Their deaths saved thousands more - and still do
on
Apollo 1
·
· Score: 5, Insightful
Nasa engineers believed that before the fire actually flashed (almost like a flashbulb, with all that exotic metal in a pure O2 atmosphere), the insulation smouldered for a bit. They decided that one way to prevent future accidents of that sort was to detect the smoke the preceededs the fire.
So they commissioned research to do so. And the result was the ionization-type smoke detector. Which you can now buy at any hardware store for as low as ten dollars, and which is required by zoning for virtually all human-habitable houses in the US and many other countries.
These devices have saved many thousands of lives so far, and will continue to do so.
These devices use a small radioactive source to ionize smoke particles, so they don't need to depend on natural ionization and can thus detect extremely miniscule amounts of smoke. This greatly increases their sensitivity, giving much earlier warning. The anti-nuclear hysteria was in full cry at the time. So it's unlikely a private company would have tried to design and market such a device for consumers. But for a NASA project, for short-term use above the atmosphere, it made sense. Once the device was done and its characteristics known, it was easy to show that a tiny amount of short-lived isotope, whose radiation doesn't leak beyond the container during the device's service life, was a miniscule risk compared to the number of lives saved. And a classic NASA spinout occurred.
So the fire and the deaths of the three astronauts was the direct cause of the invention and introduction of practical domestic smoke detectors, which otherwise certainly would not have been introduced for decades, if ever.
The purpose and effect of the DMCA is anti-competitive. wouldn't it be a beautiful irony if the DOJ took congress to court in an anti-trust trial?
They can't:
Congressional immunity. (You can't sue congress or congressmen for anything they say on the floor or any laws they pass.)
Separation of powers: (The Executive Branch can't go after the Legislative Branch by itself. In general it takes two branches to go after the third, as in impeachment. But the Congress is still more special: They can pretty much only be kicked out by each house's own rules and own internal procedures.)
Anti-trust laws only apply to the private sector. (The government IS a monopoly - by design. If you have two or more running at the same time it's usually called "civil war", though politicians often like to call it "anarchy" when it's really "polyarchy".)
Your recourse is to vote the bastards out. They're YOUR representatives, remember?
Now you might have to first deal with a lot of political machines that have corrupted the election process with everything from fake voters to gimmicked vote processing. But that's yet another ball of wax.
The other recourse is to press "reset" and start over. (That's what the Second Amendment is really for - that and keeping them in check because they know it MIGHT happen if they get TOO far out of hand and really annoy TOO many people.) But I strongly suggest you DON'T try it. Win or lose, a lot of people - probably including you - get hurt or dead and a lot of stuff that's still working gets broken. Then if it DOES work you never know in advance WHAT will boot up afterward.
So? It's NOT the business of the government or of me to guarantee any particular business a profit, or indeed that it stay in business.
But it IS the business of government to see that, when a person or organization has played by the government's rules and has something of value, they do not lose that value to someone who has NOT played by the government's rules.
You may have an argument with the rules - whether they're proper, or whether they're consistent with the rest of the rules - especially those that override lower-level rules. But that argument is not with Sony, but with the government: With the legislature (for constructing the obnoxious rule), or with the courts (for not resolving the inconsistency with other rules of the same or higher precedence).
That third article about the supposed "HCF" instruction on the 4004 is completely and utter BS. None of the instructions on the 4004 will cause it to burn up, even on the earliest production parts.
When the IBM System 360 series came out it had a large number of new opcodes (as compared with the 70x/70xx series). These were the days of CISC (Complex Instruction Set Computers), and the 360 really lived up to the name. It gave over a large amount of its word space to opcodes and opcode extensions, so it had a VERY large potential opcode space. Much of it was unpopulated, but some was populated with undocumented instructions. Further, the machine was microcoded, and the microcode was loaded when the machine powered up. (That's what floppy disks were invented for.) So the company could write new opcodes and add them later.
Of course the new machine with the ENORMOUS list of opcodes and (true) rumors of hidden undocumented opcodes quickly lead to the circulation of a humorous list of perhaps 20ish additional "new undocumented opcodes". Things like XOE (Execute Operator Immediately), EK (Electrify Keyboard), SSJ (Select Stacker and Jam), BLNK (Blink Lights), WHR (Whirr), etc. The crown jewel of this list was HCF (Halt and Catch Fire).
While this list was still funny Motorola released the 6800 single-chip microprocessor, predecessor to 650x knockoff that formed the core of the first Apple computers. To ease chip testing, the all-ones opcode threw the chip into a test mode, where it continuously incremented the program counter and performed memory reads. This wiggled all the address lines and most of the control lines, letting you know if the chip was alive and bonded.
Of course they didn't tell you about it. And of course the only way out was hard reset. And of course a jump to an unpopulated region of the address space (i.e. most of it) would leave the bus floating and generate 0xFF. And of course jumping into random data or uninitialized memory would also quickly get you an 0xFF or jump you off into unpopulated address space. So the typical behavior for a program bug was to lock up the processor beyond the ability of a debugger to function.
(Hell: I had one of the first round of solder-it-yourself evaluation kits, bent a pin on the debugger ROM putting it into the socket, and ended up with a board that booted into the test state. Was starving student and it took a couple days to get access to test equipment to find out what was wrong.)
So of course programmers, once they found out about the instruction that hung the chip in a mode where it "twiddled its thumbs at maximum speed" and got a bit warmer than usual, and couldn't get out of the mode except by hard reset, quickly christened the opcode "Halt and Catch Fire". And this became the generic term for get-stuck-in-a-test-mode instructions on microprocessors, until the chip manufacturers finally came to their senses and stopped putting such instructions into instruction sets.
What has always annoyed me are these people that build next to an airport that has been there for many years... then have the gall to complain a couple of years later about the jet noise they hear every day because of the airport that was there when they built their dream homes. If they didn't want the jet noise in the first place, they should have built somewhere else?
There's a legal doctrine about that. It's called "moving to a nuisance". Basically if you move into proximity with an annoyance that predates your move it's your fault for moving there and you have no gripe.
But enforcement of the doctrine in courts tends to be spotty in some places. Colorado and Oregon generally laugh such people out of court. But California seems to be the home of successful nuisance suits.
This kind of thing happens to small private-plane airports all the time. Developer builds devopment next to one, and after the people move in they drive the airport out of business with suits.
One such small-plane airport in Colorado came up with a great idea: After they'd gotten the suit laughed out of court, they bought up the fancy new houses that had been built next to their fence for a song. Then they put gates in the fence and ran driveways from the BACK of the carports to the taxiway. And resold the fancy houses at a significant profit to people with private planes - who NEVER complained about airport noise. B-)
I understand several other small airports in similar situations have done the same thing, or even had developers build such houses deliberately, and there's now a term for such a development - "Air Park" or something to that effect.
You know the screenwriter hasn't "gotten it" when the computer's OS terminates each interactive session by typing or saying (in a pretentious synthetic voice):
If the ID presented is proof of US citizenship, then the driver's license (or state ID, if the user merely wants an ID without driving privileges) is issued with the usual expiry date.
If the ID presented does not prove US citizenship, then proof of legal status in the US should be required. As this legal status may have an expiry date (e.g. TN or H-1B or student visa holders), then the ID or license should be issued with an expiry date no later than the expiry of the alien's status.
Simpler would be check boxes:
O US citizenship proven.
O US visa/landed status valid through (date) proven.
And ditto for Canada (due to treaty interlocks with the US and Canadian driver's license systems.)
Visa extensions and status changes could then be handled like address changes, rather than by making the driver's license self-destruct and complicating the bureaucrats' systems.
Failure to get a box checked, or presence beyond the date shown, wouldn't automatically mean they WEREN'T citizens - just that they hadn't proven it to the state driver's license issuer, or hadn't updated the license document. (So they'd better have secondary documentation available if they don't want trouble with the immigration authorities.)
Most (all?) states will issue non-drivers a state ID card, typically through the same agency that issues driver's licenses.
Essentially it's the same as a driver's license except it doesn't license you to drive. Use it to prove your identity, residency, and age, buy booze, cash checks, etc.
I don't know how many of you are aware of this, but this Gravenreuth guy, the attorney, is one of the most hated men in the German IT scene. He has been going to court over cases like these for at least ten years.
A similar thing happened in Silicon Valley. A guy got a patent on the bitwise-XOR cursor. This firm of three lawyers got hold of it. Then every time a hi-tek company was going public - WHETHER THEY USED AN XOR CURSOR OR NOT - they'd file suit. And offer to settle for some non-trivial pittance like $10,000.
Now going public is a touchy proposition, and having a suit pending against you can cost you big time, regardless of the suit's merits. So essentially all the companies knuckled under. It amounted to a small tax on going public.
And the people in question were the most hated in the Silicon Valley financial community for a decade or more. (There were rumors that a hit had been contracted...)
(Funny thing - I once worked for a company that had a patent on the BLINKING cursor, back in the days of character-only terminals. They never used it for this sort of thing - or litigated it at all. Instead they kept it in the pile of patents to be used in a countersuit against anyone who sued THEM for patent infringement.)
... forgetting to account for the radiation of energy and the resultant decrease in mass seems to be a fairly major oversight...
While we're at it, did they also overlook tides?
Just as the tidal friction of the moon on the earth is accellerating the moon (gradually moving its orbit outward while slowing the earth's rotation while friction-heating the earth's core and thrashing the oceans and atmosphere), the earth's tidal friction on the sun should be gradually moving the earth's orbit outward, at a cost to the sun's angular momentum.
This is because in both the earth/moon and sun/earth system the orbit and central body spin are in the same direction, with the central body spinning faster than the orbiting body's period. Tides raise "bumps" on the central body, which (thanks to damping from friction) are carried forward with the central body's spin and produce an accellerating force on the orbiting body.
In the sun/earth case the sun's tides on the earth also transfer some angular momentum from the earth's spin to its orbit, increasing the effect. That doesn't happen with the moon, because the moon has already used up its angular momentum and is tide-locked with the earth.
Nothing compared to the effect on Jupiter, of course. But as long as the earth's orbit isn't in a harmonic relationship with that of another major planet the effect should be nontrivial and the interactions with other planets should inetgrate out to zip.
So, was that taken into account, too? If not, the start of the bake cycle could be even further into the future.
Timothy: (do real astronomers call it the 'dark side,' when it gets plenty of light?)
brassman: If you're a RADIO astronomer, yes...the far side would be the dark side as far as you're concerned. The amount of radio crap we're spewing...
Also: The far side doesn't get light or solar radio noise reflected from the earth, while the near side sees the earth illuminated (at the nearest point: first quarter (half-lit) through full to last quarter) any time the sun is down.
Put two observatories a bit over the horizon from Earth on opposite sides and you get nearly continuous observation of the half-sky opposite the sun without interference from either the sun or the earth.
Don't put one EXACTLY opposite the earth: There's a diffuse "hot spot" of signal that diffracted around the moon there - diffuse because the moon isn't a sphere smooth down to radio or light wavelengths.
Can anyone speculate how easy it would be to sustain a stable orbit around the moon for long periods of time?
There are 5 lagrange points in a two-body system such as earth-moon. The L2 point behind the moon is unstable, but a very small amount of station-keeping thrust every now and then would keep a relay satellite there.
The moon obscures L2 from earth. But you could do a second bounce off a satellite at L4 or L5. Those are 1/6th of the way around the orbit behind and ahead of the moon and are stable second order - a satellite drifts off the potential peak but then ends up in a stable orbit around it.
With the RCA patent on field emission expiring, and practical long-life field emitters being debugged, we might have the potential for a vacuum tube rennisance.
Vacuum tubes are inherently fast. Electrons travel much faster than holes during conduction, and when traveling across the gap (where the switching takes place) they are making a single free-path hop - similar to the fast N-type FETs (gate shorter than mean-free-path) that are currently being researched.
Use a field-emission cathode and shorten the gap to something comparable to that of a transistor in an integrated circuit and you can use voltages comparable to those of an IC also - but you can also scale up voltage and power arbitrarily at the I/O "pins" without substrate breakover. Meanwhile, at the lower voltages of the internal circuitry you don't have the tip-erosion problem from ion-bombardment.
So there's potential for vacuum integrated circuits on about the same size scale as semiconductor integrated circuits, but made of glass, metal, and diamond. They could run faster than semiconductors, and do a number of other useful tricks to electrons in flight (like "bunching" for microwave amplification) that are impractical in a semiconductor. Vacuum electronics can do many things in one step that can take hundreds or thousands of steps in semiconductors.
Downside is that you don't have complimentary charge conductors, so you don't get a CMOS equivalent. (Unless you use positrons. Maybe that's what Asimov's robot brains were up to. B-) ) So you'll still drop power in resistors (or use inductors to pair up two electron tubes when you don't need the low-frequency/DC end of a signal). But you can let the whole IC get cherry-red with waste heat so that's not a problem in many applications where the power is available.
Vacuum tubes - even low-voltage vacuum ICs - are inherently immune to many harsh environmental factors (like heat and radiation) which give semiconductors heartburn.
Field emission could also give a new lease on life to many conventional vacuum-tube applications. (Tubes are still used in high-power applications at high frequencies - like radio and radar.) It's a drop-in substitute for a heated cathode.
But embed a vacuum-electronic integrated circuit to do the detail work within a cold-emission vacuum power device and you have a bunch of "killer apps". Multiple "tubes" in one vacuum bottle, and even some embedded integrated driver circuitry, had been experimented with. But now we're talking a single vacuum "tube" with a very long life (no burnout - fadeout after many years if ever) with an entere application built in.
Think a wafer-sized cellphone, a bottle-sized cellular base station or broadcast TV transmitter, or putting the whole set of electronics for an airport radar INSIDE the magnetron. Then think "one device with a guaranteed minimum life of decades" rather than "keep replacing burned-out tubes".
Now think about putting these in space probes. (Heck - once it's up you don't even need the vacuum envelope. B-) )
That got me thinking. Currently, all CRT's have one set of electron guns at the center of the screen. Would it be possible to partition the screen into, say, four areas, each of which is painted by it's own set of guns.
That's been done. But it's a problem getting the pictures to join.
Better is gun-per-pixel. That's also been done but this should make it practical.
The only flatpanel displays that look as good as a CRT to me are extreamly expensive. All the ones that are of lower cost have that wacky ghosting effect, not as bad as the old dual scan LCD displays, but it's there when playing higher framerate games..
That was recently solved - by remembering the previous frame and computing a voltage that would rapidly force the liquid crystal to the correct transparency rather than feeding it the voltage that would eventuall lead to it stabilizing at the desired transparency and letting it relax to that transparency in its own sweet time. There was an article about it maybe a month ago in slashdot.
Exepct TV-rate LCDs without ghosts as soon as this gets incorporated into the driver electronics - assuming the patent holders don't sit on it.
hot CRTs waste energy... they don't require any actual heating for operation, and I doubt (though I don't know) that they'll even get warm in operation.
The CRT's heater wastes some energy. But most of the energy consumed in a monitor is the energy dumped when the magnetic deflection field "flys back" at the end of each horizontal scan line. Some of this is recycled - into the accelleration high-voltage supply or even powering the CRT's heater - but most is just dumped as waste heat.
This is just a replacement electron gun, so it won't do anything about the deflection power waste. But see my other posting and its parent here.
I have reed an article an year ago about these screens, if they get them to work they should be really cool. They have almost all advantages of TFTs but have also almost all advantages of CRTs. They are very flat like a TFT...
I think you may have two articles conflated.
This one seems to be talking about using a diamond "forest" of cold emitters to replace the heated-cathode in a conventional electron gun, then deflecting the beam in the standard fashion, leading to an ordinary rectangular-cone CRT (but with no heater and instant-on).
You seem to be referring to another approach that was to use cold-emitters (which would also benefit from this breakthrough.):
The display consisted of a (glass) honeycomb of short individual "tubes".
Each "tube" had a single emitter "spike" (substitute "small forest") at the base.
A control electrode near the emitter (maybe substitute one per emitter in the "forest") switched it on/off and modulated the beam intensity. The voltage is near the cathode's and the voltage swing is just a couple volts, so you can use conventional transistor electronics.
(You can actually use two or more electrodes to do a matrix address and beam modulation, with the voltage gradient at the emitter tip or a space charge near it performing the computation so you don't need a separate switch per-pixel.)
The beam was accellerated along the narrow channel - the front portion of which contained an accelleration electrode with a constant high voltage - similar to a normal CRT. Difference: The beam could be bounced repeatedly between the channel walls, picking up additional electrodes by secondary emission.
The beam strikes a single phosphor dot at the end of the channel.
So you end up with something that can be fabricated (except for the cathode spike and maybe the modulation electrodes) by glass molding, vapor deposition of electrode metal, and micropipette phosphor-solution placement, and driven by essentially the same chips that run an LCD plus a single, unmodulated, high-voltage supply. The tubes are very short and the honeycomb of glass separating the individual tubes also supports the front screen, so you don't need thick heavy glass to fight 15 PSI of atmospheric pressure across more than a foot of unsupported span. Pixel placement is controlled by fabrication, so there's no sensitivity to local magnetic fields, no geometry adjustment. Of course in addition to no need to heat the cathodes there's no need to power and rapidly modulate an enormous magnetic deflection field.
And this new article tells you why we don't yet have either the cold-emission conventional CRT or the honeycomb flat-panel CRT: Positive ions from any impurities in the vacuum or kicked off the target or the sides of the channel are accellerated back toward the gun, slamming into the tip(s) and rapidly eroding it. RCA had a patent on field emission vacuum tubes but didn't feel like pursuing the technology with materials research. So the whole filed languished.
One of the biggest problems in the development of these things is that there isn't that much room between the electron emiter and the phosphorus, because of that they couldn't speedup the electron to the same speeds they get in a normal CRT and need to find new low-energy phosphoruses.
Huh? Space shouldn't be an issue. The final velocity of the electron only depends on the accelleration voltage, not the length of the path. The path only needs to be long enough to prevent arc-over along the surface of the glass (or in any residual gas in the "vacuum"), and that's a fraction of an inch.
With a conventional tube the voltage gradient also has to be low enough that the electrodes don't bend out of place. But that limit would be MUCH higher with the electrodes plated onto a glass surface or supported by the walls of a pixel-wide glass honeycomb cell, rather than by mica spacers and thin copper wire.
I expect the conventional-CRT style to come out first. It's only being held back by the RCA patent that just expired. The flat-panel might take longer, due to other patents, the need to build a "wafer" the size of the screen rather than the size of an electron-gun cathode, and possibly worse problems with tip erosion due to the limited number of tips per pixel.
The whole point to having a regular contributor to/. is to spur not only debate but action.... Maybe somebody like Lessig can help crystallize the/. community into getting real grassroots political leverage.
Personally, I'm doubtful - too many of us are frustrated, cynical, jaded, and face it, just plain lazy... We feel powerless and disenfranchised so... we stew in apathy
One source of the perception of apathy is that the opportunities to DO something are not effectively made public. If you don't spend all your time digging through obscure documents (or hire a lobbiest to do it FOR you), you don't know what the congress critters or regulators are up to.
Finding out on your own about legislation or rules that might affect you is SEVERAL full-time jobs. (Hell - they pass laws and impose regulations faster than a man can read, and the legislature doesn't even read most of what it votes on.) This stuff gets buried so deep (often deliberately) that even major corporations with entire DEPARTMENTS watching for things that affect them sometimes miss it. Responding to this crap is IMPORTANT. But people who have real work to do - and even those who don't - just can't spend enough time to hunt it down.
And that's where organizations like the EFF can come in.
I note that a number of Slashdot people (myself included) filed comments on DCMA rulemaking with the copyright office in response to a posting (within a day or so of the cutoff) that the opportunity existed.
When "action" consists of demonstrating that some people out there have an opinion, and the message can be sent by phone, email, or snail mail, the Slashdot Effect can be put to effective use. Combine that with a TIMELY article telling where to submit comments and we can flood them out.
I emphasize TIMELY because a day's notice will yeild a bunch of flames and a few well-thought-out letters, in a burst that makes it obvious they're all from a single community. Notice near the start of a comment period and a reminder a few days from the end will produce a large yeild of well-thought and well-researched comments.
This is not just speculation. This is exactly what the pro-gun movement does. The NRA (with membership second only to ARP) and its lobbying arm the ILA gets the press. But it is only one of a number of groups (GOA, Members' Councils, state orgs like CRPA, JPFO, etc.) which send out "legislative alerts" to their members when something is discovered in time to react. (They also interview and rate candidates come election time and make their voting records known.)
So instead of moaning about how apathetic the working nerds are (demotivating them further), perhaps EFF might put their staff to work to act as our early warning system, and SUBMIT SHORT ARTICLES to Slashdot when there are opportunities available for an avalanch of electronic or paper mail and phone calls. (Many of us have printers, right? We can pretty much ALL send a handwritten, hand-addressed letter, and those are even MORE effective.)
I'm sure the Slashdot editors would be happy to post such notices. They even have categories for them! Legislative and regulatory rulemaking on IP and other aspects of internet and programming freedom are PRIME examples of "News for Nerds, Stuff that Matters!". How could they resist a couple EFF staffers acting as investigative reporters - for free?
With EFF (and others) to dig up the info, Slashdot (and others) to spread the word in a TIMELY fashion and serve as dissusion and planning venues, and HORDES of internet-connected VOTERS to flood the critters with mail, we could turn a lot of this around.
But nothing will happen unless those hordes have the necessary INFORMATION to act.
And that's what the computer age is about, isn't it?
You'd think this would be an easy distinction to understand: We live in a country where 10 children are killed by hand guns every day. But Smith and Wesson doesn't worry that the FBI will come arrest them because someone used their technology to commit a crime.
The point is valid. But the statistic is bogus. You are echoing a big-lie propaganda sound bite as if it were a fact, when the true numbers are actually more than an order of magnitude lower.
Judges (and others, both in the general public and in policy-making offices) are becoming aware of the actual numbers. So continued use of this lie in arguments will tend to discredit the rest of your argument.
It is also a hotbutton issue, with the opposing view held by members of a subculture which constitutes about half the voting population. So by using a well-known big-lie from one side of this unrelated debate you risk instantly alienating half your audience.
Ten per day comes out to 3650 per (non-leap) year.
According to the National Safety Council, 230 children under the age of 15 were killed in firearms-related accidents in 1991. (That's the latest year for which figures are available at the time of the composition of the rec.guns FAQ found here.)
According to the most recent FBI Uniform Crime report (2000), available here, murder and negligent homicide of children ages 0 through 12 (by ALL means, not just firearms) totaled only 641. (The bulk of those were infants, implying most were killed by abuse or neglect rather than with-firearm murder.)
The numbers jump beginning with the 13 thorugh 16 age group, as gang-related murders begin to dominate the stats. But to get to "ten children killed by handugns per day" you need to consider "childhood" to extent into the mid-thirties.
So I recommend that, if you wish to continue using the analogy in future argumnets on the subject, you drop the bogus statistic. You can make the point even better using a real one ("robbers use handguns to kill two people per day"), or skipping the stats entirely, thus minimizing the risk of offending people who otherwise would be your allies.
Looks like some powerful players want to see Linux going toe to toe with Unix 'big iron.'
This isn't a linux issue. It's a hardware issue.
The significant thing about 'big iron' is that it's an enabling hardware technology.
Once you have it you can write firmware and software that creates the illusion that the hardware never fails
Until you have it, you can't.
The hardware described looks about right - if they handled machine checks properly. (And the fact that they even used the term implies they either did or are trying.) Basic idea: The machine catches ANY error, with enough state saved that you can:
CORRECT the error
IDENTIFY any failed components,
MOVE tasks to non-failed components or reconfigure the failing components to limp along,
NOTIFY the OS of any problem, so it can do things like start moving things off a dying component, and
pick up the computation where it left off WITHOUT the error.
When you can do this you can write a modified Linux, Windows, BeOS, or what-have-you that can do the things a mainframe can. (But you'll need to have a REALLY reliable OS for your starting point - you're now talking uptimes measured in decades. The software better not take the system down in the absense of hardware trouble, and there IS NO hardware troube. B-) )
Hot-swappable parts are more a side-effect than something key. You have to be able to hot-swap to replace a broken part with the system live. Once you have the ability to hot-swap in a replacement for a failed part and add it back into a running domain, it's trivial to generalize that to "fix" parts that were "bad" because they had never been installed.
Partitioning is also implied: You need a minimum of two domains ("virtual machine" subsets of the total device) - working (where the live system is) and diagnostic (where the maintainence guys check out the parts). Once you have that mechanism, making a LARGE number of working domains (with varying amounts of resource, including full or time-shared CPUs) is straightforward.
The guy even works three blocks from the WhiteHouse.
The software is developed in a suburb of Seattle Washington (state) and the company's security chief works in Washington (DC), nearly as far from the software department as you can get and still be in the continental US.
THAT explains the security problems in Microsoft products!
I think it doesn't make any difference whether it is open source or closed source, it's a matter of identifying them once the product is released.
So...who cares if there are problems. We'll find them eventually - as soon as someone exploits them and we hear about it.
Precicely.
If you want bug-free code you need to start at the architecture/design process (avoiding bug-prone choices), then debug as you go. It's like growing a perfect crystal - you push the impurities out as it solidifies, so only the boundary needs attention. The longer you wait, the larger your search space for each bug, and the bigger the hive of ofspring each bug has produced as new code was added to buggy code.
Security issues are a special case of "bugs", with more than the typical amount of effort needed at precoding stages to avoid building unfixable problems into the basic architecture.
I wonder if they release their code like that for QA as well. It's a matter of identifying bugs once the product is released.
My impression is that Schmidt is completely unaware that software QA, or any other pre-release potential for (securyty) bug suppression, exists. At a minimum his statement implies that Security as a department doesn't participate in architecture, design, code reviews, or QA, and that its leader either feels no need to do so, or is deliberately directing attention away from an inability to affect those stages.
That the head of security for Microsoft could emit such an answer is appalling. But it also goes a long way toward explaining the security problems in Microsoft products.
Slashdot Mirror
Nasa engineers believed that before the fire actually flashed (almost like a flashbulb, with all that exotic metal in a pure O2 atmosphere), the insulation smouldered for a bit. They decided that one way to prevent future accidents of that sort was to detect the smoke the preceededs the fire.
So they commissioned research to do so. And the result was the ionization-type smoke detector. Which you can now buy at any hardware store for as low as ten dollars, and which is required by zoning for virtually all human-habitable houses in the US and many other countries.
These devices have saved many thousands of lives so far, and will continue to do so.
These devices use a small radioactive source to ionize smoke particles, so they don't need to depend on natural ionization and can thus detect extremely miniscule amounts of smoke. This greatly increases their sensitivity, giving much earlier warning. The anti-nuclear hysteria was in full cry at the time. So it's unlikely a private company would have tried to design and market such a device for consumers. But for a NASA project, for short-term use above the atmosphere, it made sense. Once the device was done and its characteristics known, it was easy to show that a tiny amount of short-lived isotope, whose radiation doesn't leak beyond the container during the device's service life, was a miniscule risk compared to the number of lives saved. And a classic NASA spinout occurred.
So the fire and the deaths of the three astronauts was the direct cause of the invention and introduction of practical domestic smoke detectors, which otherwise certainly would not have been introduced for decades, if ever.
They can't:
Congressional immunity. (You can't sue congress or congressmen for anything they say on the floor or any laws they pass.)
Separation of powers: (The Executive Branch can't go after the Legislative Branch by itself. In general it takes two branches to go after the third, as in impeachment. But the Congress is still more special: They can pretty much only be kicked out by each house's own rules and own internal procedures.)
Anti-trust laws only apply to the private sector. (The government IS a monopoly - by design. If you have two or more running at the same time it's usually called "civil war", though politicians often like to call it "anarchy" when it's really "polyarchy".)
Your recourse is to vote the bastards out. They're YOUR representatives, remember?
Now you might have to first deal with a lot of political machines that have corrupted the election process with everything from fake voters to gimmicked vote processing. But that's yet another ball of wax.
The other recourse is to press "reset" and start over. (That's what the Second Amendment is really for - that and keeping them in check because they know it MIGHT happen if they get TOO far out of hand and really annoy TOO many people.) But I strongly suggest you DON'T try it. Win or lose, a lot of people - probably including you - get hurt or dead and a lot of stuff that's still working gets broken. Then if it DOES work you never know in advance WHAT will boot up afterward.
B-)
So? It's NOT the business of the government or of me to guarantee any particular business a profit, or indeed that it stay in business.
But it IS the business of government to see that, when a person or organization has played by the government's rules and has something of value, they do not lose that value to someone who has NOT played by the government's rules.
You may have an argument with the rules - whether they're proper, or whether they're consistent with the rest of the rules - especially those that override lower-level rules. But that argument is not with Sony, but with the government: With the legislature (for constructing the obnoxious rule), or with the courts (for not resolving the inconsistency with other rules of the same or higher precedence).
That third article about the supposed "HCF" instruction on the 4004 is completely and utter BS. None of the instructions on the 4004 will cause it to burn up, even on the earliest production parts.
When the IBM System 360 series came out it had a large number of new opcodes (as compared with the 70x/70xx series). These were the days of CISC (Complex Instruction Set Computers), and the 360 really lived up to the name. It gave over a large amount of its word space to opcodes and opcode extensions, so it had a VERY large potential opcode space. Much of it was unpopulated, but some was populated with undocumented instructions. Further, the machine was microcoded, and the microcode was loaded when the machine powered up. (That's what floppy disks were invented for.) So the company could write new opcodes and add them later.
Of course the new machine with the ENORMOUS list of opcodes and (true) rumors of hidden undocumented opcodes quickly lead to the circulation of a humorous list of perhaps 20ish additional "new undocumented opcodes". Things like XOE (Execute Operator Immediately), EK (Electrify Keyboard), SSJ (Select Stacker and Jam), BLNK (Blink Lights), WHR (Whirr), etc. The crown jewel of this list was HCF (Halt and Catch Fire).
While this list was still funny Motorola released the 6800 single-chip microprocessor, predecessor to 650x knockoff that formed the core of the first Apple computers. To ease chip testing, the all-ones opcode threw the chip into a test mode, where it continuously incremented the program counter and performed memory reads. This wiggled all the address lines and most of the control lines, letting you know if the chip was alive and bonded.
Of course they didn't tell you about it. And of course the only way out was hard reset. And of course a jump to an unpopulated region of the address space (i.e. most of it) would leave the bus floating and generate 0xFF. And of course jumping into random data or uninitialized memory would also quickly get you an 0xFF or jump you off into unpopulated address space. So the typical behavior for a program bug was to lock up the processor beyond the ability of a debugger to function.
(Hell: I had one of the first round of solder-it-yourself evaluation kits, bent a pin on the debugger ROM putting it into the socket, and ended up with a board that booted into the test state. Was starving student and it took a couple days to get access to test equipment to find out what was wrong.)
So of course programmers, once they found out about the instruction that hung the chip in a mode where it "twiddled its thumbs at maximum speed" and got a bit warmer than usual, and couldn't get out of the mode except by hard reset, quickly christened the opcode "Halt and Catch Fire". And this became the generic term for get-stuck-in-a-test-mode instructions on microprocessors, until the chip manufacturers finally came to their senses and stopped putting such instructions into instruction sets.
What has always annoyed me are these people that build next to an airport that has been there for many years ... then have the gall to complain a couple of years later about the jet noise they hear every day because of the airport that was there when they built their dream homes. If they didn't want the jet noise in the first place, they should have built somewhere else?
There's a legal doctrine about that. It's called "moving to a nuisance". Basically if you move into proximity with an annoyance that predates your move it's your fault for moving there and you have no gripe.
But enforcement of the doctrine in courts tends to be spotty in some places. Colorado and Oregon generally laugh such people out of court. But California seems to be the home of successful nuisance suits.
This kind of thing happens to small private-plane airports all the time. Developer builds devopment next to one, and after the people move in they drive the airport out of business with suits.
One such small-plane airport in Colorado came up with a great idea: After they'd gotten the suit laughed out of court, they bought up the fancy new houses that had been built next to their fence for a song. Then they put gates in the fence and ran driveways from the BACK of the carports to the taxiway. And resold the fancy houses at a significant profit to people with private planes - who NEVER complained about airport noise. B-)
I understand several other small airports in similar situations have done the same thing, or even had developers build such houses deliberately, and there's now a term for such a development - "Air Park" or something to that effect.
You know the screenwriter hasn't "gotten it" when the computer's OS terminates each interactive session by typing or saying (in a pretentious synthetic voice):
"End of Line"
Bruce Boxleitner was also the second captain on Babylon 5. Now if only they'd release that series on DVD...
For THAT I'd buy a stand-alone DVD player.
If the ID presented is proof of US citizenship, then the driver's license (or state ID, if the user merely wants an ID without driving privileges) is issued with the usual expiry date.
If the ID presented does not prove US citizenship, then proof of legal status in the US should be required. As this legal status may have an expiry date (e.g. TN or H-1B or student visa holders), then the ID or license should be issued with an expiry date no later than the expiry of the alien's status.
Simpler would be check boxes:
O US citizenship proven.
O US visa/landed status valid through (date) proven.
And ditto for Canada (due to treaty interlocks with the US and Canadian driver's license systems.)
Visa extensions and status changes could then be handled like address changes, rather than by making the driver's license self-destruct and complicating the bureaucrats' systems.
Failure to get a box checked, or presence beyond the date shown, wouldn't automatically mean they WEREN'T citizens - just that they hadn't proven it to the state driver's license issuer, or hadn't updated the license document. (So they'd better have secondary documentation available if they don't want trouble with the immigration authorities.)
Most (all?) states will issue non-drivers a state ID card, typically through the same agency that issues driver's licenses.
Essentially it's the same as a driver's license except it doesn't license you to drive. Use it to prove your identity, residency, and age, buy booze, cash checks, etc.
It was clear a few years ago that they were setting up a national ID via the drivers' license databases.
That happened when they changed the law to require the states to collect social security numbers and link them to the licenses in their databases.
(I believe the excuse used was tracking down absentee fathers who were delinquent in their child support payments.)
I don't know how many of you are aware of this, but this Gravenreuth guy, the attorney, is one of the most hated men in the German IT scene. He has been going to court over cases like these for at least ten years.
A similar thing happened in Silicon Valley. A guy got a patent on the bitwise-XOR cursor. This firm of three lawyers got hold of it. Then every time a hi-tek company was going public - WHETHER THEY USED AN XOR CURSOR OR NOT - they'd file suit. And offer to settle for some non-trivial pittance like $10,000.
Now going public is a touchy proposition, and having a suit pending against you can cost you big time, regardless of the suit's merits. So essentially all the companies knuckled under. It amounted to a small tax on going public.
And the people in question were the most hated in the Silicon Valley financial community for a decade or more. (There were rumors that a hit had been contracted...)
(Funny thing - I once worked for a company that had a patent on the BLINKING cursor, back in the days of character-only terminals. They never used it for this sort of thing - or litigated it at all. Instead they kept it in the pile of patents to be used in a countersuit against anyone who sued THEM for patent infringement.)
... forgetting to account for the radiation of energy and the resultant decrease in mass seems to be a fairly major oversight ...
While we're at it, did they also overlook tides?
Just as the tidal friction of the moon on the earth is accellerating the moon (gradually moving its orbit outward while slowing the earth's rotation while friction-heating the earth's core and thrashing the oceans and atmosphere), the earth's tidal friction on the sun should be gradually moving the earth's orbit outward, at a cost to the sun's angular momentum.
This is because in both the earth/moon and sun/earth system the orbit and central body spin are in the same direction, with the central body spinning faster than the orbiting body's period. Tides raise "bumps" on the central body, which (thanks to damping from friction) are carried forward with the central body's spin and produce an accellerating force on the orbiting body.
In the sun/earth case the sun's tides on the earth also transfer some angular momentum from the earth's spin to its orbit, increasing the effect. That doesn't happen with the moon, because the moon has already used up its angular momentum and is tide-locked with the earth.
Nothing compared to the effect on Jupiter, of course. But as long as the earth's orbit isn't in a harmonic relationship with that of another major planet the effect should be nontrivial and the interactions with other planets should inetgrate out to zip.
So, was that taken into account, too? If not, the start of the bake cycle could be even further into the future.
Timothy: (do real astronomers call it the 'dark side,' when it gets plenty of light?)
...
brassman: If you're a RADIO astronomer, yes...the far side would be the dark side as far as you're concerned. The amount of radio crap we're spewing
Also: The far side doesn't get light or solar radio noise reflected from the earth, while the near side sees the earth illuminated (at the nearest point: first quarter (half-lit) through full to last quarter) any time the sun is down.
Put two observatories a bit over the horizon from Earth on opposite sides and you get nearly continuous observation of the half-sky opposite the sun without interference from either the sun or the earth.
Don't put one EXACTLY opposite the earth: There's a diffuse "hot spot" of signal that diffracted around the moon there - diffuse because the moon isn't a sphere smooth down to radio or light wavelengths.
Can anyone speculate how easy it would be to sustain a stable orbit around the moon for long periods of time?
There are 5 lagrange points in a two-body system such as earth-moon. The L2 point behind the moon is unstable, but a very small amount of station-keeping thrust every now and then would keep a relay satellite there.
The moon obscures L2 from earth. But you could do a second bounce off a satellite at L4 or L5. Those are 1/6th of the way around the orbit behind and ahead of the moon and are stable second order - a satellite drifts off the potential peak but then ends up in a stable orbit around it.
See an explanation here
With the RCA patent on field emission expiring, and practical long-life field emitters being debugged, we might have the potential for a vacuum tube rennisance.
Vacuum tubes are inherently fast. Electrons travel much faster than holes during conduction, and when traveling across the gap (where the switching takes place) they are making a single free-path hop - similar to the fast N-type FETs (gate shorter than mean-free-path) that are currently being researched.
Use a field-emission cathode and shorten the gap to something comparable to that of a transistor in an integrated circuit and you can use voltages comparable to those of an IC also - but you can also scale up voltage and power arbitrarily at the I/O "pins" without substrate breakover. Meanwhile, at the lower voltages of the internal circuitry you don't have the tip-erosion problem from ion-bombardment.
So there's potential for vacuum integrated circuits on about the same size scale as semiconductor integrated circuits, but made of glass, metal, and diamond. They could run faster than semiconductors, and do a number of other useful tricks to electrons in flight (like "bunching" for microwave amplification) that are impractical in a semiconductor. Vacuum electronics can do many things in one step that can take hundreds or thousands of steps in semiconductors.
Downside is that you don't have complimentary charge conductors, so you don't get a CMOS equivalent. (Unless you use positrons. Maybe that's what Asimov's robot brains were up to. B-) ) So you'll still drop power in resistors (or use inductors to pair up two electron tubes when you don't need the low-frequency/DC end of a signal). But you can let the whole IC get cherry-red with waste heat so that's not a problem in many applications where the power is available.
Vacuum tubes - even low-voltage vacuum ICs - are inherently immune to many harsh environmental factors (like heat and radiation) which give semiconductors heartburn.
Field emission could also give a new lease on life to many conventional vacuum-tube applications. (Tubes are still used in high-power applications at high frequencies - like radio and radar.) It's a drop-in substitute for a heated cathode.
But embed a vacuum-electronic integrated circuit to do the detail work within a cold-emission vacuum power device and you have a bunch of "killer apps". Multiple "tubes" in one vacuum bottle, and even some embedded integrated driver circuitry, had been experimented with. But now we're talking a single vacuum "tube" with a very long life (no burnout - fadeout after many years if ever) with an entere application built in.
Think a wafer-sized cellphone, a bottle-sized cellular base station or broadcast TV transmitter, or putting the whole set of electronics for an airport radar INSIDE the magnetron. Then think "one device with a guaranteed minimum life of decades" rather than "keep replacing burned-out tubes".
Now think about putting these in space probes. (Heck - once it's up you don't even need the vacuum envelope. B-) )
That got me thinking. Currently, all CRT's have one set of electron guns at the center of the screen. Would it be possible to partition the screen into, say, four areas, each of which is painted by it's own set of guns.
That's been done. But it's a problem getting the pictures to join.
Better is gun-per-pixel. That's also been done but this should make it practical.
See another set of postings in this thred here.
The only flatpanel displays that look as good as a CRT to me are extreamly expensive. All the ones that are of lower cost have that wacky ghosting effect, not as bad as the old dual scan LCD displays, but it's there when playing higher framerate games..
That was recently solved - by remembering the previous frame and computing a voltage that would rapidly force the liquid crystal to the correct transparency rather than feeding it the voltage that would eventuall lead to it stabilizing at the desired transparency and letting it relax to that transparency in its own sweet time. There was an article about it maybe a month ago in slashdot.
Exepct TV-rate LCDs without ghosts as soon as this gets incorporated into the driver electronics - assuming the patent holders don't sit on it.
hot CRTs waste energy ... they don't require any actual heating for operation, and I doubt (though I don't know) that they'll even get warm in operation.
The CRT's heater wastes some energy. But most of the energy consumed in a monitor is the energy dumped when the magnetic deflection field "flys back" at the end of each horizontal scan line. Some of this is recycled - into the accelleration high-voltage supply or even powering the CRT's heater - but most is just dumped as waste heat.
This is just a replacement electron gun, so it won't do anything about the deflection power waste. But see my other posting and its parent here.
I think you may have two articles conflated.
This one seems to be talking about using a diamond "forest" of cold emitters to replace the heated-cathode in a conventional electron gun, then deflecting the beam in the standard fashion, leading to an ordinary rectangular-cone CRT (but with no heater and instant-on).
You seem to be referring to another approach that was to use cold-emitters (which would also benefit from this breakthrough.):
The display consisted of a (glass) honeycomb of short individual "tubes".
Each "tube" had a single emitter "spike" (substitute "small forest") at the base.
A control electrode near the emitter (maybe substitute one per emitter in the "forest") switched it on/off and modulated the beam intensity. The voltage is near the cathode's and the voltage swing is just a couple volts, so you can use conventional transistor electronics.
(You can actually use two or more electrodes to do a matrix address and beam modulation, with the voltage gradient at the emitter tip or a space charge near it performing the computation so you don't need a separate switch per-pixel.)
The beam was accellerated along the narrow channel - the front portion of which contained an accelleration electrode with a constant high voltage - similar to a normal CRT. Difference: The beam could be bounced repeatedly between the channel walls, picking up additional electrodes by secondary emission.
The beam strikes a single phosphor dot at the end of the channel.
So you end up with something that can be fabricated (except for the cathode spike and maybe the modulation electrodes) by glass molding, vapor deposition of electrode metal, and micropipette phosphor-solution placement, and driven by essentially the same chips that run an LCD plus a single, unmodulated, high-voltage supply. The tubes are very short and the honeycomb of glass separating the individual tubes also supports the front screen, so you don't need thick heavy glass to fight 15 PSI of atmospheric pressure across more than a foot of unsupported span. Pixel placement is controlled by fabrication, so there's no sensitivity to local magnetic fields, no geometry adjustment. Of course in addition to no need to heat the cathodes there's no need to power and rapidly modulate an enormous magnetic deflection field.
And this new article tells you why we don't yet have either the cold-emission conventional CRT or the honeycomb flat-panel CRT: Positive ions from any impurities in the vacuum or kicked off the target or the sides of the channel are accellerated back toward the gun, slamming into the tip(s) and rapidly eroding it. RCA had a patent on field emission vacuum tubes but didn't feel like pursuing the technology with materials research. So the whole filed languished.
One of the biggest problems in the development of these things is that there isn't that much room between the electron emiter and the phosphorus, because of that they couldn't speedup the electron to the same speeds they get in a normal CRT and need to find new low-energy phosphoruses.
Huh? Space shouldn't be an issue. The final velocity of the electron only depends on the accelleration voltage, not the length of the path. The path only needs to be long enough to prevent arc-over along the surface of the glass (or in any residual gas in the "vacuum"), and that's a fraction of an inch.
With a conventional tube the voltage gradient also has to be low enough that the electrodes don't bend out of place. But that limit would be MUCH higher with the electrodes plated onto a glass surface or supported by the walls of a pixel-wide glass honeycomb cell, rather than by mica spacers and thin copper wire.
I expect the conventional-CRT style to come out first. It's only being held back by the RCA patent that just expired. The flat-panel might take longer, due to other patents, the need to build a "wafer" the size of the screen rather than the size of an electron-gun cathode, and possibly worse problems with tip erosion due to the limited number of tips per pixel.
The whole point to having a regular contributor to /. is to spur not only debate but action. ... Maybe somebody like Lessig can help crystallize the /. community into getting real grassroots political leverage.
... We feel powerless and disenfranchised so ... we stew in apathy
Personally, I'm doubtful - too many of us are frustrated, cynical, jaded, and face it, just plain lazy
One source of the perception of apathy is that the opportunities to DO something are not effectively made public. If you don't spend all your time digging through obscure documents (or hire a lobbiest to do it FOR you), you don't know what the congress critters or regulators are up to.
Finding out on your own about legislation or rules that might affect you is SEVERAL full-time jobs. (Hell - they pass laws and impose regulations faster than a man can read, and the legislature doesn't even read most of what it votes on.) This stuff gets buried so deep (often deliberately) that even major corporations with entire DEPARTMENTS watching for things that affect them sometimes miss it. Responding to this crap is IMPORTANT. But people who have real work to do - and even those who don't - just can't spend enough time to hunt it down.
And that's where organizations like the EFF can come in.
I note that a number of Slashdot people (myself included) filed comments on DCMA rulemaking with the copyright office in response to a posting (within a day or so of the cutoff) that the opportunity existed.
When "action" consists of demonstrating that some people out there have an opinion, and the message can be sent by phone, email, or snail mail, the Slashdot Effect can be put to effective use. Combine that with a TIMELY article telling where to submit comments and we can flood them out.
I emphasize TIMELY because a day's notice will yeild a bunch of flames and a few well-thought-out letters, in a burst that makes it obvious they're all from a single community. Notice near the start of a comment period and a reminder a few days from the end will produce a large yeild of well-thought and well-researched comments.
This is not just speculation. This is exactly what the pro-gun movement does. The NRA (with membership second only to ARP) and its lobbying arm the ILA gets the press. But it is only one of a number of groups (GOA, Members' Councils, state orgs like CRPA, JPFO, etc.) which send out "legislative alerts" to their members when something is discovered in time to react. (They also interview and rate candidates come election time and make their voting records known.)
So instead of moaning about how apathetic the working nerds are (demotivating them further), perhaps EFF might put their staff to work to act as our early warning system, and SUBMIT SHORT ARTICLES to Slashdot when there are opportunities available for an avalanch of electronic or paper mail and phone calls. (Many of us have printers, right? We can pretty much ALL send a handwritten, hand-addressed letter, and those are even MORE effective.)
I'm sure the Slashdot editors would be happy to post such notices. They even have categories for them! Legislative and regulatory rulemaking on IP and other aspects of internet and programming freedom are PRIME examples of "News for Nerds, Stuff that Matters!". How could they resist a couple EFF staffers acting as investigative reporters - for free?
With EFF (and others) to dig up the info, Slashdot (and others) to spread the word in a TIMELY fashion and serve as dissusion and planning venues, and HORDES of internet-connected VOTERS to flood the critters with mail, we could turn a lot of this around.
But nothing will happen unless those hordes have the necessary INFORMATION to act.
And that's what the computer age is about, isn't it?
You'd think this would be an easy distinction to understand: We live in a country where 10 children are killed by hand guns every day. But Smith and Wesson doesn't worry that the FBI will come arrest them because someone used their technology to commit a crime.
The point is valid. But the statistic is bogus. You are echoing a big-lie propaganda sound bite as if it were a fact, when the true numbers are actually more than an order of magnitude lower.
Judges (and others, both in the general public and in policy-making offices) are becoming aware of the actual numbers. So continued use of this lie in arguments will tend to discredit the rest of your argument.
It is also a hotbutton issue, with the opposing view held by members of a subculture which constitutes about half the voting population. So by using a well-known big-lie from one side of this unrelated debate you risk instantly alienating half your audience.
Ten per day comes out to 3650 per (non-leap) year.
According to the National Safety Council, 230 children under the age of 15 were killed in firearms-related accidents in 1991. (That's the latest year for which figures are available at the time of the composition of the rec.guns FAQ found here.)
According to the most recent FBI Uniform Crime report (2000), available here, murder and negligent homicide of children ages 0 through 12 (by ALL means, not just firearms) totaled only 641. (The bulk of those were infants, implying most were killed by abuse or neglect rather than with-firearm murder.)
The numbers jump beginning with the 13 thorugh 16 age group, as gang-related murders begin to dominate the stats. But to get to "ten children killed by handugns per day" you need to consider "childhood" to extent into the mid-thirties.
So I recommend that, if you wish to continue using the analogy in future argumnets on the subject, you drop the bogus statistic. You can make the point even better using a real one ("robbers use handguns to kill two people per day"), or skipping the stats entirely, thus minimizing the risk of offending people who otherwise would be your allies.
This isn't a linux issue. It's a hardware issue.
The significant thing about 'big iron' is that it's an enabling hardware technology.
Once you have it you can write firmware and software that creates the illusion that the hardware never fails
Until you have it, you can't.
The hardware described looks about right - if they handled machine checks properly. (And the fact that they even used the term implies they either did or are trying.) Basic idea: The machine catches ANY error, with enough state saved that you can:
CORRECT the error
IDENTIFY any failed components,
MOVE tasks to non-failed components or reconfigure the failing components to limp along,
NOTIFY the OS of any problem, so it can do things like start moving things off a dying component, and
pick up the computation where it left off WITHOUT the error.
When you can do this you can write a modified Linux, Windows, BeOS, or what-have-you that can do the things a mainframe can. (But you'll need to have a REALLY reliable OS for your starting point - you're now talking uptimes measured in decades. The software better not take the system down in the absense of hardware trouble, and there IS NO hardware troube. B-) )
Hot-swappable parts are more a side-effect than something key. You have to be able to hot-swap to replace a broken part with the system live. Once you have the ability to hot-swap in a replacement for a failed part and add it back into a running domain, it's trivial to generalize that to "fix" parts that were "bad" because they had never been installed.
Partitioning is also implied: You need a minimum of two domains ("virtual machine" subsets of the total device) - working (where the live system is) and diagnostic (where the maintainence guys check out the parts). Once you have that mechanism, making a LARGE number of working domains (with varying amounts of resource, including full or time-shared CPUs) is straightforward.
The guy even works three blocks from the WhiteHouse.
The software is developed in a suburb of Seattle Washington (state) and the company's security chief works in Washington (DC), nearly as far from the software department as you can get and still be in the continental US.
THAT explains the security problems in Microsoft products!
B-)
I think it doesn't make any difference whether it is open source or closed source, it's a matter of identifying them once the product is released.
So...who cares if there are problems. We'll find them eventually - as soon as someone exploits them and we hear about it.
Precicely.
If you want bug-free code you need to start at the architecture/design process (avoiding bug-prone choices), then debug as you go. It's like growing a perfect crystal - you push the impurities out as it solidifies, so only the boundary needs attention. The longer you wait, the larger your search space for each bug, and the bigger the hive of ofspring each bug has produced as new code was added to buggy code.
Security issues are a special case of "bugs", with more than the typical amount of effort needed at precoding stages to avoid building unfixable problems into the basic architecture.
I wonder if they release their code like that for QA as well. It's a matter of identifying bugs once the product is released.
My impression is that Schmidt is completely unaware that software QA, or any other pre-release potential for (securyty) bug suppression, exists. At a minimum his statement implies that Security as a department doesn't participate in architecture, design, code reviews, or QA, and that its leader either feels no need to do so, or is deliberately directing attention away from an inability to affect those stages.
That the head of security for Microsoft could emit such an answer is appalling. But it also goes a long way toward explaining the security problems in Microsoft products.
Aren't [power conditioning devices] basically low pass filters for power? Will the networking signals survive going through them?
Depends on the device.
Surge-protectors won't touch it.
EMI filters will completely block it.
UPSes will probably block it, even in bypass mode. (They need EMI filters on their wiring due to the switching stuff inside.)