While the MS absence at Plugfest might well be nice evidence, unfortunately it will disappear: MS will go next year so no-one will be able to use it against them. But they'll send lawyers, or disempowered Linux sandbox employees. And nothing will change.
Unfortunately, MS quite simply is utterly contempuous of all courts. They do not accept that the courts should have any authority over them. They don't believe that antitrust is "real" law that should be obeyed. At best, it is a hazard to be evaded.
I do not know how to change the philosophy of such a corrupt organization. Perhaps jailing Bill Gates for 6 months on contempt charges might help. More likely not, as they will see it as arbitrary and capricious. Everything I've seen says MS has an extremely strong internal culture that resists change.
More detail? Sure: Modern CPUs are a tough heat-transfer problem. Some circuits throw off a lot of heat, and some don't. This heat first goes into the die, where it spreads in 3D. Too thin, and it can't spread before it has to cross out of the die, through the thermal goop (mfrs have to find _really_ good stuff) and into the big shiney coppyer heatslug. Too thick, and it gets too warm because silicon isn't the best thermal conductor.
About 7 years ago, I wrote a suite of open-source CPU stress-tests I called `cpuburn`. Little optimized assember pgms designed to stress different parts of the CPU. `burnK7` does precisely this FPU dot product.
Of course, I expect AMD's production testing dept to have far better code, since they will devote more job hours to it and know proprietary chip details.
Still, different parts of AMD as emailed me several times to thank me because they found the pgms useful. Great.
But these guys know what they're doing. Heat transfer from the hot multipliers has to be carefully analysed [3D finite element heat transfer analysis]. I suspect something far more mundane,
like someone reducing die or slug thickness, or a mfg problem with the die/slug gap or thermal goop.
Having ID is one thing. Maybe good or bad. Being required to carry/show it is another. My large fear is that cops will sweep IDs without probable cause. RFID would make this automatic. Too many suspect hits will get you flagged.
I would like to see citizen oversight over police databases and data displays. Plus clear rules for when the police can request ID. IOW, retain probable cause.
This is no surprise. Intel dedicated its' best people to the IA64 (Itamium) Project, and so lost about 6 years of x86 CPU microarchitecture development. Had that effort succeeded, Intel would be well in the lead. But neither the complier nor the clock made it.
So it didn't, and this was actually somewhat predictable. Intel has tried these flyers about every 10 years: first, IAPX432. next i860/960. Now IA64. This was the most costly by far. I'm a little surprised they admit it, but possibly that's because the "A" team is back on x86, and they've got a comeback cooking. It wouldn't surprise me.
I thought a write would block (or erreturn) if the TCP send buffer got full (waiting for ACKs). It'd better, 'cuz otherwise RAM will fill with waiting buffers.
any client can trivially force a great many
pagefaults on the server by requesting several large streams, then just sitting on them
(since the server app doesn't get notified of this, it will just keep going around the
ring. The TCP window will fill quickly and the buffers will be waiting on the TCP ACK
that will never come)
Ooo.. I can see this'd be nasty! Memory pressure! Wouldn't this be better handled by tuning the TCP stack, specifically smaller buffers/windows until the connection proves-out?
I'm a bit mystified. Are there actually that many CoW pagefaults from userspace write buffers? Even if the write doesn't block, they're usually followed by a read() that surely will.
I also don't think the pagefault takes nearly as many cycles with APIC.
Artists make zip ($0.50/CD) royalties. Even a platinum seller is only 500 k$. However, the label will make 10 M$!
Downloads do not threaten artists, they threaten
labels. Arguably, downloads help artists by increasing the fan-base:
Concerts & tours have been how artists have always made the bulk of their money. Pricing is almost always done on the high-end, following the monopoly revenue-maximizing strategy. A single concert of 50,000 sold seats @ $80 will net the artist 1 M$. _That's_ worth the grind! Artists do not go on tour to support their albums, they release albums to support their tours!
I go to concerts to support artists. When I buy CDs, I know I'm supporting the RIAA.
Encouragement is not supplication. I encourage people to use Linux (when appropriate) because it's good for them, not because I receive any benefit.
I have no interest in seeing Linux "take" the market. That is far too hostile a term, however it is common in marketting-for-profit circles.
I also do not see any "us" vs "them". Near as I can tell, MS-Windows people are just like Linux or BSD people. Everyone wants to get work done. MS-Windws people have been seduced by ubiquity and advertising, but are well aware of the shortcomings of their choice. Linux just has greater shortcomings for them. And support is not one of those. Desktop applications & games are.
I don't doubt that open formats/stds/source is in the public interest. That's why copyrights expire (eventually, at least in theory). Protecting IP rights is in the individual's interest. It then becomes a matter of balancing rights to achieve the desired aims (usually economic growth).
The problem is that US legislators are often unduly influenced [bought] by campaign contributions. This will tip the scale. I give you the Sunny Bono Copyright Extention Act of 1996 as evidence.
Ah, but I do _not_ want world domination. MS-Windows might even have a place. And I do understand how the world works. Disciplining bad actors is important in all directions. Kowtowing is bad.
No, not Tom. He's gone. MSFT is looking to delay the EU rulings as much as possible. US subpoenae risk a contempt-of-court rebuke in the EU, but not much more.
It actually would have hurt MSFT worse if the subpoena were approved, however improper. It would have eliminated the appeal, and cut delays. MSFT know this, so make their requests as outrageous and opposible as possible.
MSFT have learned that in a fast-changing technological world, justice delayed is as good as justice denied. Unfortunately, the legal system works by giving maximum defense latitude, and that brings in delays.
Of course this happens! In the MS-Windows world too. Everywhere. People want and need help. They ask others. Often they get it, sometimes not.
However, to universally blame the help provider is completely wrong. The asker may be intruding. The asker may be insufficiently respectful or remunerative in other ways.
Has the "reaching" argument been dismissed: Shrink-wrap EULAs are invalid because contract was agreed to (sale made) without notice/availability of the EULA terms?
I can see click-thru being valid (so long as licence agrrement click is made before CC#).
Some of the viscious ad-hominem and other epithets heaped upon Bjorn Lomborg and others has been beyonjd unseemly. Even SciAm participated in the witchhunt.
I equate the invective with a fervently held belief that the invectors doubt can withstand criticism.
Global warming is a fairly simple concept. It most likely has been occuring over the past century, but definitely withing historical norms and probably withing historical rates-of-change. The cause is much less provable. Some people blame CO2 (especially anthropogenic), when it is almost certainly an effect (ever open a warm soda?).
Seriously, nice is not a good performance tuning tool in most cases. All it does is shorten the timeslice, inducing more cache misses. The work remains the same or goes up due to cache reloads.
A good [Linux] scheduler already does process restart immediately upon unblocking to reduce latency.
But an overloaded box is still overloaded. The question is how you want it to fail. In what direction? A little cron job watching load with a shedding/restart list probably does better.
Agreed. Use the old newspool trick: `updatedb` runs like greased lightening on a partition mounted with `noatime`. ~1 min for 40 GB. All those accesstime updates eat most of the wall time. I mount all partitions NOATIME in/etc/fstab unless I _really_ need atime.
Note this only applies to "sale or distribution", not purchased product delivery. Whence the rebuttable presumption on count.
For a web visit/purchase (pull-media), the transaction is presumed consummated at the server, much like sending a written order through the mail. So transport is of private property, not goods-for-sale.
Imagine they'd granted cert and taken the case. Just what could they have decided? Overturn Miller and establish national standards? What a farce: all juries are local and would decide using local standards. So they had to leave Miller.
Maybe they could've strengthened Internet immunities. But I don't think those need strengthening: "plain brown wrapper" applies: AFAIK, the offense is in publicly displaying (often for sale) obscence material. The Internet fits neatly into older models: no problem for pulled-media (website visits), a big problem around pushed-media (pr0n email spam). 'course there are problems catching the spammers, but that doesn't mean spam should be legal.
Slashdot Mirror
Unfortunately, MS quite simply is utterly contempuous of all courts. They do not accept that the courts should have any authority over them. They don't believe that antitrust is "real" law that should be obeyed. At best, it is a hazard to be evaded.
I do not know how to change the philosophy of such a corrupt organization. Perhaps jailing Bill Gates for 6 months on contempt charges might help. More likely not, as they will see it as arbitrary and capricious. Everything I've seen says MS has an extremely strong internal culture that resists change.
Of course, I expect AMD's production testing dept to have far better code, since they will devote more job hours to it and know proprietary chip details. Still, different parts of AMD as emailed me several times to thank me because they found the pgms useful. Great.
But these guys know what they're doing. Heat transfer from the hot multipliers has to be carefully analysed [3D finite element heat transfer analysis]. I suspect something far more mundane, like someone reducing die or slug thickness, or a mfg problem with the die/slug gap or thermal goop.
I would like to see citizen oversight over police databases and data displays. Plus clear rules for when the police can request ID. IOW, retain probable cause.
So it didn't, and this was actually somewhat predictable. Intel has tried these flyers about every 10 years: first, IAPX432. next i860/960. Now IA64. This was the most costly by far. I'm a little surprised they admit it, but possibly that's because the "A" team is back on x86, and they've got a comeback cooking. It wouldn't surprise me.
I also don't think the pagefault takes nearly as many cycles with APIC.
Downloads do not threaten artists, they threaten labels. Arguably, downloads help artists by increasing the fan-base:
Concerts & tours have been how artists have always made the bulk of their money. Pricing is almost always done on the high-end, following the monopoly revenue-maximizing strategy. A single concert of 50,000 sold seats @ $80 will net the artist 1 M$. _That's_ worth the grind! Artists do not go on tour to support their albums, they release albums to support their tours!
I go to concerts to support artists. When I buy CDs, I know I'm supporting the RIAA.
Seriously, if I get an RFID card, it's going inside a Faraday cage wallet. I don't mind RFID, but I'm going to control who I display to.
I have no interest in seeing Linux "take" the market. That is far too hostile a term, however it is common in marketting-for-profit circles.
I also do not see any "us" vs "them". Near as I can tell, MS-Windows people are just like Linux or BSD people. Everyone wants to get work done. MS-Windws people have been seduced by ubiquity and advertising, but are well aware of the shortcomings of their choice. Linux just has greater shortcomings for them. And support is not one of those. Desktop applications & games are.
The problem is that US legislators are often unduly influenced [bought] by campaign contributions. This will tip the scale. I give you the Sunny Bono Copyright Extention Act of 1996 as evidence.
It actually would have hurt MSFT worse if the subpoena were approved, however improper. It would have eliminated the appeal, and cut delays. MSFT know this, so make their requests as outrageous and opposible as possible.
MSFT have learned that in a fast-changing technological world, justice delayed is as good as justice denied. Unfortunately, the legal system works by giving maximum defense latitude, and that brings in delays.
However, to universally blame the help provider is completely wrong. The asker may be intruding. The asker may be insufficiently respectful or remunerative in other ways.
Beggars cannot be choosers.
I can see click-thru being valid (so long as licence agrrement click is made before CC#).
I equate the invective with a fervently held belief that the invectors doubt can withstand criticism.
Global warming is a fairly simple concept. It most likely has been occuring over the past century, but definitely withing historical norms and probably withing historical rates-of-change. The cause is much less provable. Some people blame CO2 (especially anthropogenic), when it is almost certainly an effect (ever open a warm soda?).
A good [Linux] scheduler already does process restart immediately upon unblocking to reduce latency.
But an overloaded box is still overloaded. The question is how you want it to fail. In what direction? A little cron job watching load with a shedding/restart list probably does better.
Note this only applies to "sale or distribution", not purchased product delivery. Whence the rebuttable presumption on count.
For a web visit/purchase (pull-media), the transaction is presumed consummated at the server, much like sending a written order through the mail. So transport is of private property, not goods-for-sale.
Maybe they could've strengthened Internet immunities. But I don't think those need strengthening: "plain brown wrapper" applies: AFAIK, the offense is in publicly displaying (often for sale) obscence material. The Internet fits neatly into older models: no problem for pulled-media (website visits), a big problem around pushed-media (pr0n email spam). 'course there are problems catching the spammers, but that doesn't mean spam should be legal.
This reduces the attractiveness of SBC machines to host bots. But SBC cannot block ports like 80 (HTTP), so SBCbots can still be used for DDoS.