandwidths, but still not enough to supply the video rental biz. And this traffic will clog the nets and cause lots of problems, I would think.) >transaction charge is billed to the receiver by the sender. (Of course, >people could write stuff that would actually grab and store the text or >whatever instead of just "play" it. But I'll bet that most people will >eventually say, "why bother?" The direct access will be more convenient >and the charge so minimal.) Links could be embedded wherever there is >information, like in bibliographies or references or compilations or whatever. I agree that hypertext-like embeddings offer some hope, for things like text especially, as a reader is essentially buying the total package of texts and links, rather than just some specific text. (But any specific text becomes vulnerable to "snagging.") >Hey, consider that programmers are artists too. I write some hot >program, but I don't distribute it: I run it on one of my systems as a >sort of network pipe. People anywhere in the world pipe in their data >and get it back with a small transaction fee to me. In fact, what I'm ..... Even with much faster networks, remotely running programs introduces lags and inconveniences, not to mention security problems (e.g., "Now you can use _our_ word processor at _our_ site to write your love letters!"). If nothing else, speed of light delays will become very irksome. But I do agree that "knowledge servers" may be a way for corporate knowledge to remain bottled up in corporations (for example). -Tim -- Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)netcom.com | anonymous networks, digital pseudonyms, zero 408-688-5409 | knowledge, reputations, information markets, W.A.S.T.E.: Aptos, CA | black markets, collapse of governments. Higher Power: 2^756839 | Public Key: MailSafe and PGP available.

1 0

Tagging copyrighted text
by Marc.Ringuette＠GS80.SP.CS.CMU.EDU 11 Feb '93

11 Feb '93

Tim May and many of us argue that copyrights will become unenforceable as copying bits becomes cheaper and crypto privacy and anonymity becomes more widely available. This will mean that any static collection of bits will just be FREE. Musicians will have to make money on live performances, because they couldn't sell many recordings. Etc. A question I've been thinking about is, will Dow Jones be able to charge for its newswire? Step 1: I subscribe to Dow Jones and then relay each message to a mailing list, charging them a fraction of the original price. Step 2: Dow Jones starts changing random whitespace in the text, in an attempt to "tag" the text untraceably to trace which subscriber is leaking the information. They cancel my subscription. Step 3: I get 20 new subscriptions, and use the redundancy to cancel out Dow Jones's sneaky tagging. Step 4: ... Question: who wins? I haven't been able to work it out yet, but it may just be a simple combinatorial exercise. -- Marc Ringuette (mnr(a)cs.cmu.edu)

3 2

Re: Tagging copyrighted text
by fen＠genmagic.genmagic.com 11 Feb '93

11 Feb '93

Marc Ringuette (mnr(a)cs.cmu.edu) writes: > True, the abandonment of copyright isn't going to happen right away. I agree, too. But there are many things here to look at. Economics is a strange bird, indeed, and I pretend to know little of it. But here are some points that the future may bring into focus... 1) It is easy to copy software (music, programs, video, etc.) 2) No matter what copy protection is developed, it will always be easy. 3) People will always pay for the *production* of software (a program that needs to be written, or live music, or...) (note that the Grateful Dead, top money maker, invites tapers) 4) People will always pay for support, maintenance, storage, etc. (note cygnus) 5) Storage and indexing become more valuable as info-quantity grows. 6) Likewise, effecient and accurate filtering (access) 7) People will pay for access to the info they want *now*. 8) Human editors will team with electronic filters to provide this service. In conclusion, there will always be money to be made. In (one of) the more radical futures (that I happen to ascribe to) there will be less income from past works (ala software patents) and more for storage and timely access. But the developer of good software (the musician) will always be in demand. Fen

1 0

Re: Compressed/Encrypted Voice using Modems
by Phil Karn 11 Feb '93

11 Feb '93

At 0:19 2/10/93 -0500, Stephen D. Williams wrote: >Also, from a previous note, you wouldn't want to turn off V.42/V.42bis >since that is where the error correction is. Also, even on compressed >data, you get some additional bandwidth because it does packetized >synchronous data. This gets close to 8bits/byte instead of 10 (start, >stop). As somebody else mentioned, you do want to turn off LAPM/V.42bis when running a speech application, because speech is a real-time application and LAPM retransmissions introduce delay. Compression is not likely to add much but delay either, since the vocoder itself is already compressing the speech much more effectively than V.42bis, which is designed specifically for textual information. Almost all modern modems support synchronous terminals, although the feature is rarely used. This is how you get around wasting 2 bits out of every 10 on start and stop bits. Phil

1 0

Re: Tagging copyrighted text
by Marc.Ringuette＠GS80.SP.CS.CMU.EDU 11 Feb '93

11 Feb '93

Dear ld23178, True, the abandonment of copyright isn't going to happen right away. But I disagree with your claim that on the network of the future the artist will make a few pennies per copy by being the most convenient provider of the information. The author is very unlikely to be the most convenient and least expensive provider of their work, since they have no advantage over specialists in the information providing business. Your point about "unix pipes" across the network is an excellent one, though. Information filtering and processing services will be very valuable and harder to obtain for free. "Insert a 25c electronic coin and a pile of news articles, and I'll filter them for you." -- Marc Ringuette (mnr(a)cs.cmu.edu)

1 0

Re: Wired photo shoot at cypherpunks on Saturday
by Kevin Kelly 11 Feb '93

11 Feb '93

Thanks.

1 0

Wired photo shoot at cypherpunks on Saturday
by Eric Hughes 11 Feb '93

11 Feb '93

Kevin Kelly, editor of Wired Magazine, former list member, and former editor of Whole Earth Review, asked me to forward the following message to the list. OK, everybody, it's your chance for fame or the avoidance thereof. Eric ----------------------------------------------------------------------------- WIRED magazine is running a story by Steven Levy (Hackers, Artificial Life) on crypto rebels and the current politics of encryption. The story takes an extremely sympathetic view of the work done by this list. To illustrate this story, WIRED will have a professional photographer at Cygnus at next Satuday's cyherpunk meeting. He will set up outside of the meeting area somewhere. WIRED would like to shoot an anonymous group picture, for those willing to participate. We propose supplying blank face masks upon with individuals can write their public key. In addition we will be shooting non-anonymous portraits of some individuals, if they are willing. We would like it to be fun. We emphasize that this is all voluntary and will take place away from the meeting area, perhaps before and after the meeting. If you have any questions about this you can email me, Kevin Kelly, editor at WIRED, kk(a)well.sf.ca.us We think this article will make a big splash by bringing the issues you care about into the forefront. Having real people behind the ideas leads credence to tthe movement. We seek your cooperation.

1 0

Crypto goals
by Marc.Ringuette＠GS80.SP.CS.CMU.EDU 10 Feb '93

10 Feb '93

Y'know, just defining "what we should want" is one of the biggest challenges in moving towards the next generation of crypto. How about making a list of features we want, and/or a list of scenarios we want to be able to handle? Uncontroversial features: message privacy message anonymity message untraceability digital cash digital time stamps More difficult/controversial features: message invisibility [e.g. in low bits of video signal] untraceable access to databases [just anonymous email?] electronic black markets [just electronic cash and anonymous email?] electronic voting m-of-n trust schemes time-delay messages [are they useful?] executable but non-disassemblable code [is it possible?] tagging / un-tagging of copyrighted text [what is possible?] [explanation to follow in another message ] I know how to take a crack at implementing all of these except the last two, but would benefit from better theoretical and practical work on all of them. Most of them are obviously useful, and scenarios come easily to mind. Have I missed anything? -- Marc Ringuette (mnr(a)cs.cmu.edu)

1 0

Re: timed-release crypto
by Karl Barrus 10 Feb '93

10 Feb '93

Tim May writes: >...I could publicly post an encrypted message to this list, or in one >of the "pools," and then scatter the decryption key in several pieces >with several members of this list, paying them $1 each to "hold" their >piece for, say, a month. At the end of the month, they would fulfill >their end of the bargain by forwarding the piece they hold to some >public place or pool and the decryption key would be reconstituted... Cypherpunks, this is an experiment we can try right here right now. We certainly have the anonymous mail facilities, pools, and encryption, as well as a digital bank we could use to "pay" people :-) Splitting a decryption key is fairly easy: we could use the simple XOR method. Say you want to encrypt a message of length m and split the key among n people. Choose n-1 sequences of m random bits (coin flips, cryptographically strong pseudo-random number sequences, etc) and XOR the message with all the random sequences (call the result M'). Then your split secret is the M' and the n-1 sequences. To recover the message, calculate the XOR of M' and the n-1 sequences. Providing for an n-out-of-m scheme takes more work and number theory, but that can be done as well. /-----------------------------------\ | Karl L. Barrus | | elee9sf(a)menudo.uh.edu | <- preferred address | barrus(a)tree.egr.uh.edu (NeXTMail) | \-----------------------------------/

1 0

Speaking in Tongues
by deltorto＠aol.com 10 Feb '93

10 Feb '93

>>From: dmandl(a)shearson.com >>Subj: Re: E Pluribus Unum >> >>> From: deltorto(a)aol.com >>> >>> Fellow Cyphers, >>> >>> Eric Hughes contributed: >>> >>I applaud the Clinton administration for making itself available via >>> >>email. I do not think it advisable, however, to send a single >>> >>cypherpunks letter. Rather I urge all interested parties to compose >>> >>their own letters [etc., etc., etc., etc.] >> >>I think I've held my tongue long enough. Hmmmm. Perhaps not. :-D ddt PS: I can't afford six months, but I like (the symbolism of) the idea. There are much more effective ways of pulling their plug, no?

1 0