I've used Kirk McKusick's SoftUpdates for *BSD and been very impressed. Pulled the plug on four kernel compiles near the end. In three of the four cases, `make` just picked up the compile losing ~45 seconds. In the fourth, a `make clean` was necessary. In _all_ cases the fsck on reboot was minor. I've only lost power once in Linux during a kernel compile. I had to reinstall. It was too far gone for e2fsck.
IMHO, SoftUpdates are better than Journalled File Systems. There's no journal file to maintain, just careful ordering of the writes. Why no discussion of it for Linux?
I certainly agree that there is no reason or basis that software publishers should be allowed to grant themselves
immunity from direct or consequential defective product liability. It isn't as if they allow their customers to
see the defects [sharp edges] or otherwise judge their product. Nor should they escape wilful negligence
and treble punative damages when they refuse to fix known bugs.
But I am a bit afraid of too much change too quickly. I'd like to offer the SPA an olive branch and a carrot:
Allow publishers to limit liability _iff_ they make source [& the means to compile it] available people who
own a legal copy. Plus disclaim "deriviate work" on the resulting patches so they could be shared openly.
Then at least users could fix things and to some extent be masters of their own fate. The software wouldn't
need to be GPL because the copyright owner could restrict copying, but the patches would be GPL.
This is what should be in UCITA iso the onesided theft currently written.
What about power / heat ?
on
Inside the Itanium
·
· Score: 3, Informative
A nice architectural overview, but there's no mention of power. IIRC, Itanium sucks back 125 Watts!
Power/heat this high gives system designers problems, plus it can't be easy getting ~100 Amps to&from a chip.
Otherwise, AFAIK, Linux has working ia64 so code size can be compared. I'd expect 4x x86.
Much as I like this sort of law, I wonder if it will pass Constitutional tests sure to be launched by the various industry groups. It could be seen as affecting Interstate Commerce, which would be in violation of the Interstate commerce clause.
IANAL, but a service contract is different unless it specifies a given expiry time. Either party is free to withdraw from the contract at any time, giving whatever notice is required.
This is how ISPs, credit card companies, banks, etc. can get away with modifying terms. They are giving you a new service every month, so can ask for a new contract. Old service they cannot change.
Contracts where it is important that service continue have fixed expiries and/or evergreen provisions requiring long notice. Did you put any such terms in your ISP contract? No? Then your ISP probably put terms in favorable to it.
Well, at least they're not claiming they can beat Shannon's limit. 10 Gbit/s doesn't sound too unreasonable for coax. If you can't drive the frequency higher than say 500 MHz, you could still encode a constellation of 20 bits per transition.
This is similar to what modems do. AFAIK, they still don't run any faster than 3750 baud (Hz),
but they can encode up to 15 bits per wave to get 56kbit/sec. If the line isn't so quiet, they cannot distinguish all 15 bits, so the modems have to negotiate a constellation with fewer bits.
My question is how this will work with an ethernet-like collison detection system that AFAIK cable modems use. The jam signals could get ugly, and I'm not sure you can carry as my info on broadband as baseband systems. Or how cable decoders will cope.
I've tried out some Gigabit and it's not bad, but not easy. More for routers & backbones than desktops. Maybe for servers if they can shovel more than 10-40 Mbyte/s.
In my tests with `ttcp`, the best I can get is around 32 MByte/s between two PCs plugged together.
To get higher, you need PCI 64/66. The normal PCI bus can carry 4bytes*33 MHz = 133 MB/s (1066 Mb/s) but only during bursts. There is significant setup time, and the bursts are fairly short. Maybe I could get better throughput if I tweak the PCI registers, but I risk starving some other device.
The media jump on every Court case that will sell air/print -- irrespective of whether there are cameras in the courtroom. Please point to the specific harm. Sequestering a jury is less harmful to society than limiting the freedom of the press, especially about an important operation of government [justice]. Keeping secrets is harmful. Everyone else in the courtroom is a professional who should perform better under scrutiny, not worse.
Your "tearing, mangling and persuing mundane details" doesn't just describe the media, it also describes the Court process. Have you ever sat through a real court case? "The wheels of justice may grind slowly, but they grind exceedingly fine." In this sense, the media is accurite.
As for the OJ criminal trial, I don't think the presence of cameras influenced the case one bit. It sure did influence the public, though. Calling the outcome a "travesty" betrays a vengeful motive, or at least a misunderstanding of "the true goal of the proceedings" of criminal justice. It is not to punish wrongdoers, but to maintain order by punishing wrongdoers. Subtile but big difference. Do you know why it is better that 100 guilty walk free than 1 innocent be convicted? Hint -- it has to do with freedom, and IMHO has resulted in Anglo-Saxon countries enjoying greater economic development than those without this legal tradition. Or doesn't a impeached, perjuring, racist star prosecution witness count towards "reasonable doubt"?
As for Bill Clinton, his trial was by the Senate with no cameras present. How did cameras or the media circus win him leniency? [Where is that transcript?] They exposed most of his wrongdoing.
I will answer your question: Yes, coverage of the MS proceedings is vital. On the principle of open courts alone. Sure, some people may be mislead by the media. Better that than rumours and innuendo. Fortunately, we will get coverage. It's only the depositions that are private, probably as they should be until admitted into evidence.
Fully agreed that it will not make it to Court. However, biz executives have become remarkably short-term thinkers over the past decade:) They might well decide to cut off their noses to spite their face. They may not have much choice, since movies have investors who don't are about the MPAA, all they want is profit on this project.
As for the 2nd Am, I'm not sure it hasn't been invoked in Court cases. IMHO, "well regulated" means what it did at the time of writing -- well adjusted sights (sighted-in/zeroed in modern parlance). The alternative meaning of "subjected to rules and regulations" would need some proving to convince me. But others might be more easily convinced.
This decision was not rendered by any Federal or State Court, but by an Arbitration Panel. We do not know the composition of that panel, nor what (MPAA?) rules it was attempting to enforce.
I would hope that if this came before a Court, they would dismiss the complaint. Parody is protected expression, and will necessarily involve mimicing sight-gags. But that may be meaningless if the MPAA has effective control over trailer distribution. OTOH, such monopoly control could be the basis for suit. IANAL.
`grep key/proc/interrupts` already gives a nice count under Linux. Remember that all keys generate one interrupt when pressed, and one when released. IIRC, some generate more than one int per press (extended keycodes?), and there may be an issue with debouncing.
This was a case of MAI employees leaving to join Peak to service MAI computers. It was also a case of undisputed licence (machines & software acquired by third parties) not a shrink- or click-wrap.
IANAL, but this case seems to have been very poorly defended by a defendant that did plenty of wrong things & the courts wanted to punish. Some more obvious defenses that were apparently not used:
unreasonable restraint of trade to prohibit competition in servicing MAI computers,
agency -- Peak was doing the copying to RAM at third-party specific request, just like a 3rdP employee.
Peak wan't party to licence with third-party for the machine under service, so 17 USC 117 governs, not licence. 3rd parties should have been persued for licence violation.
Just because all wrongdoers want to {shred docs, use crypto, own guns,} does NOT mean all {doc shredders, crypto users, gunowners,} are wrongdoers. This is a logical fallacy. If all A are B, it does not follow that all B are A.
In a society that presumes innocence, you shouldn't even need to provide counter-examples. But here goes: docs are expensive to store and extremely expensive to retrieve and review even if totally innocent or even exculpatory. Docs are often requested for a fishing expedition unrelated to the complaint. Docs can be misread and misused by attorneys who are by definition extremely partisan.
Launching a satellite from Alaska? Why? There's free delta-V available from the Earth's rotation if you launch from further south. You don't need to be on the equator.
Is it on some really wierd polar orbit where you don't want eastward velocity? They'd better really map out the space junk then, because most of it will be coming at them fast.
I don't mind devil's advocates because I believe all positions should be discussed thoroughly.
In this case, the "adequate return" argument is easily defeated: Businesses use an Internal Rate of Return (IRR), External Rate of Return (ERR) or some other form of discounted cash-flow analysis (Net Present Value) to decide whether an investment is worth making. Typically, the interest rate used (hurdle rate) is 10-20%. With this interest rate, any revenues in the out years (say 20+) get discounted to insignificance.
The result is the length of time does not impact the decisions to invest [create] using modern methods of financial analysis.
Current ATX tower case designs are fairly good at minimizing complaints. The only area of obvious "fat" is the thickness which in many cases has grown excessive (8+ inches) without any discernable benefit except estetics and tower stability (on carpet, without base).
If you want smaller, you'll have to give something up. Like you, the last thing I want to give up is standardized components. I like reliability, and don't want some of the bare-bones small systems I've seen sold.
I made a simple layout with a std ATX PSU over & blocking the PCI slots (horrors!) but leaving one and the AGP slot clear. 3.5" drive in front, CDROM/DVD over the DIMMs & the CPU clear. About 12 x 12 x 5 inches. Clamshell case for better accessibility.
Of course `rm -R *` isn't enough -- it just unlinks files, but doesn't delete datablocks. To delete datablocks, try the -P option which overwrites the file data before unlinking. Unfortunately, this option is not available on GNU `rm` which is used on most Linux systems.
On behalf of the research team, EFF then filed a lawsuit seeking a clear determination that publication and presentation of this and other related research is speech protected under the US Constitution both at this conference and at other conferences in the future.
IANAL, but I don't think you can get such a determination from any US Court. For better or worse, they only deal in suits with full adversarial participation.
Had the EFF sued for defamation of character [by the RIAA alleging that Felton was a lawbreaker] or some other tort then the case probably couldn't be so easily dismissed. But nor would it be Federal, either.
AFAIK, The BIG pox was called the Great Pox and it is now called syphilis. This disease has mutated considerably over the centuries, and now has much less acute/obvious symptoms than in the past.
It was/is a very serious disease, and has confusing symptoms in many body systems. IIRC Dr Lister said: "he who knows the Great Pox knows medecine."
This really _doesn't_ look good for Cisco, the self-styled networking solutions company.
If I were the CIO of DoubleClick.net, CNN, MSNBC, or another top-10 site would I want to entrust my operations to Cisco? If their hefty price premium doesn't buy outstanding quality, what does it buy?
Last I checked, the Fed had issued >350 G$ in folding paper notes. Thats ~$1200 for every US resident. More than half of it is in hundreds. To be sure, alot of it is overseas. About 1/3 if a comparison with Canada holds. The drug trade and other illegal activity also probably eats up a fair bit, although I'd imagine they launder it {convert to bank deposits] fairly quickly. The question remains, and I haven't found any good answer in 10 years of looking -- who's holding [hoarding] the cash? $800/capita is more than I can believe. Around $100 would strike me as a reasonable number, even after taking all the till cash into account. Where is the rest? 200,000 people with 1 M$ each? 2M with 100k$? I've done anayses over time and in different OECD countries. They're not much different.
Is this exploit self-patching?
on
More On Tragedy
·
· Score: 1
I wonder what's going through the minds or airline pilots and their security chiefs in the aftermath of the WTC collapses.
Are they going to allow their planes to be seized by hijackers [cooperate for ransom], or are they simply going to put the helm over into a dive rather than surrender? Or try aggressive flying (barrel rolls) before that?
The whole matter is very macabre, but needs to be thought through. IMHO, the total tower pancake collapses was an unexpected payoff to terror. What are the odds? Is it better to deliberately crash than possibly be used as a kinetic/incendary projectile?
Slashdot Mirror
IMHO, SoftUpdates are better than Journalled File Systems. There's no journal file to maintain, just careful ordering of the writes. Why no discussion of it for Linux?
immunity from direct or consequential defective product liability. It isn't as if they allow their customers to
see the defects [sharp edges] or otherwise judge their product. Nor should they escape wilful negligence
and treble punative damages when they refuse to fix known bugs.
But I am a bit afraid of too much change too quickly. I'd like to offer the SPA an olive branch and a carrot:
Allow publishers to limit liability _iff_ they make source [& the means to compile it] available people who
own a legal copy. Plus disclaim "deriviate work" on the resulting patches so they could be shared openly.
Then at least users could fix things and to some extent be masters of their own fate. The software wouldn't
need to be GPL because the copyright owner could restrict copying, but the patches would be GPL.
This is what should be in UCITA iso the onesided theft currently written.
Power/heat this high gives system designers problems, plus it can't be easy getting ~100 Amps to&from a chip.
Otherwise, AFAIK, Linux has working ia64 so code size can be compared. I'd expect 4x x86.
Much as I like this sort of law, I wonder if it will pass Constitutional tests sure to be launched by the various industry groups. It could be seen as affecting Interstate Commerce, which would be in violation of the Interstate commerce clause.
This is how ISPs, credit card companies, banks, etc. can get away with modifying terms. They are giving you a new service every month, so can ask for a new contract. Old service they cannot change.
Contracts where it is important that service continue have fixed expiries and/or evergreen provisions requiring long notice. Did you put any such terms in your ISP contract? No? Then your ISP probably put terms in favorable to it.
Absolutely right! I meant to say "symbols", not "bits" per constellation or transition! Serious brainfade.
This is similar to what modems do. AFAIK, they still don't run any faster than 3750 baud (Hz),
but they can encode up to 15 bits per wave to get 56kbit/sec. If the line isn't so quiet, they cannot distinguish all 15 bits, so the modems have to negotiate a constellation with fewer bits.
My question is how this will work with an ethernet-like collison detection system that AFAIK cable modems use. The jam signals could get ugly, and I'm not sure you can carry as my info on broadband as baseband systems. Or how cable decoders will cope.
In my tests with `ttcp`, the best I can get is around 32 MByte/s between two PCs plugged together.
To get higher, you need PCI 64/66. The normal PCI bus can carry 4bytes*33 MHz = 133 MB/s (1066 Mb/s) but only during bursts. There is significant setup time, and the bursts are fairly short. Maybe I could get better throughput if I tweak the PCI registers, but I risk starving some other device.
Your "tearing, mangling and persuing mundane details" doesn't just describe the media, it also describes the Court process. Have you ever sat through a real court case? "The wheels of justice may grind slowly, but they grind exceedingly fine." In this sense, the media is accurite.
As for the OJ criminal trial, I don't think the presence of cameras influenced the case one bit. It sure did influence the public, though. Calling the outcome a "travesty" betrays a vengeful motive, or at least a misunderstanding of "the true goal of the proceedings" of criminal justice. It is not to punish wrongdoers, but to maintain order by punishing wrongdoers. Subtile but big difference. Do you know why it is better that 100 guilty walk free than 1 innocent be convicted? Hint -- it has to do with freedom, and IMHO has resulted in Anglo-Saxon countries enjoying greater economic development than those without this legal tradition. Or doesn't a impeached, perjuring, racist star prosecution witness count towards "reasonable doubt"?
As for Bill Clinton, his trial was by the Senate with no cameras present. How did cameras or the media circus win him leniency? [Where is that transcript?] They exposed most of his wrongdoing.
I will answer your question: Yes, coverage of the MS proceedings is vital. On the principle of open courts alone. Sure, some people may be mislead by the media. Better that than rumours and innuendo. Fortunately, we will get coverage. It's only the depositions that are private, probably as they should be until admitted into evidence.
As for the 2nd Am, I'm not sure it hasn't been invoked in Court cases. IMHO, "well regulated" means what it did at the time of writing -- well adjusted sights (sighted-in/zeroed in modern parlance). The alternative meaning of "subjected to rules and regulations" would need some proving to convince me. But others might be more easily convinced.
I would hope that if this came before a Court, they would dismiss the complaint. Parody is protected expression, and will necessarily involve mimicing sight-gags. But that may be meaningless if the MPAA has effective control over trailer distribution. OTOH, such monopoly control could be the basis for suit. IANAL.
`grep key /proc/interrupts` already gives a nice count under Linux. Remember that all keys generate one interrupt when pressed, and one when released. IIRC, some generate more than one int per press (extended keycodes?), and there may be an issue with debouncing.
This was a case of MAI employees leaving to join Peak to service MAI computers. It was also a case of undisputed licence (machines & software acquired by third parties) not a shrink- or click-wrap.
IANAL, but this case seems to have been very poorly defended by a defendant that did plenty of wrong things & the courts wanted to punish. Some more obvious defenses that were apparently not used:
unreasonable restraint of trade to prohibit competition in servicing MAI computers,
agency -- Peak was doing the copying to RAM at third-party specific request, just like a 3rdP employee.
Peak wan't party to licence with third-party for the machine under service, so 17 USC 117 governs, not licence. 3rd parties should have been persued for licence violation.
Bad facts make bad decisions make bad law.
In a society that presumes innocence, you shouldn't even need to provide counter-examples. But here goes: docs are expensive to store and extremely expensive to retrieve and review even if totally innocent or even exculpatory. Docs are often requested for a fishing expedition unrelated to the complaint. Docs can be misread and misused by attorneys who are by definition extremely partisan.
Is it on some really wierd polar orbit where you don't want eastward velocity? They'd better really map out the space junk then, because most of it will be coming at them fast.
In this case, the "adequate return" argument is easily defeated: Businesses use an Internal Rate of Return (IRR), External Rate of Return (ERR) or some other form of discounted cash-flow analysis (Net Present Value) to decide whether an investment is worth making. Typically, the interest rate used (hurdle rate) is 10-20%. With this interest rate, any revenues in the out years (say 20+) get discounted to insignificance.
The result is the length of time does not impact the decisions to invest [create] using modern methods of financial analysis.
If you want smaller, you'll have to give something up. Like you, the last thing I want to give up is standardized components. I like reliability, and don't want some of the bare-bones small systems I've seen sold.
I made a simple layout with a std ATX PSU over & blocking the PCI slots (horrors!) but leaving one and the AGP slot clear. 3.5" drive in front, CDROM/DVD over the DIMMs & the CPU clear. About 12 x 12 x 5 inches. Clamshell case for better accessibility.
Of course `rm -R *` isn't enough -- it just unlinks files, but doesn't delete datablocks. To delete datablocks, try the -P option which overwrites the file data before unlinking. Unfortunately, this option is not available on GNU `rm` which is used on most Linux systems.
IANAL, but I don't think you can get such a determination from any US Court. For better or worse, they only deal in suits with full adversarial participation.
Had the EFF sued for defamation of character [by the RIAA alleging that Felton was a lawbreaker] or some other tort then the case probably couldn't be so easily dismissed. But nor would it be Federal, either.
It was/is a very serious disease, and has confusing symptoms in many body systems. IIRC Dr Lister said: "he who knows the Great Pox knows medecine."
First they ignore you
then they laugh at you
then they fight you
then you win.
Microsoft has clearly stepped through to the fighting.
If I were the CIO of DoubleClick.net, CNN, MSNBC, or another top-10 site would I want to entrust my operations to Cisco? If their hefty price premium doesn't buy outstanding quality, what does it buy?
Makes the computer run slow
unexplained disk activity
makes files disappear randomly
causes machine lockups
Last I checked, the Fed had issued >350 G$ in folding paper notes. Thats ~$1200 for every US resident. More than half of it is in hundreds.
To be sure, alot of it is overseas. About 1/3 if a comparison with Canada holds. The drug trade and other illegal activity also probably eats up a fair bit, although I'd imagine they launder it {convert to bank deposits] fairly quickly.
The question remains, and I haven't found any good answer in 10 years of looking -- who's holding [hoarding] the cash? $800/capita is more than I can believe. Around $100 would strike me as a reasonable number, even after taking all the till cash into account. Where is the rest? 200,000 people with 1 M$ each? 2M with 100k$?
I've done anayses over time and in different OECD countries. They're not much different.
Are they going to allow their planes to be seized by hijackers [cooperate for ransom], or are they simply going to put the helm over into a dive rather than surrender? Or try aggressive flying (barrel rolls) before that?
The whole matter is very macabre, but needs to be thought through. IMHO, the total tower pancake collapses was an unexpected payoff to terror. What are the odds? Is it better to deliberately crash than possibly be used as a kinetic/incendary projectile?