For the same reason it's not likely (but you never know) that there's any advantage to adding an off-chip L3 cache.
A price/performance advantage or a performance advantage? I would imagine that for a chip running at 1.1 GHz that access to a L3 cache running at 500 MHz would be far more perferable to access of some PC-266 Double Rate DRAM. The inhibiting factor becomes how much that 500 MHz DDR-DRAM is going to set you back, plus the complexity costs in you northbridge design, and that zingers that throws into cache coherenecy in multiple CPU designs.
Even the performance advantage disappears if the hit rate isn't pretty high, because you insert at least one cycle into your access latency while you check tags for a hit. I'm not familiar with the studies themselves but from second-hand sources it looks like there's precious little predictability left in the access stream once L1+L2 get done with it.
dpilot argued that the L3 cache on K6-3 systems adds measurably to performance; I certainly can't prove dpilot wrong. I do know that there's a pretty fierce tradeoff between complexity and speed, though, and sometimes you get more milage out of keeping it simple but screaming fast.
L1 cache is the memory closest to the actual computing functions. It runs at CPU speed, but because larger => slower it can't be very big; usually measured in Kbytes. It caches the most-used memory addresses.
L2 cache is larger and slower than L1. Until recently, L2 was implemented by separate RAM devices attached to the CPU. The original Pentium (socket 7) L2 cache was on the processor's front-side bus, between it and the system controller. This became a serious speed limiter and newer processors added a back-side bus strictly for cache (one reason that the CPU modules appeared.) Back-side bus cache runs around 400 MHz plus three or so bus cycles added latency. At 800 MHz this starts to get ugly.
Moving the L2 cache on-chip may not let it run much faster (typically CPU/2 or CPU/2.5) but it cuts the pipeline latency, and latency reduction is what cache is all about. Also, being on-chip makes it much less expensive to use wide busses so the L2 could, for instance, transfer an entire cache line to the L1 in a single cycle.
L1+L2 cache is so good at removing nonrandom accesses from the memory stream that appears on the front-side bus that what actually makes it to the DRAM is almost completely random-access. That's why packet-based memory (e.g., RAMBUS) do so poorly compared to their sustained bandwidth: the bandwidth is never sustained, it's just the first cycle that counts.
For the same reason it's not likely (but you never know) that there's any advantage to adding an off-chip L3 cache. The hit rate would be too low to be worth the trouble of checking for a hit.
Somehow I don't think His Honor intends to include Win2k under the prohibition. Which means that the injunction is Windows dressing. The MICROS~1 specific JVM will still be the de facto standard.
Also - @home has a strict AUP *against* security scans. They would be in violation of their own AUP to take action like what this guy has mentioned in the article. I was not able to locate their online AUP, but searching here or here should reveal it. If nothing else, I will scan it in and post it, as I still have the copy I signed.
@Home will probably just ban users from having static IP addresses[*], running servers, and running Linux (because it's potentially "dangerous").
They already do on at least two counts. You can't run servers of any kind (e.g., shared printers on your LAN) and you have to use their Special Modified Version of Infernal Exploder to access account info. Most @Home systems also don't use static IP.
WTF is this? You know how many people I know that setup @home service with an old Linux box to proxy off the cable connection to the rest of the house? I know about 4 people that this is seriously going to piss off. They don't spam and their systems are very secure, but it looks like they are going to be @Home's sacrificial lamb to the UDP.
@Home says they're going to enforce their AUP, and your friends complain because they're currently violating it by running Linux boxen? Didn't they read the AUP before signing up? @Home is very clear that servers of all kinds (file, print, telnet, you name it) are no-nos. Don't like it? Don't sign with @Home.
And yes, we walk the walk. We paid for the wiring all the way from the curb to the server and then they changed the AUP. Still on 56k.
Maybe the video card outfits should consider including a real modem on the high-end video adaptors. Save a slot, speed up network gaming, and add only minimally to the cost.
I'm curious about what fuel he plans to use: butane, propane, beer, what? These things are plastic coated aluminum, right? They don't burn so easily or so brightly. Maybe microwaving them would be more effective.
Ah, child, you're too young to remember old-fashioned flashbulbs. Nothing but aluminum wire in a glass&plastic envelope, and man! were those suckers bright. Nothing, and I mean nothing burns like aluminum in oxygen.
Since this is a geek forum, here's a little science experiment. Wear welding goggles and sunscreen (no, I'm not joking about the sunscreen.) Do it far from anything combustible, like yourself or your car.
Get a little bit of the usual magnesium tape (you can light Mg with a match) and use it to start a small piece of aluminum foil supported by something you don't care about (because it's gonna be ruined.) The glare will be visible from orbit.
Oh, year: wear white or something you don't mind getting bleached by the UV. And trust me on the sunscreen.
Chris, messages to abuse@home.net go to/dev/null. Including yours. If there were someone reading them the UDP wouldn't have been called in the first place!
As others have pointed out, there is a fundamental difference between horizontal and vertical integration. MICROS~1 is dangerous to end-users (I avoid the term 'consumers') because they have a lock on a complete horizontal layer. AOL-T/W won't be the only way to get content; we can always watch MGM movies over RCN pipes and get our news from C|NET.
If AOL-T/W cut off CNN to non-AOL subscribers they'll feel it far more than we will. If they deny AOL subscribers access to the New York Times they'll just lose subscribers. If they only allow Warner Brothers movies over their pipes then they'll lose lots of subscribers. I don't think Case is that stupid.
It would be nice if JK did a little homework before preaching. The watershed case for the antitrust law of vertical integration was the Brown Shoe case back in (IIRC) 1952. Brown, with only 5% market share, wanted to merge to create a vertically integrated firm. Its competitors objected, because the increased efficiency would have cut into their profits, and the USDOJ blocked the merger. The USSC ruled that integration which threatened competitors but benefited consumers was not against the law; that harm to the consumer was required.
AOL-T/W may be objectionable for a lot of reasons (for that matter AOL is objectionable all on its own) but it's certainly not an antitrust matter either de jure or practically. What it will do is put pressure on other information marketing firms to get their act together in the face of a new standard of efficiency -- which IMHO we should all applaud.
IBM's PC had a 16-bit processor, while CP/M was only an 8 bit OS, that was the problem.
Nice try, no cigar. CPM-86 had been around for several years at that point. I should know; I still have my original 8" CPM-86 disk complete with three-digit serial number.
Cute. Here we keep hearing about how maybe Linux is an OK server OS but Win* is the only choice of the desktop, and what happens? We get a big push towards using an NT server with Linux desktops!
Actually, we've been doing something similar to support legacy X apps (OK, X makes this much easier.) We have legacy servers sufficient for the workload, and the obsolete apps run on them transparently (the local jobs are just RJE scripts.) Given the gorking huge pile of legacy WinCode, this is something that corporate admins are going to need someday regardless.
The problem with your analysis is that you're comaparing mechanisms with systems. A zero-sum mechanism can be part of a non-zero-sum system; that's exactly the case in all of the examples, law included: The examples you cite are not zero-sum games; the goal of standards review in engineering, peer review in science, and open-source development is NOT to make "the other guy lose", but "to find the bugs and fix them to arrive at the truth".
In standards development/review, one party's proposals are accepted and another's rejected. Often at considerable cost (I'm talking about millions of $$) to the losing party.
In peer review, one party's theory becomes accepted and another's falls into disrepute, with corresponding spoils (e.g. tenure, status, grants) to the parties in question.
In open-source software, someone becomes a Code God and another gets Stoopid Points for off-by-one errors, with possible career consequences.
In law, as you point out, one party prevails and the other eats dirt.
Ah Ha! The perceptive sutdent points out, that's not the whole picture.
In standards work, the intention is that the chosen proposal has greater net utility than the rejected one, often because it suits the needs of more parties or because creates greater long-term functionality.
In science, the two theories are not of equal value; the system supposedly selects the prevailing theory because it's more accurate than the rejected one.
In open-source the resulting code is (as you note) supposedly more useful.
In law, the guilty are more likely to be punished and the innocent less so.
If you are laboring under the delusion that the goal of any court proceeding is "to get to the truth", I cannot help you.
I am, as you say, "laboring under the delusion" tha the goal of the entire legal system is Truth and Justice. Which is not to say that I'm naive enough to think that the system always works. Any more than that standards bodies never set stupid standards, that the prevailing scientific theories are always better than the ones laughed at, or that open-source software development always produces optimal software.
The fundamental oversight in your analysis is that these little head-butting contests have side effects (a phenomenon that should be familiar to techies.) Standards bodies not only have winners and losers, but influence the course of technology. Science not only has Stephen Hawking and Archimedes Plutonium but enables human advancement. Open-source software not only has the Gnome and KDE teams, but lands on millions of desktops. And the law not only has winners and losers but keeps Bob Vickers out of circulation.
Have you actually served on a jury lately? I have. And I can tell you that all of the main players there really were there for more than scoring in some pointless game, and the result really was the truth as nearly as we are given to find it.
In engineering, there are design goals that can be measured against and optimized. In science, there's an objective truth being sought by both sides of a debate. In open source, the quality of software speaks for itself; if you don't like a feature, you fork the code.
Those design goals aren't universal. Everybody brings their own agendas to the table, and There Can Be Only One (e.g., DDR and RamBus.)
You're giving science the benefit of being judged by it's noblest intentions, and law by its worst failures.
Forking the code is almost always suicidal, as has been pointed out ad nauseum. In practice, There Can Be Only One.
But in law, it's still trial by combat - the only way for one side to win is to make the other side lose. Primitive mammalian "us-vs-them" behavior at its finest.
You make it sound like we might as well flip a coin, or that the only thing that matters is the legal teams. Which, except in our more dramatic moments, we all know to be silly. It really does matter what the facts in the case are; I've seen defense lawyers work wonders trying to raise doubts and be totally overwhelmed by the evidence.
By the way, don't dis that head-butting. It works, too, because of side-effects. The winner, on average, fathers a healthier bunch of kids than the loser would. Which renders the game a net gain for the herd in the same way that you and I are the net beneficiaries of Clarence Darrow and William Jennings Bryan.
Don't knock adversary systems. After all, most standards bodies work that way, and work very well (speaking from personal experience.) Scientific publication works that way: publish something original in Science and there will instantly be a horde of PhDs trying to prove you wrong; if your idea survives it's likely good. An adversary system is still the best known way to weed out errors, and that's why (drum roll, please) open-source software development is so effective: it's another adversary system.
So, if you please, why are adversary proceedings good enough for science, standards, and open-source software but not for the mechanics of the larger societies that we live in?
From the standpoint of purely Darwinian social evolution, a zero-sum game with nontrivial costs just wouldn't last, and certainly not become as universal as it has.
Seen another way, there are plenty of examples of positive-sum law. Imagine what the cost to society would be of a total lack of traffic laws -- half the cars driving on the left, half on the right, some stopping on red, some on green, some drunk, etc.
We all recognize that conventions are necessary in technical applications (there's no real reason why we shouldn't run ftp on port 71 instead, or for that matter use different CRC polynomials) Law serves the same function in society that standards do in technical communities: they allow us to coexist, which has benefits greater than the costs they impose.
I telecommute from hotel rooms. Rooms rented by the Company, in fact. I also do lots of Company work on a Company laptop while flying on Company-paid air trips.
Seems like these are far more promising opportunities for OSHA: regulating business hotel accomodations, regulating laptop ergonomics, and regulating airline seating used for business travel. Just think of the fun they'll have with the FAA over seating standards.
I hate to say it, but RH seems to be blowing it on what's supposed to be their forte: service & support. I was having some problems with a clean install from a Cheapbytes CD and didn't want to mess around, so I bought a boxed copy of 6.1. That barfed on three different machines -- fatal crash of the install reporting file corruption. I went back to the Cheapbytes CD, found out what the problem with it was, and continued.
So maybe anyone can have a bad CD sneak out. I tried to report the problem to RH (more as a courtesy than anything) and found out I needed to register to do that. (You'd think that they'd accept bug reports from anyone, but no.) Then, despite having a paid-for boxed copy I couldn't register. The web submission barfs.
Looks like Young needs to do a reality check if he wants to keep those billions.
The usual industry solution to muddled claims of "trade secrets" (and copyright, BTW) is to run a clean-room exercise. Now, as I understand it, CSS is so insecure that a brute-force crack based entirely on published (as in indisputably legit hardcopy) data would only take hours to days on modest hardware, and once one key is available all keys are readily obtained.
Kewl.
What say someone undertakes exactly that? Work up the brute-force crack code in a clean room, then be prepared to actually crack the sucker right in open court or at least before witnesses. Publish the keyfinding code right in the Court's own transcript. Let 'em censor that!
Newer processors have increasingly smaller distances between the pathways on the surface of the chip. Since this distance is smaller, radiation can cause the electrical charge to "jump" across.
Not quite. Ionizing radiation induces carrier pairs in the silicon, possibly causing conduction when it shouldn't happen. Charged particles (esp. alpha) actually deposit charge where it can cause trouble. The most susceptible circuits are memories; dynamic memories actually store data in charge states, and static memories are intentionally very low-energy. Newer processes use smaller features and lower supply voltages, which reduces the charge needed to change state.
Unfortunately, for speed/power reasons all leading-edge processors use dynamic logic (similar to dynamic memory, except that the charges can be even smaller since the storage time is so much less.)
There are ways to minimize these effects with shielding,
Shielding is actually worse than useless, since cosmic rays are too penetrating to block effectively, but they do kick out secondary radiation from the shielding.
I'd love to know whether they're using the original Intel 486 or one of the newer 486 replacements that use static logic (which is handy in really-low-power systems since it lets you start and stop the clock at any time.)
Funny how history repeats itself. William Marshall was the chief military advisor to Henry II, as remarkable a king as FDR was a President. (It's after him that the term "Knight Marshall" comes to us.) Like his distant namesake, he was widely reknowned for absolute integrity.
We need to keep in mind that tax software is the ultimate antithesis to operating software. (The Magic Cauldron for details.) The requirements change -- dramatically -- every year. Unlike operating software, there's very little accumulated value over time. This means that in effect the only thing you're buying with tax software is a service, and the s/w is just the delivery vehicle for that service.
Slashdot Mirror
For the same reason it's not likely (but you never know) that there's any advantage to adding an off-chip L3 cache.
A price/performance advantage or a performance advantage? I would imagine that for a chip running at 1.1 GHz that access to a L3 cache running at 500 MHz would be far more perferable to access of some PC-266 Double Rate DRAM. The inhibiting factor becomes how much that 500 MHz DDR-DRAM is going to set you back, plus the complexity costs in you northbridge design, and that zingers that throws into cache coherenecy in multiple CPU designs.
Even the performance advantage disappears if the hit rate isn't pretty high, because you insert at least one cycle into your access latency while you check tags for a hit. I'm not familiar with the studies themselves but from second-hand sources it looks like there's precious little predictability left in the access stream once L1+L2 get done with it.
dpilot argued that the L3 cache on K6-3 systems adds measurably to performance; I certainly can't prove dpilot wrong. I do know that there's a pretty fierce tradeoff between complexity and speed, though, and sometimes you get more milage out of keeping it simple but screaming fast.
L1 cache is the memory closest to the actual computing functions. It runs at CPU speed, but because larger => slower it can't be very big; usually measured in Kbytes. It caches the most-used memory addresses.
L2 cache is larger and slower than L1. Until recently, L2 was implemented by separate RAM devices attached to the CPU. The original Pentium (socket 7) L2 cache was on the processor's front-side bus, between it and the system controller. This became a serious speed limiter and newer processors added a back-side bus strictly for cache (one reason that the CPU modules appeared.) Back-side bus cache runs around 400 MHz plus three or so bus cycles added latency. At 800 MHz this starts to get ugly.
Moving the L2 cache on-chip may not let it run much faster (typically CPU/2 or CPU/2.5) but it cuts the pipeline latency, and latency reduction is what cache is all about. Also, being on-chip makes it much less expensive to use wide busses so the L2 could, for instance, transfer an entire cache line to the L1 in a single cycle.
L1+L2 cache is so good at removing nonrandom accesses from the memory stream that appears on the front-side bus that what actually makes it to the DRAM is almost completely random-access. That's why packet-based memory (e.g., RAMBUS) do so poorly compared to their sustained bandwidth: the bandwidth is never sustained, it's just the first cycle that counts.
For the same reason it's not likely (but you never know) that there's any advantage to adding an off-chip L3 cache. The hit rate would be too low to be worth the trouble of checking for a hit.
Somehow I don't think His Honor intends to include Win2k under the prohibition. Which means that the injunction is Windows dressing. The MICROS~1 specific JVM will still be the de facto standard.
... with the obvious potted-pork smell.
Maybe.
you only need the @home pluggin to access your user information.
IOW, you do have to be using "approved" software. Or is this plugin available for Linux?
The server issue varies by agreements(AUP) with the local cable operator.
Not according to the @Home AUP
Nope, thermite if for graduate-level pyros. Plain ol' aluminum foil is for PYR201-level stuff.
As for lighting Mg with a match, all I can say is "been there, done that."
Also - @home has a strict AUP *against* security scans. They would be in violation of their own AUP to take action like what this guy has mentioned in the article. I was not able to locate their online AUP, but searching here or here should reveal it. If nothing else, I will scan it in and post it, as I still have the copy I signed.
The @Home AUP
Pay particular attention to the all-inclusive ban on "servers," broadly defined.
@Home will probably just ban users from having static IP addresses[*], running servers, and running Linux (because it's potentially "dangerous").
They already do on at least two counts. You can't run servers of any kind (e.g., shared printers on your LAN) and you have to use their Special Modified Version of Infernal Exploder to access account info. Most @Home systems also don't use static IP.
WTF is this? You know how many people I know that setup @home service with an old Linux box to proxy off the cable connection to the rest of the house? I know about 4 people that this is seriously going to piss off. They don't spam and their systems are very secure, but it looks like they are going to be @Home's sacrificial lamb to the UDP.
@Home says they're going to enforce their AUP, and your friends complain because they're currently violating it by running Linux boxen? Didn't they read the AUP before signing up? @Home is very clear that servers of all kinds (file, print, telnet, you name it) are no-nos. Don't like it? Don't sign with @Home.
And yes, we walk the walk. We paid for the wiring all the way from the curb to the server and then they changed the AUP. Still on 56k.
Maybe the video card outfits should consider including a real modem on the high-end video adaptors. Save a slot, speed up network gaming, and add only minimally to the cost.
I'm curious about what fuel he plans to use: butane, propane, beer, what? These things are plastic coated aluminum, right? They don't burn so easily or so brightly. Maybe microwaving them would be more effective.
Ah, child, you're too young to remember old-fashioned flashbulbs. Nothing but aluminum wire in a glass&plastic envelope, and man! were those suckers bright. Nothing, and I mean nothing burns like aluminum in oxygen.
Since this is a geek forum, here's a little science experiment. Wear welding goggles and sunscreen (no, I'm not joking about the sunscreen.) Do it far from anything combustible, like yourself or your car.
Get a little bit of the usual magnesium tape (you can light Mg with a match) and use it to start a small piece of aluminum foil supported by something you don't care about (because it's gonna be ruined.) The glare will be visible from orbit.
Oh, year: wear white or something you don't mind getting bleached by the UV. And trust me on the sunscreen.
Chris, messages to abuse@home.net go to /dev/null. Including yours. If there were someone reading them the UDP wouldn't have been called in the first place!
As others have pointed out, there is a fundamental difference between horizontal and vertical integration. MICROS~1 is dangerous to end-users (I avoid the term 'consumers') because they have a lock on a complete horizontal layer. AOL-T/W won't be the only way to get content; we can always watch MGM movies over RCN pipes and get our news from C|NET.
If AOL-T/W cut off CNN to non-AOL subscribers they'll feel it far more than we will. If they deny AOL subscribers access to the New York Times they'll just lose subscribers. If they only allow Warner Brothers movies over their pipes then they'll lose lots of subscribers. I don't think Case is that stupid.
It would be nice if JK did a little homework before preaching. The watershed case for the antitrust law of vertical integration was the Brown Shoe case back in (IIRC) 1952. Brown, with only 5% market share, wanted to merge to create a vertically integrated firm. Its competitors objected, because the increased efficiency would have cut into their profits, and the USDOJ blocked the merger. The USSC ruled that integration which threatened competitors but benefited consumers was not against the law; that harm to the consumer was required.
AOL-T/W may be objectionable for a lot of reasons (for that matter AOL is objectionable all on its own) but it's certainly not an antitrust matter either de jure or practically. What it will do is put pressure on other information marketing firms to get their act together in the face of a new standard of efficiency -- which IMHO we should all applaud.
IBM's PC had a 16-bit processor, while CP/M was only an 8 bit OS, that was the problem.
Nice try, no cigar. CPM-86 had been around for several years at that point. I should know; I still have my original 8" CPM-86 disk complete with three-digit serial number.
Cute. Here we keep hearing about how maybe Linux is an OK server OS but Win* is the only choice of the desktop, and what happens? We get a big push towards using an NT server with Linux desktops!
Actually, we've been doing something similar to support legacy X apps (OK, X makes this much easier.) We have legacy servers sufficient for the workload, and the obsolete apps run on them transparently (the local jobs are just RJE scripts.) Given the gorking huge pile of legacy WinCode, this is something that corporate admins are going to need someday regardless.
The examples you cite are not zero-sum games; the goal of standards review in engineering, peer review in science, and open-source development is NOT to make "the other guy lose", but "to find the bugs and fix them to arrive at the truth".
Ah Ha! The perceptive sutdent points out, that's not the whole picture.
If you are laboring under the delusion that the goal of any court proceeding is "to get to the truth", I cannot help you.
I am, as you say, "laboring under the delusion" tha the goal of the entire legal system is Truth and Justice. Which is not to say that I'm naive enough to think that the system always works. Any more than that standards bodies never set stupid standards, that the prevailing scientific theories are always better than the ones laughed at, or that open-source software development always produces optimal software.
The fundamental oversight in your analysis is that these little head-butting contests have side effects (a phenomenon that should be familiar to techies.) Standards bodies not only have winners and losers, but influence the course of technology. Science not only has Stephen Hawking and Archimedes Plutonium but enables human advancement. Open-source software not only has the Gnome and KDE teams, but lands on millions of desktops. And the law not only has winners and losers but keeps Bob Vickers out of circulation.
Have you actually served on a jury lately? I have. And I can tell you that all of the main players there really were there for more than scoring in some pointless game, and the result really was the truth as nearly as we are given to find it.
In engineering, there are design goals that can be measured against and optimized. In science, there's an objective truth being sought by both sides of a debate. In open source, the quality of software speaks for itself; if you don't like a feature, you fork the code.
But in law, it's still trial by combat - the only way for one side to win is to make the other side lose. Primitive mammalian "us-vs-them" behavior at its finest.
You make it sound like we might as well flip a coin, or that the only thing that matters is the legal teams. Which, except in our more dramatic moments, we all know to be silly. It really does matter what the facts in the case are; I've seen defense lawyers work wonders trying to raise doubts and be totally overwhelmed by the evidence.
By the way, don't dis that head-butting. It works, too, because of side-effects. The winner, on average, fathers a healthier bunch of kids than the loser would. Which renders the game a net gain for the herd in the same way that you and I are the net beneficiaries of Clarence Darrow and William Jennings Bryan.
Don't knock adversary systems. After all, most standards bodies work that way, and work very well (speaking from personal experience.) Scientific publication works that way: publish something original in Science and there will instantly be a horde of PhDs trying to prove you wrong; if your idea survives it's likely good. An adversary system is still the best known way to weed out errors, and that's why (drum roll, please) open-source software development is so effective: it's another adversary system.
So, if you please, why are adversary proceedings good enough for science, standards, and open-source software but not for the mechanics of the larger societies that we live in?
Nope. Don't buy it.
From the standpoint of purely Darwinian social evolution, a zero-sum game with nontrivial costs just wouldn't last, and certainly not become as universal as it has.
Seen another way, there are plenty of examples of positive-sum law. Imagine what the cost to society would be of a total lack of traffic laws -- half the cars driving on the left, half on the right, some stopping on red, some on green, some drunk, etc.
We all recognize that conventions are necessary in technical applications (there's no real reason why we shouldn't run ftp on port 71 instead, or for that matter use different CRC polynomials) Law serves the same function in society that standards do in technical communities: they allow us to coexist, which has benefits greater than the costs they impose.
I telecommute from hotel rooms. Rooms rented by the Company, in fact. I also do lots of Company work on a Company laptop while flying on Company-paid air trips.
Seems like these are far more promising opportunities for OSHA: regulating business hotel accomodations, regulating laptop ergonomics, and regulating airline seating used for business travel. Just think of the fun they'll have with the FAA over seating standards.
I hate to say it, but RH seems to be blowing it on what's supposed to be their forte: service & support. I was having some problems with a clean install from a Cheapbytes CD and didn't want to mess around, so I bought a boxed copy of 6.1. That barfed on three different machines -- fatal crash of the install reporting file corruption. I went back to the Cheapbytes CD, found out what the problem with it was, and continued.
So maybe anyone can have a bad CD sneak out. I tried to report the problem to RH (more as a courtesy than anything) and found out I needed to register to do that. (You'd think that they'd accept bug reports from anyone, but no.) Then, despite having a paid-for boxed copy I couldn't register. The web submission barfs.
Looks like Young needs to do a reality check if he wants to keep those billions.
The usual industry solution to muddled claims of "trade secrets" (and copyright, BTW) is to run a clean-room exercise. Now, as I understand it, CSS is so insecure that a brute-force crack based entirely on published (as in indisputably legit hardcopy) data would only take hours to days on modest hardware, and once one key is available all keys are readily obtained.
Kewl.
What say someone undertakes exactly that? Work up the brute-force crack code in a clean room, then be prepared to actually crack the sucker right in open court or at least before witnesses. Publish the keyfinding code right in the Court's own transcript. Let 'em censor that!
Newer processors have increasingly smaller distances between the pathways on the surface of the chip. Since this distance is smaller, radiation can cause the electrical charge to "jump" across.
Not quite. Ionizing radiation induces carrier pairs in the silicon, possibly causing conduction when it shouldn't happen. Charged particles (esp. alpha) actually deposit charge where it can cause trouble. The most susceptible circuits are memories; dynamic memories actually store data in charge states, and static memories are intentionally very low-energy. Newer processes use smaller features and lower supply voltages, which reduces the charge needed to change state.
Unfortunately, for speed/power reasons all leading-edge processors use dynamic logic (similar to dynamic memory, except that the charges can be even smaller since the storage time is so much less.)
There are ways to minimize these effects with shielding,
Shielding is actually worse than useless, since cosmic rays are too penetrating to block effectively, but they do kick out secondary radiation from the shielding.
I'd love to know whether they're using the original Intel 486 or one of the newer 486 replacements that use static logic (which is handy in really-low-power systems since it lets you start and stop the clock at any time.)
Funny how history repeats itself. William Marshall was the chief military advisor to Henry II, as remarkable a king as FDR was a President. (It's after him that the term "Knight Marshall" comes to us.) Like his distant namesake, he was widely reknowned for absolute integrity.
Not only misquoted, but misattributed:
They that can give up essential liberty to obtain a little temporary safety, deserve neither liberty nor safety.
- Benjamin Franklin, Historical Review of Pennsylvania (1759)
We need to keep in mind that tax software is the ultimate antithesis to operating software. (The Magic Cauldron for details.) The requirements change -- dramatically -- every year. Unlike operating software, there's very little accumulated value over time. This means that in effect the only thing you're buying with tax software is a service, and the s/w is just the delivery vehicle for that service.