- cypherpunks-legacy@lists.ogf.org

Re: Secure voice software issues
by jim＠tadpole.com 12 Aug '93

12 Aug '93

V.42 adds about 200ms to my slip link. I'd rather do without it, thanks.

1 0

Re: [uk.transport] Speed Camera with OCR
by wcs＠anchor.ho.att.com 12 Aug '93

12 Aug '93

Rick Tait, R.Tait(a)bnr.co.uk, reports Marcus Reynold's note that there's a new > > speed trap in Kent. Apparently the radar clocks your speed, a TV camera > > records your number plate & then a real time OCR system picks out the > > actual number & flashes it up on a digital display a short distance > > further up the road. [...] > > The ability to read a car's plate & decypher the number in real time > > without any human intervention must take a bit of work. There's a lot of automated-highway research going on; most of it has been oriented towards radio+smartcards, which are easy to make accurate, and while they're not totally cheap, you can palm off much of the cost on the car-owner and use cheap equipment in the tollbooth. License-plate reading in real time is tough, if you're trying to get both speed and accuracy from a video image of a moving car; our neural-net folks were looking into it about three years ago, and it seemed like they'd be pretty good at it if they could get a bit more funding:-) ; they've mainly been concentrating on envelope-readers for Post Offices and railroad-car identifiers. Embarrassing speeders is a special case, because the accuracy doesn't really have to be 100% - you catch the cars on video, so if you decide to prosecute anybody, you can have a human read the picture and verify that you had the right license plate number. Meanwhile, back on the road, if your computer misidentifies the speeding driver half the time, he may assume the speed trap was nicking the other fast car next to him, and you've scared a lot of other drivers into slowing down because it *looks* like you've could have nicked them; so you've accomplished the safety objective. It's scary stuff, because the technology is mostly there, and a government that wanted to pay for development could get it done by a number of research companies. If it's too expensive for widespread use now, computer horsepower keeps getting cheaper every year, especially when what you need is DSP-crunchers and gate-arrays for neural nets, rather than general-purpose systems. Meanwhile, the smartcard people are mostly not using digicash, they're using no-privacy systems and appealing to the convenience for the commercial driver, whose company probably doesn't object. Dave Mandl comments: > Hmmm...most drivers I've known would be thrilled to get off with just > "embarrassment." In fact, they'd probably be proud to have it broadcast > to everyone else on the road (except the lawmen, naturally) that they'd > just been clocked at 110 MPH or whatever. A number of years ago, Ohio tried a system that displayed your speed (but not your license plate) as you went by - they quickly discovered that they had to cut off the display at something like 80 or 85 MPH to prevent just that effect (though adding the license plate number would cut down on the exhibitionism a bit.) Bill Stewart

1 0

Re: [uk.transport] Speed Camera with OCR
by dmandl＠lehman.com 12 Aug '93

12 Aug '93

> From: R.Tait(a)bnr.co.uk > I have just seen an item on the the BBC South East News about a new > speed trap in Kent. Apparently the radar clocks your speed, a TV camera > records your number plate & then a real time OCR system picks out the > actual number & flashes it up on a digital display a short distance > further up the road. The idea is to embarrass drivers by letting them > & others see their speed & registration. I don't know what type of > equipment they are using, but it sounds like a really neat trick. > The ability to read a car's plate & decypher the number in real time > without any human intervention must take a bit of work. Has anybody > heard how reliable this system really is ? Hmmm...most drivers I've known would be thrilled to get off with just "embarrassment." In fact, they'd probably be proud to have it broadcast to everyone else on the road (except the lawmen, naturally) that they'd just been clocked at 110 MPH or whatever. But somehow, I doubt that this data is used only for embarrassment, don't you? --Dave.

2 1

[uk.transport] Speed Camera with OCR
by R.Tait＠bnr.co.uk 12 Aug '93

12 Aug '93

I thought this might be of interest to fellow cypherpunks, especially to those of us in the UK. Has anything like this been done in the US or Canada? -Rick ------- Start of forwarded message ------- Newsgroups: uk.transport From: mreyno(a)sound.demon.co.uk (Marcus Reynolds) Subject: Speed Camera with OCR Reply-To: mreyno(a)sound.demon.co.uk Distribution: world X-Mailer: cppnews $Revision: 1.20 $ Organization: Sound & Vision BBS +44 (0)932 252323 Date: Wed, 11 Aug 1993 20:34:43 +0000 I have just seen an item on the the BBC South East News about a new speed trap in Kent. Apparently the radar clocks your speed, a TV camera records your number plate & then a real time OCR system picks out the actual number & flashes it up on a digital display a short distance further up the road. The idea is to embarrass drivers by letting them & others see their speed & registration. I don't know what type of equipment they are using, but it sounds like a really neat trick. The ability to read a car's plate & decypher the number in real time without any human intervention must take a bit of work. Has anybody heard how reliable this system really is ? I can see a few more uses for such a system than merely embarrassing speeding motorists, if it can work reliably & cheaply. How's this for starters:- City centre entry control & road pricing without the fancy toll booths, just an itemised bill at the end of the month. Tracking stolen vehicles, check every plate against the DVLC wanted list. Tracking any vehicle the State wants to track. Hello Big Brother. -- Marcus Reynolds (Bandwidth saving sig) ------- End of forwarded message -------

1 0

Physical Privacy
by Duncan Frissell 12 Aug '93

12 Aug '93

To: cypherpunks(a)toad.com W >This might be a step away from 'cypher' W >topics, but, Duncan, I was W >curious as to wether you publish any manuals W >on privacy, and/or can W >suggest books or sources for privacy and W >financial privacy info. Cypherpunks is dedicated to digital privacy. One cannot practice digital privacy without an understanding of physical privacy techniques (and vice-versa these days). Shameless marketing propoaganda will be sent by Email. Duncan Frissell "Privacy is a type of information that has its polarity reversed; I imagine it as anti-information. In a world where everything is connected to everything -- where connection and information and knowledge are dirt cheap -- then disconnection and anti-information and no-knowledge become expensive." -- Kevin Kelly But not *too* expensive. Frissell & Associates, custom crafted legal regimes since 1969. --- ~ WinQwk 2.0b#0 ~ Unregistered Evaluation Copy

1 0

Re: Key-Escrow (black) Humor
by George A. Gleason 12 Aug '93

12 Aug '93

How many government agents does it take to screw in a lightbulb? a1: None, they don't screw. a2: None, they only screw the general public. a3: That question is forbidden under Don't Ask / Don't Tell. -gg

2 1

Privacy and mail-lists
by Duncan Frissell 12 Aug '93

12 Aug '93

To: cypherpunks(a)toad.com W >This might be a step away from 'cypher' W >topics, but, Duncan, I was W >curious as to wether you publish any manuals W >on privacy, and/or can W >suggest books or sources for privacy and W >financial privacy info. Cypherpunks is dedicated to digital privacy. One cannot practice digital privacy without an understanding of physical privacy techniques (and vice-versa these days). Shameless marketing propoaganda will be sent by Email. Duncan Frissell "Privacy is a type of information that has its polarity reversed; I imagine it as anti-information. In a world where everything is connected to everything -- where connection and information and knowledge are dirt cheap -- then disconnection and anti-information and no-knowledge become expensive." -- Kevin Kelly But not *too* expensive. Frissell & Associates, custom crafted legal regimes since 1969. --- WinQwk 2.0b#0

1 0

Chaos harnessed for encryption / Fluctuations and Or
by gtoal＠an-teallach.com 12 Aug '93

12 Aug '93

In article <19930811185438.1.TK(a)ROCKY.AI.MIT.EDU> tk(a)reagan.ai.mit.edu writes: > with some op amps. If you want to play, there's an article in SciAm > this month on building a chaotic "encryption" machine. It probably > provides acceptable security if you use triple DES on signals prior to > sending them. That was smoke and mirrors by people who understand chaos better than they understand encryption. All it boils down to is a synchronised pair of (not very good) PRNGs. It's not a substitute for a one-time pad by a long chalk, which is how they seem to be using it. Basically they've just reinvented every schoolboy's exor encryption program all over again, this time with the latest 'chaos' buzzword thrown in to make it sound hip. Forget it. It's a dead end. G === Personal mail to gtoal(a)gtoal.com (I read it in the evenings) Business mail to gtoal(a)an-teallach.com (Be careful with the spelling!) Faxes to An Teallach Limited: 031 662 4678 Voice: 031 668 1550 x212

1 0

Secure voice software issues
by gtoal＠an-teallach.com 12 Aug '93

12 Aug '93

In article <9308111916.AA03336@servo> karn(a)qualcomm.com writes: > Voice calls are different, as the availability of a two-way path lets > you do things much more securely. If you generate a session key with > DH and use PGP/RSA *only to sign the exchanges*, not to encrypt the > session key, then even if your RSA secret key is later compromised, it > would not compromise those session keys that had already been created, > used and destroyed. Thanks for that explanation, that bit hadn't sunk in with me! This makes me think... something similar would be a good extension to SMTP wouldn't it? DH exchange of keys before sending point to point mail? With the user's public keys being picked up via their .mailrc or .pgpkey or something... (It would only happen if both SMTP's supported it and both users had made their public key available to the mail system) I'm thinking of ways of automatically and easily encoding all traffic by default, to avoid line snooping. I'm not suggesting this as an alternative to explicitly encrypting things you want to keep private. You could still do that too. G === Personal mail to gtoal(a)gtoal.com (I read it in the evenings) Business mail to gtoal(a)an-teallach.com (Be careful with the spelling!) Faxes to An Teallach Limited: 031 662 4678 Voice: 031 668 1550 x212

1 0

Secure voice software issues
by gtoal＠an-teallach.com 12 Aug '93

12 Aug '93

In article <9308110421.AA29116(a)jobe.shell.portal.com> hfinney(a)shell.portal.com writes: > A couple of comments on the cryptophone idea. > > First, there has seemed to be general agreement in our earlier discussions > of this concept that the hard part is compressing the voice to the point > where it can go over commonly-available modems. The government-standard > CELP algorithm is too slow for general-purpose home computers. You need > an algorithm that can operate in real time and compress intelligibly down > to about 13K bits per second. It has to be either able to compress and > decompress simultaneously or else you need some switching logic to decide > which person is talking and which is listening at each moment, with both > sides reversing roles in synchrony. Tony Robinson (who some of you might know already) - the author of the lossless 'shorten' sound compression program - is working on an ADPCM version. Currently compresses at 3 bits per (8 bit) sample. This plus a sampling rate of 4000hz instead of 8000hz (keeping it easy, on a sun) just squeezes in under the bandwidth limitation of v32bis, with a little left over for protocol overhead. Graham PS Has the 'netphone' list died or did I just drop off when I moved house like I lost my cypherpunks subscription? I remember there was talk of letting it drop because the situation seemed to be in control, but I haven't seen much progress since on vat et al, and the underground effort to do something on soundblasters that everyone said 'shut up about' has had more than enough time to come up with the goods... === Personal mail to gtoal(a)gtoal.com (I read it in the evenings) Business mail to gtoal(a)an-teallach.com (Be careful with the spelling!) Faxes to An Teallach Limited: 031 662 4678 Voice: 031 668 1550 x212

1 0