My favorite example for disputing Slashdot's law of universal crack-ability is digital cable in the USA: PowerKEY (Scientific-Atlanta) and DigiCipher (General Instruments/Motorola) have not been cracked.
Cryptography actually does work if implemented correctly. It's true brute force can be applied to most methods, but it's impractical if the message is long enough (e.g. a movie) and the key is changed often enough. How long would it take to decrypt a movie encrypted with GPG if each 1 megabyte chunk were encrypted seperately?
As far as attacking the keystream, remember, you need a crib (see Cryptonomicon) to attempt a brute force attack. If they keys are truly random (see early Netscape for a counter-example), you don't have a crib. You don't know when your brute force attack has found the right key.
There are pseudorandom number generators that have exceptionally long periods, but if they wanted, cable companies can afford to install 1U hardware truly-random number generators in every headend.
Also, I know it's obsolete, but Circuit City's Divx was never cracked. They expected they only had a few years
SD (secure digital) cards haven't been cracked either. There's probably not enough good stuff on 'em to bother yet, but when Music Net & Pressplay start offering downloads to "secure" portable players, I expect it will take more than a year to crack it. I'd put money on it.
Sony & Panasonic's smallest 1080i sets are 32" for somewhere between $1500 & $2000, so you're right that a typical store probably won't have a sub-$1k set. (CT-32HX41 Panasonic & KV-32HS20 Sony).
I wish the 480p EDTV spec had gone somewhere. There's got to be some middle ground between the $450 27" Sony I got & that $900 Samsung. There's 480i Wegas (optionally XBR) with a nicer tube, but it's still interlaced dammit.
Except for small unidentifyable electronic components, I suppose most of the devices you carry are at least closed boxes. The hardware engineers I work with frequently carry prototypes back & forth between Japan or Taiwan & California.
These are digital cameras with missing or wrong manufacturer markings on them, often with bad fit between the plastic shell and the switches (which are soldered to PCBs inside) and usually a serial port dongle (so we can develop software for them) hanging by its wires which enter the body of the camera through a rough-hewn hole. They often don't work properly or boot with a lot of nasty beeps as they enter the calibration code (thus they can't be demonstrated as cameras to screeners).
They're often more delicate and certainly more valuable than a final-production camera-- there are less than 20 in the world and there are blue wires on the PCBAs. Therefore, they're usually in the engineer's carry on.
They look about as much like a small poorly-disguised bomb as you might want, but no-one's been hassled yet.
Nintendo claims they never take a loss on hardware sales (though that's hard to believe considering the Gamecube only costs $199).
Tricky accounting might let Nintendo say this without technically lying even if the real cost of all the parts + putting 'em together exceed the wholesale cost of a Gamecube.
Suppose IBM, MoSys and NEC want Nintendo to be successful and they're also on the weak side in the negotiation (don't know if that was the case, but read on). Nintendo negotiates incredibly cheap prices for the ASICS so the Gamecube really is cheap (to Nintendo) to build.
The chip guys aren't making much or are maybe even losing some money, but the contract guarantees a certain number of chips per quarter (at a particular price) will be bought over the next few years & they believe their process guys can get the cost down & they'll eventually make money.
It's the same idea as sell below cost now, but reduce the price slower than the cost falls. You're just moving the loss (and the potential profit gain from cost reductions) outside the primary manufacturer.
Of course, what really happens is contracts always get renegotiated so if Nintendo doesn't sell enough, they buy fewer chips & the penalty is softer than originally agreed. If the chip guys drive their cost way down & Nintendo gets wind of it, they renegotiate the price down.
Oh, and the engineers who enabled the cost reductions get nothing but complaints about how they didn't do enough or how fallout went up because 20 cents was saved on a crappier regulator.
I just don't believe this. Copyright owners have sued novel devices claiming contributory infringement, but they usually lose (RIAA v. Diamond, Universal v. Sony). Photocopiers still exist without a "paper tax" paid to book publishers.
Kodak does support watermark detection and won't copy prints made on Portra III paper, but they've discontinued the paper because "the reality is that there are many more non-Kodak methods for making high quality copies that are readily available to consumers today."
http://www.hjpro.com/news_copyright.cfm
I still see Kodak Picture Maker kiosks at my local stores....
how far will MS take their 'hardware lust'? how close to being Apple will they get before the afore mentioned giants turn on them?
Whoa!
Good fscking point!
However, an Xbox2++ might eat away at the home market for PCs which doesn't seem that profitable anyway. Microsoft doesn't seem interested in going after the higher margin powerdesktop/workstation/server market. Might Dell & Co. just let 'em have a bit of home & concentrate on business?
Grabbing the mpeg-2 stream from a satellite box is a great idea. Something like 5C encryption on the stream piped over USB would keep the MPAA happier.
The TiVo uses a hardware real-time encoder because the main CPU is wimpy. A Pentium 4 @ 2.2GHz is probably getting close to doing MPEG-2 @ 720x480x60 in real time with some quality. Creative's requirements (Video Blaster Movie Maker) for 352x240 MPEG-1 or 2 (at probably reduced quality as well as resolution) is a 350MHz PIII.
Instead of compromising on quality, though, you could add an external MPEG encoder, connected via USB like the ADS Tech. Instant DVD (under $200). That's expensive now (compared with $300 for the Xbox), but give it time.
{Yes, full-speed USB is capable of carrying video. 12 megabits/sec > 5-6 mbps of DVD. USB won't carry lightly compressed video like DV (3.6 megaBYTES/sec). For that you do need firewire or high-speed USB.}
The Xbox's hard drives are little small for PVR, though. An external USB connected drive with encryption would be a good idea.
If we're putting the hard drive & possibly video encoder in an external box, the need for the xbox is diminished, though. If we're doing analog PVR from any source, the xbox can provide UI & control. If we're doing PVR from satellite, though, might as well run the UI on the satellite box & make the add-on just a USB hard drive with CPRM.
Suppose your an IT manager in a company that uses non-libre software.
You want to keep on the up & up with licenses 'cause you a) think it's wrong to copy w/o 'em b) don't want the BSA hassle and/or c) want to keep track of what machines run what software so you can keep 'em from breaking.
So you do your own internal audits. When the BSA sends their threats, you say "we've audited & we're compliant." When they tell you to turn over the results of the audit you tell them politely to fuck off & stop wasting their time. Maybe mentioning the method by which you audit calms 'em down a bit.
I suppose this will piss them off even more than if you hadn't done an audit, but their attitude that only they know how to count might annoy a judge enough do deny an order to hand over your audit or let the BSA do one of their own.
Bands playing covers are supposed to pay royalties to the songwriter (usually through the Harry Fox Agency).
The only way to avoid this mess is to see music & theater that's 1) written by the performers 2) written by people that died a LONG time ago 3) written by commission from the performers (San Francisco Mime Troupe).
Even public domain plays and scores can be a problem because a derivative work of a public domain work can be copyrighted (e.g. a modern arrangement of a jazz standard based on a Tin Pan Alley Tune (which might be barely old enough to be p.d. in the first place).
Whether or not the DMCA considers intent is still a matter of debate. Ashcroft, in his department's reply brief in the Felten v. RIAA case said that intent was paramount- Felten didn't hack SDMI to copy some Britney toons, but for research, so it wasn't actionable (& I guess the warning letter was just bad advice). There is also a specific provision for research, but the Justice Department specifically mentioned intent.
(2) No person shall manufacture, import, offer to the public, provide, or otherwise traffic in any technology, product, service, device, component, or part thereof, that -
(A) is primarily designed or produced for the purpose of circumventing a technological measure that effectively controls access to a work protected under this title;
(B) has only limited commercially significant purpose or use other than to circumvent a technological measure that effectively controls access to a work protected under this title; or
(C) is marketed by that person or another acting in concert with that person with that person's knowledge for use in circumventing a technological measure that effectively controls access to a work protected under this title.
This, to me, looks kind of like the four tests of fair use. Unfortunately, it's not the intention of whether the device was intended for infringing use, but whether it was intended for use in circumvention (even if otherwise legal) at all that's tested.
This looks like a pretty awful contradiction. 1201c1 says "(c) Other Rights, Etc., Not Affected. - (1) Nothing in this section shall affect rights, remedies, limitations, or defenses to copyright infringement, including fair use, under this title." so you're supposedly still entitled to fair use, but there's no way to exercise that right.
What makes you think that this functionality cannot be integrated onto the same chip that does the normal digital-to-analog conversion?
Yep- this is the key technology making encrypted content MUCH harder to access in "pristine" form. For example, Broadcom has a digital cable "settop on a chip" that has the CPU, MPEG decoder, decryption hardware (DES3 in realtime needs hardware 'cause CPU is too wimpy) and video DAC on one chip.
Soundcards with signed drivers are another step in this direction.
One weakness is that keys must still be passed around. Some of the schemes for protecting session keys are quite good and the key decrypters are also in hardware, though sometimes not on the same chip (though they use encrypted comm. btwn chips).
Simple software-only cracks will be a thing of the past once the encryption moves out of software and into the video and soundcards, with DVD-ROM drives and network interfaces participating in secure channels for keystreams.
Be, Inc.'s auction next week includes a few logic analyzers & high speed o'scopes. Get 'em now before they're declared "hackers tools."
Stupidity + Monopoly = Success beyond your wildest dreams.
A few short years ago when IBM wasn't looking so hot (Taligent/Kaleida) and the Baby Bells were going great gangbusters (partially due to the Telecommunications Act), pundits commenting on the Microsoft case thought a breakup might actually be good for MS.
The thought was that IBM, which didn't get broken up, had to be vewwwy careful to avoid the appearance of abusing monopoly power, giving rise to DEC and eventually Compaq/Dell/Microsoft. If their lawyers hadn't held them back from using more than FUD against the mini, we might still be using mainframes.
The baby bells and AT&T were allowed to do business with fewer restriction. There's regular FCC tariff crap, but the Telecomm Act helped ease that. For a while, AT&T and some of the babies were doing quite well. In some respects, Verizon & esp. SBC might still be said to be doing well. Their customer service may suck worse than ever, but are they making money?
....the Sound Blaster 16. 44.1kHz 16-bit stereo is CD quality...
Someone else has already pointed this out about the SB16, but I'll do it again a different way.
You are confusing word width measurements in different parts of the audio chain. I agree that 16 bits is enough for most music in most listening environments as a distribution format.
Let's divide the audio processing chain into a few parts. How 'bout
assuming an all-digital chain from the ADC at the studio to the DAC at the listener.
You want more than 16 bits in acquisition because you don't want to have to either apply heavy compression in the analog domain, ride the gain like crazy (manual compression) or lose the quiet passages in the noise.
After acquisition, you could do some AGC/compression before storing the samples in a shorter word, but storage is cheap.
When you do processing or mixing, you end up adding samples from multiple source (multiple tracks or the outputs of multiple delay lines for things like reverb). This increases the word size ( 0x7fff + 0x7fff = 0xfffe which doesn't fit in a 16-bit signed word). You can round towards zero (please don't truncate) to get rid of the new LSB, but if you keep the wider width until the bitter end you'll introduce fewer artifacts (remember, we might be talking about a 128 channel digital console!).
You can master to 16 bits if your output is going to be CD, but you might want to do the compression/equalization stuff as one step to a 24 bit master then do something like Sony's Super Bit Mapping to get 16 bits. You also might be mastering for release in both DVD-A/SACD and CD.
Distribution in 16 bits is cool, but the only way to make sure all 16 bits are actually information and not noise is to handle the above steps properly (see the RHCP Californiacation as a counter-example).
Reproduction through a perfect 16 bit DAC would be fine, but most 16 bit linear DACs only gave you (when people still used 'em) 14 to 15 1/2 bits of information. Oversampling filters increases word width and there is real benefit to converting those bits which is why you'll see things like 8x 18bit outputs (on 10 year old audiophile CD players). Today, everything seems to use delta-sigma DACs with very high sample rates and effective word widths of well over 16 bits.
The last thing on the soundcard is the analog path from the DAC to the jack. The SB16 puts the DAC output through an incredibly crappy digitally controlled analog mixer along with the CD, synth and other sources.
That's why you don't get 96dB of SQNR out of an ostensibly 16 bit soundcard.
---
Soundcards aren't only for reproduction. They are production tools also and the Audigy with a 1394 interface and front-mount I/O certainly fits the bill for amateur production. There is a spectrum of artist abilities and budgets so it's unreasonable to expect everyone who's authoring to have high-dollar equipment.
---
What's a "standard PCI Sound Blaster 16" by the way? The "Live" was the first PCI soundcard from Creative (not counting rebadged Ensoniq cards) and if there's some sound improvement to go with the bus improvement, I'll take it.
It's odd that the software I trust the least to protect my assets (my work on the hard drive that I'd prefer a worm not eat) is trusted the most by the OS to protect someone else's assets.
Consider digital cable set-top boxes (Scientific-Atlanta Explorers and Motorola DCTs). Security still unbroken.
Private key is inside a secure micro that is never supposed to give it up. It only decrypts session keys (which change every few seconds so can't effectively be shared).
It's true that at this point the unencrypted content is in the clear on a couple of memory busses (possibly mixed with a lot of other traffic), but the content tends to be ABC, HBO and pay-per-view movies. In other words, not that valuable. The important thing is securing the access for folks that don't pay not the content for those who do.
In any case, set-tops-on-a-chip like Broadcom's mean that the CPU and MPEG decoder are on the same die. Moving the DENC (digital to analog converter for video) onto the MPEG decoder eliminated one exposed path & this will eliminate some others. They'll have to get embedded DRAM on the chip to really make it secure though. Then it'll be output of the QAM decoder in, analog video out.
These techniques can be used for PCs too. Soundcards are just the beginning-- look for secure CPUs, hard disks and video cards in the future.
This stepping-stone to a secure operating system is sometimes called "Secure Boot."
It's also called the xbox.
That's why there are folks attacking the BIOS of the xbox (search your favorite p2p network for xbox.bin), trying to figure out how to boot unsigned software, over at xboxhacker
But dvbackup barely works. It doesn't even control the transport directly, which every $100 DV video editing package is capable of.
I think this would be great as the tapes are cheap and available at grocery stores (I'm thinking Digital8 which can use 8mm or Hi8 tapes and is miniDV-compatible). The camcorders start around $700 which isn't that cheap, but not bad for a backup drive you can take with you & make kid videos.
It's funny, though, that the reason I peeked at this article in the first place is that I have many gigs of kid video for which the only backup is the VHS (bad!) tapes I've sent my parents.
The "dailies" and a copy of the edited-down project on 8mm will survive a disk crash, but I'd have to re-build the whole project just to do a small tweak (not a big deal for kid videos, but you might feel differently about your Episode II spoof) and I worry about fire, earthquake & theft.
If I had another camcorder, I could at least back up the dailies.
Besides digital video & photography, all my work would probably fit on a single CD.
It's a strange twist that commercial video houses which might survive without every scrap of video they've ever worked with can afford backup easier than parents who don't want to lose one frame.
A generic (Honda, or something) gasoline generator is only a hundred bucks or so
Honda doesn't make generic generators, Coleman does. Even those aren't $100. I just bought a 5kw Coleman @ Costco for $440. It's the cheaper series with the Tecumseh side-valve 4 cycle engine (instead of the overhead valve Briggs & Stratton or an overhead cam Honda).
It is LOUD. I'd read about this before I bought it, but I had no idea until I started it up. I went running for the hearing protection (muffs- I hate plugs) right away. I'd planned to break it in for 1/2 hour, but I
A good Honda generator in the 5kw class costs more like $1900 (Northern Tool catalog). You can get Honda engines in gensets by other mfg.'s for less but Honda makes several classes of engine.
It's not just the noise (which is why I delved into this article in the first place- I look forward to the day when quiet generation is cheaper)- the cheap gens eat themselves up in no time- something like 100 hours for mine (it does have the cast iron cylinder liner for maybe 2x the life).
Smaller gens are cheaper- a 50cc, 1kw, 57dB @ 7.7 yards Honda is $690 and you can run a computer off it, but not the blower motor in your furnace, your refrigerator or your septic pump (my house's big draws), so put on a sweater & don't bother with that bathroom break & forgeddabout that steak.
Um, what technology does your sub-$4000 RPTV use? Video projectors in general use LCD, DLP (Texas Instruments' micro-mirror tech) or a set of 3 cathode ray tubes (one each for red, green and blue). RPTVs tend to use CRTs.
No shadow mask removes some problems with direct-view CRTs, but introduces other ones.
Direct-view tubes have a very wide field of view, matched only by expensive panels.
-M
Ashcroft/DOJ says circumvention research is legal
on
DMCA 2, Freedom 0
·
· Score: 1
One thing that came out of this is that the DOJ claimed that the *intent* of creating a circumvention device matters. If it's used only to demonstrate a weakness, it's kosher. We'll see whether they hold to their own opinion when they're on the other side of the aisle.
"Plaintiffs' alleged conduct is not proscribed by the statute"
"While Plaintiffs' computer programs have the additional capability of actually circumventing access controls, they are allegedly not designed or marketed for the purposes of actually getting access to the copyrighted material itself."
The DOJ asserts that Dmitry fails the "is primarily designed or produced for the purpose of
circumventing protection afforded by a technological measure" test of 1201(b) while ignoring 1201(c):
(c) Other Rights, Etc., Not Affected. - (1) Nothing in this section shall affect rights, remedies, limitations, or defenses to copyright infringement, including fair use, under this title.
(2) Nothing in this section shall enlarge or diminish vicarious or contributory liability for copyright infringement in connection with any technology, product, service, device, component, or part thereof.
(3) Nothing in this section shall require that the design of, or design and selection of parts and components for, a consumer electronics, telecommunications, or computing product provide for a response to any particular technological measure, so long as such part or component, or the product in which such part or component is integrated, does not otherwise fall within the prohibitions of subsection (a)(2) or (b)(1).
(4) Nothing in this section shall enlarge or diminish any rights of free speech or the press for activities using consumer electronics, telecommunications, or computing products.
By the way, USB-connected cable modems (if they follow the trend of DSL modems) are going to make NAT more of a pain in the ass. They can play all sorts of nasty games in the class driver on the host that will be hard and/or illegal to replicate on Linux or BSD.
Slashdot Mirror
C-SPAN only covers some committee hearings & they might be covering this, but their website is misbehaving right now so I can't tell for sure.
Capitol Hearings will have it via Real audio starting at 9:30 A.M. (I assume that's Eastern time).
They don't offer archives.
I'm glad I got streambox before it was sued out of existance!
My favorite example for disputing Slashdot's law of universal crack-ability is digital cable in the USA: PowerKEY (Scientific-Atlanta) and DigiCipher (General Instruments/Motorola) have not been cracked.
Cryptography actually does work if implemented correctly. It's true brute force can be applied to most methods, but it's impractical if the message is long enough (e.g. a movie) and the key is changed often enough. How long would it take to decrypt a movie encrypted with GPG if each 1 megabyte chunk were encrypted seperately?
As far as attacking the keystream, remember, you need a crib (see Cryptonomicon) to attempt a brute force attack. If they keys are truly random (see early Netscape for a counter-example), you don't have a crib. You don't know when your brute force attack has found the right key.
There are pseudorandom number generators that have exceptionally long periods, but if they wanted, cable companies can afford to install 1U hardware truly-random number generators in every headend.
Also, I know it's obsolete, but Circuit City's Divx was never cracked. They expected they only had a few years
SD (secure digital) cards haven't been cracked either. There's probably not enough good stuff on 'em to bother yet, but when Music Net & Pressplay start offering downloads to "secure" portable players, I expect it will take more than a year to crack it. I'd put money on it.
-M
Can I tell you a story about how the 2000HD got developed?
Nah, better not, but kudos to the guys that developed it that it works at all.
I have yet to see in the stores an HDTV which is either 19" or 25" or 27".
Here's a 27" HDTV for 900 bucks.
Sony & Panasonic's smallest 1080i sets are 32" for somewhere between $1500 & $2000, so you're right that a typical store probably won't have a sub-$1k set. (CT-32HX41 Panasonic & KV-32HS20 Sony).
I wish the 480p EDTV spec had gone somewhere. There's got to be some middle ground between the $450 27" Sony I got & that $900 Samsung. There's 480i Wegas (optionally XBR) with a nicer tube, but it's still interlaced dammit.
Except for small unidentifyable electronic components, I suppose most of the devices you carry are at least closed boxes. The hardware engineers I work with frequently carry prototypes back & forth between Japan or Taiwan & California.
These are digital cameras with missing or wrong manufacturer markings on them, often with bad fit between the plastic shell and the switches (which are soldered to PCBs inside) and usually a serial port dongle (so we can develop software for them) hanging by its wires which enter the body of the camera through a rough-hewn hole. They often don't work properly or boot with a lot of nasty beeps as they enter the calibration code (thus they can't be demonstrated as cameras to screeners).
They're often more delicate and certainly more valuable than a final-production camera-- there are less than 20 in the world and there are blue wires on the PCBAs. Therefore, they're usually in the engineer's carry on.
They look about as much like a small poorly-disguised bomb as you might want, but no-one's been hassled yet.
Most /. readers are NOT the richest 25 percent of the world, at least I'm not...
Really? R U sure you mean the world and not the western world or first world or something?
More than 1/3 of the world's population lives in China and India. 5 of the 6 billion people on Earth live in "less developed" countries.
The poorest 10% of Americans are still better off than two-thirds of the world population.
Nintendo claims they never take a loss on hardware sales (though that's hard to believe considering the Gamecube only costs $199).
Tricky accounting might let Nintendo say this without technically lying even if the real cost of all the parts + putting 'em together exceed the wholesale cost of a Gamecube.
Suppose IBM, MoSys and NEC want Nintendo to be successful and they're also on the weak side in the negotiation (don't know if that was the case, but read on). Nintendo negotiates incredibly cheap prices for the ASICS so the Gamecube really is cheap (to Nintendo) to build.
The chip guys aren't making much or are maybe even losing some money, but the contract guarantees a certain number of chips per quarter (at a particular price) will be bought over the next few years & they believe their process guys can get the cost down & they'll eventually make money.
It's the same idea as sell below cost now, but reduce the price slower than the cost falls. You're just moving the loss (and the potential profit gain from cost reductions) outside the primary manufacturer.
Of course, what really happens is contracts always get renegotiated so if Nintendo doesn't sell enough, they buy fewer chips & the penalty is softer than originally agreed. If the chip guys drive their cost way down & Nintendo gets wind of it, they renegotiate the price down.
Oh, and the engineers who enabled the cost reductions get nothing but complaints about how they didn't do enough or how fallout went up because 20 cents was saved on a crappier regulator.
Moderation: +1 bitter
Citation?
I just don't believe this. Copyright owners have sued novel devices claiming contributory infringement, but they usually lose (RIAA v. Diamond, Universal v. Sony). Photocopiers still exist without a "paper tax" paid to book publishers.
Kodak does support watermark detection and won't copy prints made on Portra III paper, but they've discontinued the paper because "the reality is that there are many more non-Kodak methods for making high quality copies that are readily available to consumers today."
http://www.hjpro.com/news_copyright.cfm
I still see Kodak Picture Maker kiosks at my local stores....
how far will MS take their 'hardware lust'? how close to being Apple will they get before the afore mentioned giants turn on them?
Whoa!
Good fscking point!
However, an Xbox2++ might eat away at the home market for PCs which doesn't seem that profitable anyway. Microsoft doesn't seem interested in going after the higher margin powerdesktop/workstation/server market. Might Dell & Co. just let 'em have a bit of home & concentrate on business?
Grabbing the mpeg-2 stream from a satellite box is a great idea. Something like 5C encryption on the stream piped over USB would keep the MPAA happier.
The TiVo uses a hardware real-time encoder because the main CPU is wimpy. A Pentium 4 @ 2.2GHz is probably getting close to doing MPEG-2 @ 720x480x60 in real time with some quality. Creative's requirements (Video Blaster Movie Maker) for 352x240 MPEG-1 or 2 (at probably reduced quality as well as resolution) is a 350MHz PIII.
Instead of compromising on quality, though, you could add an external MPEG encoder, connected via USB like the ADS Tech. Instant DVD (under $200). That's expensive now (compared with $300 for the Xbox), but give it time.
{Yes, full-speed USB is capable of carrying video. 12 megabits/sec > 5-6 mbps of DVD. USB won't carry lightly compressed video like DV (3.6 megaBYTES/sec). For that you do need firewire or high-speed USB.}
The Xbox's hard drives are little small for PVR, though. An external USB connected drive with encryption would be a good idea.
If we're putting the hard drive & possibly video encoder in an external box, the need for the xbox is diminished, though. If we're doing analog PVR from any source, the xbox can provide UI & control. If we're doing PVR from satellite, though, might as well run the UI on the satellite box & make the add-on just a USB hard drive with CPRM.
Suppose your an IT manager in a company that uses non-libre software.
You want to keep on the up & up with licenses 'cause you a) think it's wrong to copy w/o 'em b) don't want the BSA hassle and/or c) want to keep track of what machines run what software so you can keep 'em from breaking.
So you do your own internal audits. When the BSA sends their threats, you say "we've audited & we're compliant." When they tell you to turn over the results of the audit you tell them politely to fuck off & stop wasting their time. Maybe mentioning the method by which you audit calms 'em down a bit.
I suppose this will piss them off even more than if you hadn't done an audit, but their attitude that only they know how to count might annoy a judge enough do deny an order to hand over your audit or let the BSA do one of their own.
-M
Theater companies have to pay for performance licenses. This is true even for non-profit companies & even in educational settings. High schools have been prosecuted for acting without a license.
Bands playing covers are supposed to pay royalties to the songwriter (usually through the Harry Fox Agency).
The only way to avoid this mess is to see music & theater that's 1) written by the performers 2) written by people that died a LONG time ago 3) written by commission from the performers (San Francisco Mime Troupe).
Even public domain plays and scores can be a problem because a derivative work of a public domain work can be copyrighted (e.g. a modern arrangement of a jazz standard based on a Tin Pan Alley Tune (which might be barely old enough to be p.d. in the first place).
Here's 1201a2
This, to me, looks kind of like the four tests of fair use. Unfortunately, it's not the intention of whether the device was intended for infringing use, but whether it was intended for use in circumvention (even if otherwise legal) at all that's tested.
This looks like a pretty awful contradiction. 1201c1 says "(c) Other Rights, Etc., Not Affected. - (1) Nothing in this section shall affect rights, remedies, limitations, or defenses to copyright infringement, including fair use, under this title." so you're supposedly still entitled to fair use, but there's no way to exercise that right.
{mod parent up!}
What makes you think that this functionality cannot be integrated onto the same chip that does the normal digital-to-analog conversion?
Yep- this is the key technology making encrypted content MUCH harder to access in "pristine" form. For example, Broadcom has a digital cable "settop on a chip" that has the CPU, MPEG decoder, decryption hardware (DES3 in realtime needs hardware 'cause CPU is too wimpy) and video DAC on one chip.
Soundcards with signed drivers are another step in this direction.
One weakness is that keys must still be passed around. Some of the schemes for protecting session keys are quite good and the key decrypters are also in hardware, though sometimes not on the same chip (though they use encrypted comm. btwn chips).
Simple software-only cracks will be a thing of the past once the encryption moves out of software and into the video and soundcards, with DVD-ROM drives and network interfaces participating in secure channels for keystreams.
Be, Inc.'s auction next week includes a few logic analyzers & high speed o'scopes. Get 'em now before they're declared "hackers tools."
-M
Stupidity + Competition = Disaster
Stupidity + Monopoly = Success beyond your wildest dreams.
A few short years ago when IBM wasn't looking so hot (Taligent/Kaleida) and the Baby Bells were going great gangbusters (partially due to the Telecommunications Act), pundits commenting on the Microsoft case thought a breakup might actually be good for MS.
The thought was that IBM, which didn't get broken up, had to be vewwwy careful to avoid the appearance of abusing monopoly power, giving rise to DEC and eventually Compaq/Dell/Microsoft. If their lawyers hadn't held them back from using more than FUD against the mini, we might still be using mainframes.
The baby bells and AT&T were allowed to do business with fewer restriction. There's regular FCC tariff crap, but the Telecomm Act helped ease that. For a while, AT&T and some of the babies were doing quite well. In some respects, Verizon & esp. SBC might still be said to be doing well. Their customer service may suck worse than ever, but are they making money?
....the Sound Blaster 16. 44.1kHz 16-bit stereo is CD quality...
Someone else has already pointed this out about the SB16, but I'll do it again a different way.
You are confusing word width measurements in different parts of the audio chain. I agree that 16 bits is enough for most music in most listening environments as a distribution format.
Let's divide the audio processing chain into a few parts. How 'bout
1) acquisition
2) tracking
3) processing
4) mixing
5) mastering
6) distribution
7) reproduction
assuming an all-digital chain from the ADC at the studio to the DAC at the listener.
You want more than 16 bits in acquisition because you don't want to have to either apply heavy compression in the analog domain, ride the gain like crazy (manual compression) or lose the quiet passages in the noise.
After acquisition, you could do some AGC/compression before storing the samples in a shorter word, but storage is cheap.
When you do processing or mixing, you end up adding samples from multiple source (multiple tracks or the outputs of multiple delay lines for things like reverb). This increases the word size ( 0x7fff + 0x7fff = 0xfffe which doesn't fit in a 16-bit signed word). You can round towards zero (please don't truncate) to get rid of the new LSB, but if you keep the wider width until the bitter end you'll introduce fewer artifacts (remember, we might be talking about a 128 channel digital console!).
You can master to 16 bits if your output is going to be CD, but you might want to do the compression/equalization stuff as one step to a 24 bit master then do something like Sony's Super Bit Mapping to get 16 bits. You also might be mastering for release in both DVD-A/SACD and CD.
Distribution in 16 bits is cool, but the only way to make sure all 16 bits are actually information and not noise is to handle the above steps properly (see the RHCP Californiacation as a counter-example).
Reproduction through a perfect 16 bit DAC would be fine, but most 16 bit linear DACs only gave you (when people still used 'em) 14 to 15 1/2 bits of information. Oversampling filters increases word width and there is real benefit to converting those bits which is why you'll see things like 8x 18bit outputs (on 10 year old audiophile CD players). Today, everything seems to use delta-sigma DACs with very high sample rates and effective word widths of well over 16 bits.
The last thing on the soundcard is the analog path from the DAC to the jack. The SB16 puts the DAC output through an incredibly crappy digitally controlled analog mixer along with the CD, synth and other sources.
That's why you don't get 96dB of SQNR out of an ostensibly 16 bit soundcard.
---
Soundcards aren't only for reproduction. They are production tools also and the Audigy with a 1394 interface and front-mount I/O certainly fits the bill for amateur production. There is a spectrum of artist abilities and budgets so it's unreasonable to expect everyone who's authoring to have high-dollar equipment.
---
What's a "standard PCI Sound Blaster 16" by the way? The "Live" was the first PCI soundcard from Creative (not counting rebadged Ensoniq cards) and if there's some sound improvement to go with the bus improvement, I'll take it.
-M
Mary jane's too expensive in my neck of the woods to stick to the imperial system. At over $10/gram, sometimes even 1/4 oz. is too much.
I know old fogeys who talk about four finger lids (1 oz.) and dime bags: about 1/4 oz for $10 (too cheap to use a precise scale).
It's odd that the software I trust the least to protect my assets (my work on the hard drive that I'd prefer a worm not eat) is trusted the most by the OS to protect someone else's assets.
somebody, somewhere, is going to find it.
Consider digital cable set-top boxes (Scientific-Atlanta Explorers and Motorola DCTs). Security still unbroken.
Private key is inside a secure micro that is never supposed to give it up. It only decrypts session keys (which change every few seconds so can't effectively be shared).
It's true that at this point the unencrypted content is in the clear on a couple of memory busses (possibly mixed with a lot of other traffic), but the content tends to be ABC, HBO and pay-per-view movies. In other words, not that valuable. The important thing is securing the access for folks that don't pay not the content for those who do.
In any case, set-tops-on-a-chip like Broadcom's mean that the CPU and MPEG decoder are on the same die. Moving the DENC (digital to analog converter for video) onto the MPEG decoder eliminated one exposed path & this will eliminate some others. They'll have to get embedded DRAM on the chip to really make it secure though. Then it'll be output of the QAM decoder in, analog video out.
These techniques can be used for PCs too. Soundcards are just the beginning-- look for secure CPUs, hard disks and video cards in the future.
-M
This stepping-stone to a secure operating system is sometimes called "Secure Boot."
It's also called the xbox.
That's why there are folks attacking the BIOS of the xbox (search your favorite p2p network for xbox.bin), trying to figure out how to boot unsigned software, over at xboxhacker
But dvbackup barely works. It doesn't even control the transport directly, which every $100 DV video editing package is capable of.
I think this would be great as the tapes are cheap and available at grocery stores (I'm thinking Digital8 which can use 8mm or Hi8 tapes and is miniDV-compatible). The camcorders start around $700 which isn't that cheap, but not bad for a backup drive you can take with you & make kid videos.
It's funny, though, that the reason I peeked at this article in the first place is that I have many gigs of kid video for which the only backup is the VHS (bad!) tapes I've sent my parents.
The "dailies" and a copy of the edited-down project on 8mm will survive a disk crash, but I'd have to re-build the whole project just to do a small tweak (not a big deal for kid videos, but you might feel differently about your Episode II spoof) and I worry about fire, earthquake & theft.
If I had another camcorder, I could at least back up the dailies.
Besides digital video & photography, all my work would probably fit on a single CD.
It's a strange twist that commercial video houses which might survive without every scrap of video they've ever worked with can afford backup easier than parents who don't want to lose one frame.
-M
A generic (Honda, or something) gasoline generator is only a hundred bucks or so
Honda doesn't make generic generators, Coleman does. Even those aren't $100. I just bought a 5kw Coleman @ Costco for $440. It's the cheaper series with the Tecumseh side-valve 4 cycle engine (instead of the overhead valve Briggs & Stratton or an overhead cam Honda).
It is LOUD. I'd read about this before I bought it, but I had no idea until I started it up. I went running for the hearing protection (muffs- I hate plugs) right away. I'd planned to break it in for 1/2 hour, but I
A good Honda generator in the 5kw class costs more like $1900 (Northern Tool catalog). You can get Honda engines in gensets by other mfg.'s for less but Honda makes several classes of engine.
It's not just the noise (which is why I delved into this article in the first place- I look forward to the day when quiet generation is cheaper)- the cheap gens eat themselves up in no time- something like 100 hours for mine (it does have the cast iron cylinder liner for maybe 2x the life).
Smaller gens are cheaper- a 50cc, 1kw, 57dB @ 7.7 yards Honda is $690 and you can run a computer off it, but not the blower motor in your furnace, your refrigerator or your septic pump (my house's big draws), so put on a sweater & don't bother with that bathroom break & forgeddabout that steak.
-M
Um, what technology does your sub-$4000 RPTV use? Video projectors in general use LCD, DLP (Texas Instruments' micro-mirror tech) or a set of 3 cathode ray tubes (one each for red, green and blue). RPTVs tend to use CRTs.
No shadow mask removes some problems with direct-view CRTs, but introduces other ones.
Direct-view tubes have a very wide field of view, matched only by expensive panels.
-M
One thing that came out of this is that the DOJ claimed that the *intent* of creating a circumvention device matters. If it's used only to demonstrate a weakness, it's kosher. We'll see whether they hold to their own opinion when they're on the other side of the aisle.
In their own words:
"Plaintiffs' alleged conduct is not proscribed by the statute"
"While Plaintiffs' computer programs have the additional capability of actually circumventing access controls, they are allegedly not designed or marketed for the purposes of actually getting access to the copyrighted material itself."
The DOJ asserts that Dmitry fails the "is primarily designed or produced for the purpose of
circumventing protection afforded by a technological measure" test of 1201(b) while ignoring 1201(c):
(c) Other Rights, Etc., Not Affected. - (1) Nothing in this section shall affect rights, remedies, limitations, or defenses to copyright infringement, including fair use, under this title.
(2) Nothing in this section shall enlarge or diminish vicarious or contributory liability for copyright infringement in connection with any technology, product, service, device, component, or part thereof.
(3) Nothing in this section shall require that the design of, or design and selection of parts and components for, a consumer electronics, telecommunications, or computing product provide for a response to any particular technological measure, so long as such part or component, or the product in which such part or component is integrated, does not otherwise fall within the prohibitions of subsection (a)(2) or (b)(1).
(4) Nothing in this section shall enlarge or diminish any rights of free speech or the press for activities using consumer electronics, telecommunications, or computing products.
By the way, USB-connected cable modems (if they follow the trend of DSL modems) are going to make NAT more of a pain in the ass. They can play all sorts of nasty games in the class driver on the host that will be hard and/or illegal to replicate on Linux or BSD.