Slashdot Mirror

Yeah, well I'm headed for Denver with a digital recorder, a well disguised camera and a stereo pair of shotgun mics.
Feel free to head over to http://www.aeg.r.clueless.com/ and put in a PRE-Order for your bootleg box set including comedy outtakes from
AEGs last board and stockholders meeting. The first 100 sets include a free T-Shirt with every band logo on it and locations to buy tickets once they are sold out.

how will you *know* that the source they claim went into the product is unmodified?

The idea is that I can examine the product's source code, recompile it using my own tools, and then use my own recompiled version. Yes, it's possible that the compiler I choose may be specifically designed to recognize that source code, and then inject a trojan into the binary that wasn't in the source - but of course, I've examined the source code of my compiler and built that from source, too!

Now the compiler that I've used to compile my compiler could be designed to recognize its source, and inject the code into the binary that will recognize the product's source code, and inject a trojan into the product's binary. It's been done before (http://www.clueless.com/jargon3.0.0/back_door.html).

But if the compiler and product come from independent entities, the odds are much greater that I'm using what I think I'm using than if I just blindly trust the product's manufacturer to provide an unmodified binary.

Clear?

The Wheel of time has turned again. GPUs are now general-purpose massively-parallel computers; they will be folded back into the CPU core, so that the general purpose CPU gains massive parallelism. Kind of like SIMD, but on the order of a million operations per instruction instead of 8.

The next 10 years will consist of a new type of external graphics hardware being built, which will of course, be folded into the CPU at the end of it.

In a similar vein, maybe a divinely inspired programmer can spec out the perfect message passaging schema from God, but until mere mortal CS PhD's actually do it, the rest of us are doubting Thomases.

At the risk of overworking my Christian metaphors, perhaps microkernels are an example of creationism. How do you spec out the required message passing until you have experience with the implentation? How do you revise the message passing to fix problems without breaking the implementation?

ship a slip of paper with the iPod with instructions to break it open to remove the battery.

Optionally, you can write in for two extra sets of arms.

I don't really care if the original poster was going for funny. There's a little TLA for this situation, usually reserved for fr1st psot lusers: YFI.

Besides the whole thing reeks of "haha only serious." A real case of Napoleon Chihuahua going on here.

But the mailing list server would have to take on additional load since they send mail to so many users.

No they wouldn't. Client-side whitelists would be an unavoidable part of any hashcash rollout. After all, flag days are untenable- there WILL be a transitional period when not all email software supports the protocol, and users will need a way to exchange messages during that time. The simplest workaround is to allow users to input a list of addresses they're willing to recieve from without an attached hash.

Given that such a system will exist, mailing list will certainly take advantage of it. When you sign up for a ML, it will instruct you to add the ML to your local whitelist. The only time the servers software will send hashed messages is when a previous non-hashed message has been rejected- and in that case, the body of the 2nd message will just be a repeated instruction to whitelist the ML, or your subscription will be cancelled.

It has nothing to do with the band. Phishing is derived from phreaking and fishing.

Everyone should already know this, so please don't mod me up.

GDS causes problems on public terminals? Easy solution: DDTT!

Here's a citation for that quote. The Internet. On the edge of collapse since 1981.

is indistinguishable from a rigged demo. Apologies to Asimov.

Er, don't you mean Clarke , as the AC also pointed out?

One of the many forks of the Jargon File also has an appropriate entry for this topic which also includes the version of the maxim that you used.

"Bill Gates's intellectual property guru talks to Brad Stone..."

Please, folks, not everyone is a guru.

It's bad enough that hackers lost their good name.

All of this talk about IE, Photoshop, and "pngcrush" just shows that PNG is still not living up as totally suitable replacement for GIF.

PNG seems to be a good example of the second-system effect, where a simple implementation is replaced by "elephantine feature-laden monstrosity", that took a long time to develop, and is difficult to implement correctly. When the largest graphics companies, Adobe & Macromedia, have problems, you have to wonder.

Not that PNG doesn't have its uses, but if the goal was to create a patent-free replacement, it would have been a lot quicker and easier to come up with a "GIF2" rather than the be-all-end-all replacement. Anyway, it shouldn't come as a shock that people lived with the GIF patent for many years, and that PNG will probably never fully replace GIF. Just something for the "Burn All GIFs" crowd to consider.

The canonical geek sport is volleyball. Always has been.

Yes, this is the same type of process. Fluid when handled gently, but it becomes rigid when subjected to a sharp impact.

If you jab it, it feels hard and your finger won't go in very far. You can pour it slowly, but you can grab a clump of it, almost as if it's a solid. This kind of fluid is called dilatant. It becomes more viscous when agitated or compressed.

The cornstarch mixture is sometime called ooblick.

Thats pretty much the canonical example of why DWIM is a bad idea...

Ahem. Re-read the entry on "discipline and bondage". You'll see that C is precisely not the sort of thing they're referring to. They're referring to languages that don't give you enough rope to hang yourself because they've used up all that rope tying you up and making you conform to their idea of architectural purity. A language that operates at the same level as C but doesn't give you pointers or bitfields because they're "icky" would be an example of a B&D language (Java doesn't need pointers because its "platform", the JVM, doesn't require them ... though one could argue the java platform is indeed into BDSM)

Second, of course Netscape 4 runs faster on your old machine than does Mozilla (even though you really mean Firebird, since Mozilla was always meant as a testbed, and Firebird was meant to actually be a stand-alone browser like old versions of Netscape). Most software released 6 years ago will be smaller and less featureful than its modern equivalents, and Mozilla does a lot of stuff that Netscape couldn't dream of.

Third, HTML vs XHTML+CSS+etc. You are not a web designer. Even if someone's paying you to do it, you simply can't hold that opinion and be anything more than a Dreamweaverist or Frontpager. Why? Because noone in their right mind thinks the old, kludge-ridden, variably-parseable, non-standard markup language that is HTML before HTML4.01/STRICT is better in any way than XHTML. Which is easier: changing the global font by editing each and every <font> tag in your entire website or editing a single CSS file? Repeat for colors, layout, link styles, etc. There's no way I'd ever go back to doing all that stuff by hand.

Sometimes rewrites are wasteful (see also second-system effect). Your examples, though, are awful. You listed several things that are not good enough, and that needed to be rewritten.

Viruses aren't self-replicating. The smallest things that can handle that task are bacteria.

A virus needs to hijack the synthesis machinery of a cell in order to make more viruses. As noted by someone else on this thread, a virus is more like a quine--code that generates another copy of itself as output, but you still need a compiler to execute it.

In the sense of independent self-replicating machinery, cells are God's nanotech.

I'm not sure that we have come to the point of understanding where we can control nanobots.

That's definitely true, given that we haven't come to the point where we can build self-replicating nanobots.

Always be sure to mount a scratch monkey.

Real Programmers (TM) use INTERCAL.

All other computer languages, including, but not limited to, C, C++, or any other computer language, now known or later developed, is for lusers.

All other languages, including but not limited to C, Objective C, FORTRAN, C++, Java, COBOL, C#, Pascal, BASIC, and all other languages, compiled or interpreted, now known or later developed, shall bow before the majesty of INTERCAL, the One True Programming Language.

Bow before me for my operating system and all the programs that run on top of it are written entirely in INTERCAL, the master of all programming languages.

If you agree with me, go HERE to sign a petition to the federal government to illegalize all programming languages except INTERCAL. If you disagree with me, hear now and hear well: Real Programmers (tm) use INTERCAL.

INTERCAL is a registered trademark of Compiler Language With No Pronounceable Acronym Company, Incorporated.

But remember to always mount a scratch fly!

In brief: What science fiction do you recommend to people who hate science fiction?

The origin of Sturgeon's Law, widely quoted as "Ninety percent of everything is crap.", is related in the following anecdote:

"When people talk about the mystery novel," Ted [Sturgeon] said, as I remember, "they mention The Maltese Falcon and The Big Sleep. When they talk about the western, they say there's The Way West and Shane. But when they talk about science fiction, they call it 'that Buck Rogers stuff,' and they say 'ninety percent of science fiction is crud.' Well, they're right. Ninety percent of science fiction is crud. But then ninety percent of everything is crud, and it's the ten percent that isn't crud that is important. And the ten percent of science fiction that isn't crud is as good as or better than anything being written anywhere."

-- James Gunn, The New York Review of Science Fiction #85, September 1995

I have never read much science fiction, I as cannot bear bad science fiction. But, if I were given a reading list of very good science fiction, I think I would enjoy it greatly.

As such, can you give me a science fiction reading list that contains unassailable writing? What science fiction is both highly "literary" and just plain good? What science fiction will satisfy even the most "high-brow"/pretentious tastes?

here.

guru meditation n. Amiga equivalent of `panic' in UNIX (sometimes just called a `guru' or `guru event'). When the system crashes, a cryptic message of the form "GURU MEDITATION #XXXXXXXX.YYYYYYYY" may appear, indicating what the problem was. An Amiga guru can figure things out from the numbers. Generally a guru event must be followed by a Vulcan nerve pinch.

This term is (no surprise) an in-joke from the earliest days of the Amiga. There used to be a device called a `Joyboard' which was basically a plastic board built onto a joystick-like device; it was sold with a skiing game cartridge for the Atari game machine. It is said that whenever the prototype OS crashed, the system programmer responsible would calm down by concentrating on a solution while sitting cross-legged on a Joyboard trying to keep the board in balance. This position resembled that of a meditating guru. Sadly, the joke was removed in AmigaOS 2.04.

Although this looks promising on paper, one can but wonder if that turbine locomotiveis yet another boondoggle (sp?).

Given that the turbine's fuel appetite does not significantly changes when the turbine goes from idle to warp factor nine (unlike a diesel), one wonder what fuel economy will be with a SINGLE 5000 hp turbine engine. Okay, granted, with hotel power (to light-up the cars and air-condition/heat them), you still suck some power from the engine when the train is stopped.

One would think that a sensible way to address this problem would be to use 10 smal 500 HP engines whose number that kick-in would depend of the power needed at a time. But of course, this would mean higher maintenance costs and more chances for something to go wrong. However, modular design could make it easy to replace a turbine.

Case in point: the old United Aircraft Turbotrain, designed in the sixties, had 6 Pratt & Whitney PT-6 turbine engines, each one of which could be replaced by three mechanics in less than 30 minutes (this was a NICE train: the bar was in the engine's cab, so you could look at the track ahead whilst sipping a beer, and switch to a stiffer drink whenever the train missed a school bus or a gasoline truck at a crossing).

Even if we end-up with a super-magical turbine locomotive that runs all the time and doesn't suck fuel all the time as well, we'd face a little problem that is caused by the turbines's very suitableness for powering aircraft: low weight.

Of course, low weight means less power needed to go at high speed. But is means also less weight on drivers. Perhaps railroads will be clipping newspapers coupons looking for sand clerance sales (sand can be injected right in front of driving wheels to boost adhesion if the wheels start to slip Wheels will definitely slip if there isn't enough weight on them)...

So, one wonders of the suitableness of a turbine locomotive-hauled passenger train. Will it slip? Will it haul? I'm afraid that a turbine
engine will have to be weighted up... But that weight need not be always deadweight. Big cities call for big commuter train traffic: the thing electrification is for. So, why not add a pantograph and power transformer allowing for full-power operation under catenary when approaching terminals? At least, this will reduce downtown air pollution.

Wouldn't a better way be to have distributed traction throughout the train? You keep power generation in a lightweight power car (it would hardly be a locomotive anymore), and have traction motors throughout the train itself, so to take advantage of the weight there, too. Smaller traction motors, too, or at least, bigger derated ones. The first french TGVs had powered axles under the passenger coaches, and the Hikari Japanese bullet-trains running on the Shinkansen, as well as the newest german ICE trains have distributed powered axles through the trains (and the ICE-3 trains are real neat, too because the front seats of the first cars look on the track ahead, over the engineer's shoulders).

But of course, one hits other problems, such as safely sending traction power throughout the train. You're talking at something like 1000 amps at 600 volts there. Coupling/uncoupling cars would cause problems, and at each car, you also have connections that can go wrong. 100 years ago, in Paris, a subway train caught fire, killing more than 100 people. The cause: high-intensity traction power sent through flexible cables throughout the train. Such lessons from the past are not easily forgotten...

An lighter articulated train would be better in this respect, but then, you end-up with with an unflexible consist.

A normal train car has two 4 wheel trucks, one at each end. On an articulated train, adjacent cars share the same truck. Since trucks are rather heavy, you end up with significant weight savings (a 10 car normal train has 20 trucks, whereas an articulated 10 car train has only nine count them right, and don't do a fencepost error). The downside is that you cannot easily remove or add a car, they are all stuck together, so you have an inflexible train which can't be adjusted for varying loads.

But, again, adding cars and removing them is expensive, more expensive than hauling around empty seats (or it seems, looking the way some MBAs with adding machines seem to think in railroad adminive departments). But, after all, the french TGVs are articulated, so this is less a problem it might see.

Aha! Let's compromise on, oh, four car articulated, self-contained (1 first class parlour/club-car, 2 second class coaches, bar car & checked luggage/bicycle space with reversible control cab) units, two of which could be powered by one power car. So a 16 car train could be feasible, and you can retain some flexibility.

And then, do we have a tilt-train ? Tilt-trains are attractive, but is still one more thing that can go wrong. And with motorized trucks, you have less room to put the needed power-banking mechanisms...

A tilt train is a train that will tilt in the curves to compensate for cant deficiency . Cant deficiency is when the track is banked less than what would be needed for the speed the train goes through the curve. So, to prevent people from being tossed around curves, you simply tilt the train inwards, much like an airplane that does a banked turn. The new Bombardier Amtrak Acela train is a tilt-train, as well as the Bombardier VIA Rail LRC s that have been running for more than 20 years in Canada.

There are two kind of tilt-trains: passive-suspention and active-suspension . Passive-suspension tilt trains are simply hung down and swing out in curves, whilst active suspension trains have electronic acceleration sensors that control hydraulic rams that tilt the carbodies. The old United Aircraft Turbo train and the old Talgo Pendular trains had passive suspension.

It should work politically: engineers looove that kind of contraptions! And politicians looove to be associated with forward-thinking technology... But what kind of engineers? Aircraft engineers are clueless about railroad problems (one should remember the woes suffered by the late UAC turbo train), and railroad engineers are justifiably wary of sleek lightweight technology that falls apart at the slightest rail joint...

I am afraid that having efficient turbine power for high-speed passenger trains would end-up in a costlier, less flexible exercise than using electrified off-the-shelf technology in the long run...

2. What does TCPA / Palladium do, in ordinary English?

Its obvious application is to embed digital rights management (DRM) technology in the PC. The less obvious implications include making it easier for application software vendors to lock in their users.

4. How does it work?

likely implementation in the first phase of TCPA is a `Fritz' chip - a smartcard chip or dongle soldered to the motherboard.

5. What else can TCPA and Palladium be used for?

TCPA can be used to implement much stronger access controls on confidential documents. For example, you might arrange that your soldiers can only create word processing documents marked at `confidential' or above, and that only a TCPA PC with a certificate issued by your own armed forces can read such a document. This is called `mandatory access control', and governments are keen on it. The Palladium announcement implies that the Microsoft product will support this. Once TCPA is widespread, corporations can do this too - and so, for that matter, can the Mafia.

(believe it or not, much government-developed software is pretty damn good.)

OF COURSE government-developed software is pretty damn good!!!
They don't have fucking marketroids breathing down their necks!!!

--

Since everyone is lambasting Mass. as being clueless, the clueless.com site seems stunningly appropriate. It *cough* even mentions who invented the Internet. (no, not THAT guy).