A Stuckist Net – you want in
Letters Register readers are dreamers - but not schemers.
That's the conclusion from your response to our story, The Stuckist Net - what is your post-Palladium future? - asking how a non-Palladium world might be sustained if the biggest names in the computer industry continue to capitulate to . We suggested that in India and China are two countries willing to capitalize on the PC industry's reluctance to face down the US entertainment lobby.
Both India and China have indigenous populations, thriving local film industries, who would not feel Hollywood wrath as acutely. They'd also welcome the chance to lead a technology market, rather than become assemblers or service suppliers.
"Maybe in a century or so, they'll ship charity-technology to the underdeveloped West", muses one correspondent.
Lots of you want in, but far less of you responded to the challenge of describing the bits and pieces. What machines would we run? Who could control the pipes? For that matter, what pipes?
A few of you put your thinking caps on.
Is it even feasible? Practical Stuckism
1. acquire inexpensive, used equipment, primarily laptops, for parts...
2. Learn how to build ruggedized equipment by studying the innards of damaged ruggedized laptops (e.g. how to shock-mount hard drives so they don't crash when a laptop is dropped, how to shock-mount LCDs, how to waterproof a case, and etc). Design and construct a basic, easily customizable rugged frame on which any acquired parts can be retrofitted. In my case, I'm building molds with which I can cast as many parts as I need (I was into special effects when I was a kid, so I have a lot of fabrication knowledge, from clay sculpting to plaster casting, to plastic injection molding using household items -- no joke).
3. Start ripping apart the used laptops, putting their parts in cheaply molded, ruggedized chassis' and build a set of computers that'll last twenty or thirty years and be able to be repaired easily as necessary.
Basically, the goal is an eternal, nearly unbreakable computer.
This is kind of a paranoid approach, but if you don't want to have Palladium shoved down your throat in a few years, you've only got around two years left to scrounge parts. I'm going to have a stack of nearly indestructible laptops running Linux hidden away in MY closet. And, I'll probably have to buy a Palladium computer at some point, but at least I'll have the real thing tucked away somewhere safe. In fact, I should have at least a dozen of them altogether, plus all the molds and measurements.
OK, what about the processors?
They could clone x86, or license a non-Intel instruction set such as ARM cheaply, or SPARC for no cost at all. That's the easy part. Manufacturing requires huge capital investment
Not entirely true. There exist many large programmable devices (FPGAs and CPLDs) which are very capable of being programmed to act like microprocessors and microcontrollers, and the programmers for these chips can be as simple as a US$30 kit or a US$100 cable. Some require a bit more sophisticated devices, but development kits are almost universally available for less than US$1000, software and all. In the case of Xilinx FPGAs (and possibly others) the software's free.
Granted they won't be quite as fast as a dedicated mask processor, and there will be some waste (unused gates, extra circuitry) but it is certainly not the prohibitive entry cost of an entire chip fab plant. It's also worth noting that masked ASICs can be had for only US$30,000 startup costs and maybe US$5 per chip. (Large FPGAs go for maybe US$50 per chip) (but then you have to find an ASIC maker (most are in Taiwan) who'll keep mum about what's in the chip. They all do - Not to do so would be business suicide).
Here's where it becomes silly. Since they're 'digital devices' they must have copy protection built in, even into the FPGA (if certain laws are passed). So there'll be an area of the FPGA that you can't program anymore - But how is it ever going to be accessed? If you don't program any connections between such circuitry there won't be any connections. The pins can be left unconnected on the board. It can't enforce anything at the bit level - Unless a 1 or a 0 becomes copyrighted. If the incoming data is analyzed for copyright codes and checksums, scramble the bus, and/or set it to a different width. Ignore the bus pins entirely, and wire up the data to uncommited I/O pins. Use funky data alignment. Rewrite your program counter to use Gray Code. Everything's programmable, and will look like unmitigated garbage to a DRM circuit. And unmitigated garbage has to be functional, because how else are you going to process any generic mathematical function on any incoming data? The function c = a + b, if it is to be mathematically useful, must process any data for a, b, and c. It's going to look like garbage.
As an aside, Lattice Semiconductor still makes programmable logic chips with only eight single-bit registers, and legally, they'd have to find a way to incorporate a DRM scheme into one byte of data AND program combined, and still leave some space for user-programmability? That would be fascinating to see. I'll bet they wouldn't write it in C#. And if there isn't any in these little chips, then several of them can be ganged together to make a half-decent 68000 clone. Faster, and Fritzfree...
Circuit boards to mount these chips can be had for setup costs of ~$500 and $40 each from any of a large number of small vendors. One could, theoretically, build one's own Fritz-free CPU board for not too much money. Remember those '486 Accelerator' boards that used to plug into 386 motherboards? Same idea. No, they're not going to run at 2.53GHz, they'd be lucky to get 100MHz, but that's fast enough for Linux. Remember the HEAP gun from Neal Stephenson's Cryptonomicon? It would not be that difficult to make a HEAP computer.
I have made a business out of making custom-built ISA cards for specialized industrial applications, and I'll be damned if I want anything to do with a Fritz/Palladium/whatever 'Hardware Certification Scheme'. It would cost me a fortune to have my cards certified even though they have nothing at all to do with copyrighted material!
Larry Kollar has some interesting thoughs about free pipes:-
Well, since I've been thinking along the same lines for a while, I have to agree with you (except that I've been calling it GeekNet).
One minor disagreement -- I don't think hardware is much of a problem. Today's kit is fast enough for most purposes, especially when coupled with an efficient OS and a lightweight window manager. By the time all the new hardware goes on the Fritz, non-chipped hardware should be even more powerful. As the not-so-techie types "up"grade, I expect the "old, useless" kit to quickly become free for the taking.
Communications may sound like an obstacle, but I'm old enough to remember when 2400 baud dialup modems moved a *lot* of mail and news around the world. The Stuckists could get started right now: wireless LANs for local links, V.90 modems and dirt-cheap long distance (at least in the US) or slower packet radio (ham) links for long-haul. In that scenario, I would expect UUCP to make a comeback for transporting mail and news over dialups; the packet radio links would be suitable for interactive, low-bandwidth connections like chat or telnet. A few strategic encrypted tunnels through the mainstream PigopolistNet might be possible, but I wouldn't want to rely on them.
The result is certainly no Utopia. You would have islands of local connectivity, and selfish goobers could easily clog the long-haul pipes with large files (MP3s and the like) unless a local BOFH eradicates such folks quickly and mercilessly. But that same limitation would also work to the Stuckists' advantage: given its total unsuitability for media piracy, the government's corporate masters would see no reason to outlaw it. Meanwhile, geeks (who generally love a technical challenge) would work quietly to extend the effective bandwidth of those slow links. Who knows what kind of breakthroughs in compression, DSP, or encoding technologies could result in such an environment?
Maybe it's time that geeks took their ball and found a new park. Count me in.
But for a Stuckist net and Stuckist PCs, the best option is going to be ARM - Linux already runs on it, and it's surely not beyond the wit of man to come up with a big version of the Sharp handheld, suitable for desk work - or even to hack one to get decent external graphics and USB disks and keyboards usable.
The internet thus far has been based on all computers voluntarily following the same protocols in the same ways. As long as there aren't compelling reasons to break this universal consensus, it certainly simplifies things. But if the media companies get their way and create these compelling reasons, they will find the consensus to be a fragile one.
Once the decision has been made to leave the standard internet to those who would strangle it, adding one more layer of abstraction and creating a universe of internets won't be that difficult. It won't require separate wiring or even the assent of those who own the wires: the IP doesn't care what is above it in the protocol stack. That users are free to pick and choose from numerous mappings from URLs to IP addresses will quickly become the new consensus, and those who would regulate how two computers they don't own should communicate will be back to square one.
Firewall of China
I see the "great firewall of China" being applied to every country along geographical borders... There WILL be a black market in general purpose computers, mark my words. (start stockpiling your old hardware now). Every time entrenched interests outlaw tools that increase personal freedom a black market develops. Firearms. Narcotics. Now computers.
The USA in league with China...both veering toward fascism from different directions...gentleman's agreement so those in power stay in power (we'll pretend Tianeman sq. never happened if you pay us to build your great wall)...just wait until the hackers realize that makes the bedfellows with the hairy armpit anti-G8 protestors...
We need an active campaign on the lines of "We're NOT buying CDs from the pigopolists. We're NOT going to the movies", writes David Cefai along with many others.
There needs to be a concerted effort to stop these laws and by concerted, I dont mean an internet boycott or similar. It has to be made public. If there is a public boycott world wide, everyone saying that for 1 week there will be no purchases of software, DVD's, CD's etc. When these companies lose a couple of billion dollars because people do the only thing that they can legally do, which is decide to NOT buy a product, then we might see these laws and these moves stopped.
I think the Western companies/governments continued failure to adopt Linux and Open Source solutions could very well be their undoing.
While you see countries like china eagerly adopting this new model, and improving it, most European and American companies pretend it doesn't exist. If this trend continues, the balance of power will shift radically. Students in 'progressive' countries will have access to specifications of protocols, programs and hardware. While their Western colleagues are stuck with a box they can't open, with no technical documentation. Messing with it is a greater offense than murder.
This will create a generation here that is technologically illiterate. Sure they can play 'snowwhite', but compiling a program, improving a protocol? These kids will never have 'played' with computers like many of us did. On the other hand, countries that turned away from closed systems will have a breed of techies that grew up with systems that came with full source-code and RFC's. Ready to help their less-proficient fellow-humans. Maybe in a century or so, they'll ship charity-technology to the underdeveloped West...
My gut-feeling is that we're moving to an all-out war between the Open-communtity and the establishment. I think we'll either see the demise of Microsoft and the like (which I'm hoping), or the demise of open systems in Europe and America. The fight could last for another couple of years or so, with some remaining splinter-groups afterwards, but I think it will shift one way or the other. Quite possibly by shifting the balance of political power in the world too.
Guy Van Sanden
Where do I sign up? Seriously, this is a big deal. We're heading fast towards a totally regulated environment in which casual website use is tracked and every email sent/received is monitored, in which the efficiencies of the internet are lost (no p2p applications, only centralised hierarchical systems that log every activity), and in which one has no privacy even in one's own home. A world in which you can't buy general purpose devices, only TIVOs and PlayStations. It's amazing how blase folks are about this: folks are unwilling to articulate the arguments against something they are see as inevitable. Either that or they are in denial - they can't believe it will happen. But it is looking more & more likely as time goes by. Speaking as an info-sec professional: the governmental argument (that the measures proposed will defeat terrorism & crime) is specious. One can't help but suspect the agenda is both simpler and more sinister: that government fundamentally does not trust the populace, and it is determined to monitor/control what the populace sees & does. The film & record industries are of course entirely happy to be part of that as they continue to push the argument that the internet is eating into their revenues(*). It is horrifying to see all this coming to pass. I am not a "political" person but this is driving me to become one.
Paul G Smith
I think that Sony et al will get what they want. The US will lose its edge, there was a .dot com that just announced they are moving their coding to Bangalore for the lower cost. WSJ 08-19-2002. The anti crypto laws will reduce the flow of knowledge anyway, to the NSA even. But then what do you expect from doddering old fools like Hollings
It was a rude awakening for me when you mentioned the threat of of the Internet becoming a one way medium. I realize in many ways it is because of the combination of limited IP address space and OSes that are totally insecure by default. I have already helped to many friends with "securing" ADSL connections.
Currently the safest way to connect a windows machine to the net includes either a NAT or a firewall, making the machine unable to work as a server dishing out information, instead it passively sucks whatever it can of the net.
RIAA might just as well lobby for laws that requires Internet connected computers to be NATed and firewalled, and P2P will effectively be killed
dead. Outlaw the odd socks proxy as a circumvention device, finishing touch, and the net will lay at the feet of giant media.
To counter this we need more IP addresses, IPv6 anyone ?
Frank A. Stevenson
Leave us alone - all we want is pizza
In a brilliant letter, Brian Hurt writes:-
Paraphrasing Sun-Tzu, don't ever fight the enemy on ground of his choosing. Fight the enemy on ground of your choosing. So long as there exists an option, even if it is that old, obsolete 2.4GHz P4 whose only benefit is that it isn't Palladium-crippled, then eventually freedom will win out.
94% of House races and 85% of Senate races are won by the cannidate with the most money
Here's another statistic for you: The top 1% wealthiest Americans own well over 50% of the nations wealth.
The best we can do is drive the price to buy a politician up. Temporarily. We can't outbid them, they have money than the rest of us put together.
In any case, even if you disbelieve this highly pessimistic opinion of the US political scene, one other thing to remember is that Microsoft, Intel, and Disney are overtly, openly, plutocracies. The only votes you get are the ones you explicitly buy- hint, they're called *shares*. And you can buy as many votes as you like, all you can afford. They don't need Holling's bill (as anything other than an excuse, and simply *proposing* it is enough of an excuse) to impose to impose an Orwellian (in the strictest sense of the world) regime on us.
Simply defeating the Hollings Bill isn't sufficient. We have to actively outlaw what they're doing. We have to pass an anti-Hollings bill. In this era when an American's right to representation and a fair trial seems to have been repealed (search under "Unlawful Combatant") our chance of getting such a bill passed makes that hell-bound snowball look like a sure thing.
Leads into libertarian wet-dream.
The problem with the 'Candide-like' geek population (/.ers, war chalkers, whatever) is that many of them cannot believe that such an apocalyptic situation as Palladium (i.e. various major labels/hardware and software co.s) owning mainstream PCs utterly, and the complete demolition of fair-use rights.
One of the key symptoms of the problem you describe is that (as usual in any impending spod-related disaster) one can divide the vocal geek populace into three groups - those whom are gloom mongering, those whom are being cynical about the situation, and the those whom are being cynical about the gloom mongers.
The fact is that most people whom are up on current geek affairs (Reg readers, aforementioned /.ers, etc) already have at least a good clue as to what is going on. There's no point poking at those palid lumps of lard. The real mission should be breaking out the message outside of the geek enclave and out into the mainstream, whilst avoiding FUDish scaremongering.
Christophe Dupont, UK
Well here's the Candide case put succinctly
1) My feeling on the Pigopolists' intention to restrict the Internet by controlling protocols is that it just plain won't work. I have sufficient faith in rebel hackers that they will always find a way of defeating blocks based in firewalls & routers. In any case there are _indirect_ ways of (ab)using IP protocols to carry data outside of the official data content part of the packet. These will be exploited...
2) It's clear that what the Pigopolists want is for the Internet to be turned into yet another home shopping channel. I think it is unclear that there is a need for this channel; in any case, IP is a very messy way of providing it. Technically the best way to tackle this argument is to provide a set of protocols more suited to "entertainment" and to allow these to be heavily filtered/monitored/Fritzed or whatever. Though I think there should still be plenty of mileage in use of the courts to preserve traditional consumer freedoms.
3) I fail entirely to see how the Pigopolists are going to prevent home-based wireless networks merging into area & regional networks, when _consumer_ wireless equipment becomes common enough & people start using their computer systems as routers in order to enable them to interconnect with each other. Even if there are means to keep these nets from being connected to the Internet trunks, WAN connections could be provided by ham radio - though personal experience with University networks, and the corporate demand for home access to "secure" corporate networks suggests that interconnection at many, many points is inevitable.
This idea goes right back to the original spirit of the Internet ... a network entirely owned by and operated by its users ... BT, ATT etc. will no doubt whinge, obfuscate and try to obstruct the development (because there is no place for them in this structure) but I think it is inevitable. The recent development of "warchalking" is just the first primitive step in this direction ... in the long run, freeloading onto commercial wireless nets won't work because effective access controls will be deployed, but freeloading onto a ubiquitous uncontrolled network owned and operated by its users is unstoppable.
The phrase "stuckist net" tends to give the impression that such a network would be in some ways frozen in time - in fact the Pigopolists' network would be the one with frozen protocols; the phrase "stuck pig" (as in "bleeding like a") springs to mind. The correct phrase for the new network would be "the open internet", or simply "freenet", assuming that trade names can be reclaimed.
4) Those of us who care enough are already stockpiling "open" hardware. Personally I probably have enough to last out my natural life already.
To which I responded-
I think you're in denial :-)
1. It doesn't matter what the "rebel hackers" produce if it can't run on the hardware that 99 pc of the population use. Palladium will mark it as unsafe. Is that a victory?
2. You can only warchalk as far as your pipe allows you. If it blocks content, you're fucked.
I appreciate your romantic view of things, but can you address the issues in the article? It discusses new hardware to run this on, etc.
The "rebel hackers" don't appreciate the magnitude of the problem yet - working on the assumption that PCs will remain open, and routers will remain neutral.
It's worse, one academic points out:-
With all the focus in The Register as of late on the effects of Palladium and TCPA on the business and communications sectors of the computing community, perhaps you neglect the ill affects on the more meager sectors of the computing community which will undoubtedly loose big. I personally speak of the high-performance scientific computing community, which relies heavily on both the ability to modify at will our resources without reliance on imposed certification and the freedom to communicate software and datasets with worldwide collaborators. Scientific members of the Stuckist Net, as you see it, will be effectively walled off from a variety of interactions that both those within and outside of the Stuckist Net heavily rely on for scientific discourse and innovative research.
First off, members of large institutions like universities will most likely be forced to comply with the computing standards adopted by their host institution. Noting the aversion to litigation American universities rightfully suffer from (not to mention bad memories from a long history of being an easy target for crackers), it will be likely that many such institutions will require Palladium or other TCPA certification of its computing resources. At this point, one might envision an administrative nightmare for the scientific computationalist who is forced to conform to a TCPA. Will every memory upgrade, motherboard replacement, or RAID card crash cost me countless hours of recompiling home-brewed scientific software, or worse, terabytes of binary data that no longer can be used since my hardware configuration has changed and the embedded ceritifcation is no longer good?
Never mind the cost to scientists who rely on custom hardware to certify their IC boards with Palladium or other TCPA standards (this would make many current scientific efforts simply not cost effective).
Second, assume I successfully fight off administration attempts require TCPA compliance and avoid surrendering precious CPU and memory to an OS whose overly bloated, non-modifiable kernel sucks some of the life out of my ~200 compute nodes. As a member of the Stuckist Net there will be no easy (possible?) way for me to share binary data from my modded machine with TCPA'ed collaborators, submit scientific documents to TCPA'd online journals, or apply for grants through TCPA'd government websites. I will loose access to many commercial software apps (IDL, high performace compilers, etc.) or even many stand alone hardware products (nice switches, oscilioscopes, etc.) which may require TCPA of the computers they interface with. At this point I will certainly be stuck, forced to choose between the cost of high quality, non-TCPA research (that I have control over but that will be difficult to bring out from the basement) or overly expensive, worse performing, TCPA compliant research (that I have to recertify at an unknown cost four times a month). Either way, I don't look forward to it.
Dept. of Astronomy,
Radio hams isn't a great metaphor to choose, point out several readers:-
Ham didn't really die out (not that it's dead by any stretch of the imagination) because of regulations, etc. There are just far simpler and less expensive ways of communicating. The die hard ham users are still there, and I would bet that they are quite happy that not just anybody can or will just jump on a radio and start broadcasting whatever idiotic rubbish comes to mind.
Which is true. The analogy is that with the circumvention tools underground, and requiring expertise beyond what the average Joe is prepared to learn, it could resemble amateur radio.
After all, all PCs will be safe appliances. Who'd want to mess around with wires and aerials?
Bill Softky is brings us back to earth:-
I'm really not sure how much TCPA will be capable of really doing what the bad guys want to do.
If TCPA restricts how I can save and open Word documents, can't I just cut-and-paste the content into an email to leak to a respectable (cough, cough) journalist? If TCPA restricts copying MP3s or CDs, can't I just put the microphone of my home CD-burner in front of the loudspeaker? If TCPA restricts DVD playing, can't I just aim my home video camera at the monitor during playback?
Only if TCPA finds a way of blocking content by its true *perceptual* signature (independent of font, sampling rate, screen resolution etc) will I worry about it as a threat rather than an annoyance.
Thanks for your letters and the huge response. I'm still reading them. Don't miss Barlow vs Bill this evening, via BBC2 or Real Player.
Sponsored: Today’s most dangerous security threats