Yes; that was understood. A subtle lurking point, however, is that once the third world countries have attained first world status, we'll not have to worry about this so much.
First world nations tend to have negative population growth rates, except by immigration influx, or population growth amongst recent-generation immigrants.
but when a US court of appeal overturn the damages awarded by a lower court to a rightful holder of a patent, it's just sheer justice done to a strategic company.
You must not hang out here very much. Pretty much the entire Slashdot crowd hates process patents. In the scheme of things--i.e., to the degree that we, the gallery, can have a scheme of things--the rationale behind the hatred of process patents is pretty good. I.e., those of us you are accusing of siding with the "convicted monopolist" in favor of your "rightful holder of a patent" are not doing what you say at all. We are opposing the notion that there ought to be a "rightful patent" on subjects like these. That we are doing so in the context of "siding" (SIC!) with what is otherwise our favored whipping boy, should help you grab a clue: this is not a matter of "siding" at all, but a matter of principle.
Process patents are Evil. They should be jettisoned from our collective legal systems with all possible haste.
I'm with ya. There's always a good remote option; you can even get remote serial port access if you must. While those systems add a nominal expense to the equipment budget, they save on the staff budget far more than they cost. And as far as these data centers go, if everything is wired up right, there's usually enough staff to directly visit the rare exception.
In a virtualized server environment, the problem of keeping the hypervisors up isn't yours. The problem of keeping the virtual servers running, may or may not be yours, but if it is, you will have a control client access layer that is akin to access to the HW.
It's not merely an issue of allowing more competition. For example, here in California cable TV is not a state-granted monopoly. And yet, you will find close to zero overlapping cable TV regions. Why? Because it makes little economic sense to the operators to do that. One operator, having paid for infrastructure, can lower prices in its region to below what a new competitor could afford, because the new competitor, having to lay down duplicate infrastructure, will be taking it over the barrel on paying for its new infrastructure. So the new operator just shies off from the whole thing. It's really a kind of willful collusion, but there's nothing evil about it. It's just good, obvious business sense.
At best, you can hope for the phone company, the cable company, and maybe some new third leg of wireless operators to form some kind of three way competitive market for delivery services. I don't think this is nearly enough, however, for any thing at all resembling competition. Markets with relatively small numbers of participants tend to engage in huge amounts of tacit collusion. Basically, it's very easy for the various players to watch each other's prices, set similar price points, and become lax about the whole thing. The victim is the consumer.
Real competition occurs in thriving markets where new competitors enter with innovations that lower the fundamental cost basis of their products. This forces competitors to adopt similar innovations or die. This seldom happens in small markets with a static set of competitors, because they're all set in their ways, and know the others are set in their ways. I.e., they can happily never change a thing and GET AWAY WITH IT.
So basically, don't hold your breath on any kind of real competition occurring here. While I'm a big fan of competitive markets, I'm a big cynic on this market. On a bad day, in a bad mood, I think we should just regulate the entire thing.
The most likely explanation is the difference in how controllers handle a marginal drive. For example, 10 years ago it was rare to have a RAID solution that re-wrote a bad block rather than just failing out the whole drive.
This is an interesting insight. I know there is a sea of differences in how enterprise SAN controllers do this. The very best ones will actually REBOOT hard drives, on the ground that most hard disk errors are actually firmware bugs. This was determined by the vendor, DDN, through extensive interviews with disk manufacturers. And you are right on, in observing that most disk drive problems are just failed sectors--the rest of the disk is generally fine. DDN talks this up quite a lot; their architecture is heavily built around real-time recovery (hence allowing drive reboot mid stream) and incremental rebuild (hence not having to rebuild whole drives in response to lost sectors).
Other vendors may do things like this also, but this is one of those cases where this is what the vendor "sells," if you get my drift. I.e., a core competency area.
I'd never considered the matter for DAS raid controllers. You think?
It's possible to have whole bad lots, even from the enterprise lines. I'm hearing this doesn't happen so often now, but when you're the person experiencing the bad lot, this can give one hell of a bad impression. Supposing this case, the phenomenon would be sampling bias.
I still avoid WD like the plague, even though I am aware they have long since fixed the quality problems that led to the bad lots in the 90's... can't help myself. Lost like 3 drives on a personal system in a row...
I agree with you. It's not the corporate tax rate. The situation is far more complicated.
Foreign companies have cheap labor. Foreign countries have lower standards of living. They don't have minimum wages. Foreign countries don't have environmental protection. They don't have labor protection, or worker safety laws. Foreign countries, even when they have equivalent tax burdens, have tax structures that favor exports, often dramatically.
Some of these things we could do something about. For example, some kind of VAT or sales tax instead of an income tax would seriously fire up our position as an exporter. This isn't just a matter of the corporate tax rate, either. It would remove all the hidden cost on workers from their own personal income taxes from the export supply chain. Once you know that, you wonder why we continue to allow the income tax (or any payroll tax) at all.
Other things we have to wait to fix themselves, such as foreign country standards of living.
A few things, such as lack of worker safety protection or environmental protection as a basis of cheap labor in an outsourced country are cause for moral concern. Arguably, these things will change with increases in the standards of living and the empowerment of people in those counties, but I don't personally care for that answer. I've often thought that we ought to apply a graduated import tax on foreign source countries that is inversely proportional to certain moral/ethical positions of the laws in the source country. I.e., if the country has no child labor protection laws, or the laws are known not to be enforced, you tack on X% export tax as punishment. You'd need to make sure a huge swath of the first world was on the same sheet of music with this, creating external pressure for social improvement in the source country, through partial denial to the world's largest markets.
A is not really complying, though. Which is, as you say, splitting hairs. As an aside, the FSF always wants compliance item C, and generally takes an encouraging enlightened approach in their contacts with infringers. Or so I hear.
I wonder how many companies, having found themselves in violation, have selected A, for real? I bet very few.
Citation please. I lurk on slashdot a lot and don't remember seeing such a thing, at least recently.
I don't particularly remember it recently, either. Perhaps the Slashdot public is getting more astute about copyright and contract law? Be that as it may: I've often seen poster or posters positing down this path, and from more than one angle. For example, I recall reading a recommendation once that someone publish some source code, not really theirs to give, under the GPL, because it was irreversible when done. Not so. Only the true and original owner can release rights like that.
But it can't really surprise you that a long time poster like me has seen it all, right?:-)
Anyway, my point is the current reality: a company faced with compliance has its choices. Consider the fringe case. An employee (or even a manager!) inappropriately integrates some GPL code into a much larger 1,000,000 line system. Let's suppose a valuable proprietary work. You don't really think the larger company is legally bound to do anything, right? They're not. They merely have to cease the infringement, and perhaps pay damages, depending.
In addition, the police officer is protected from a mistaken arrest--the citizen is not. If you conduct a citizen's arrest, you MUST be right, or there will be consequences.
Well. It's split hairs because you're familiar with the rules. An oft-repeated slashdot theme is "forcing" a company to distribute its source, or in some cases discussion of hostile access to source and repuplication of their source without permission. I'm not accusing you of this, it's just that I've seen a great deal of misconception of the legal consequences of a GPL violation around here,...
This is not correct. An entity finding itself in a position to comply with the GPL or not can indeed choose not; it's just that there will be many consequences. One such consequence could be being forced to rewrite one's software to a non GPL platform, and perhaps, pay damages to the original GPL copyright holder.
As resource bases contract and the world goes back to a solar economy,...
You mean after the nuclear economy takes off and we run out of fissibles? Don't you think making prognosticians about things thousands of years in the future is a little far sighted for your crystal ball?:-)
Yes, but only for a portion of the document. I believe the class of algorithms they use do something along the lines of sliding windows and hashing, which is what they do with the documents when they receive them, and what they store in their databases, that they can determine matches.
They are first party to copies made in entirety to their servers, aren't they? (This question is rhetorical, I don't know the answer) A better way would be to have the hashcode algorithm run by the submitter, and only the hashcodes uploaded. This would be, IMO, a sustainable fair use argument.
Statutory damages for copyright violation don't require any proof of harm at all. It's coded into the law, per se.
Your idea to attack proof of actual distribution is a good one, however, because the "making available" angle they've been using was made up out of whole cloth: there's nothing in the law for it at all. They're hoping to get it woven into jurisprudence without first weaving it into black letter law, abracadabra, like.
I wouldn't be suprised at all if they're pushing for legislation to make "making available" itself actionable. It would certainly make their life easier, eh?
Slashdot Mirror
Your position appears to be, implicitly, that immigrants into first world nations neither integrate themselves nor integrate their children?
C//
Yes; that was understood. A subtle lurking point, however, is that once the third world countries have attained first world status, we'll not have to worry about this so much.
C//
First world nations tend to have negative population growth rates, except by immigration influx, or population growth amongst recent-generation immigrants.
C//
but when a US court of appeal overturn the damages awarded by a lower court to a rightful holder of a patent, it's just sheer justice done to a strategic company.
You must not hang out here very much. Pretty much the entire Slashdot crowd hates process patents. In the scheme of things--i.e., to the degree that we, the gallery, can have a scheme of things--the rationale behind the hatred of process patents is pretty good. I.e., those of us you are accusing of siding with the "convicted monopolist" in favor of your "rightful holder of a patent" are not doing what you say at all. We are opposing the notion that there ought to be a "rightful patent" on subjects like these. That we are doing so in the context of "siding" (SIC!) with what is otherwise our favored whipping boy, should help you grab a clue: this is not a matter of "siding" at all, but a matter of principle.
Process patents are Evil. They should be jettisoned from our collective legal systems with all possible haste.
C//
I'm with ya. There's always a good remote option; you can even get remote serial port access if you must. While those systems add a nominal expense to the equipment budget, they save on the staff budget far more than they cost. And as far as these data centers go, if everything is wired up right, there's usually enough staff to directly visit the rare exception.
C//
In a virtualized server environment, the problem of keeping the hypervisors up isn't yours. The problem of keeping the virtual servers running, may or may not be yours, but if it is, you will have a control client access layer that is akin to access to the HW.
C//
Hmmm. Are you saying that the number of total instructions cleared per clock, is double for Nehalem over P4?
The multicore revolution more than obviously accomplished far more than that, but I wonder about single threaded performance, ...
C//
If we allowed more competition,...
It's not merely an issue of allowing more competition. For example, here in California cable TV is not a state-granted monopoly. And yet, you will find close to zero overlapping cable TV regions. Why? Because it makes little economic sense to the operators to do that. One operator, having paid for infrastructure, can lower prices in its region to below what a new competitor could afford, because the new competitor, having to lay down duplicate infrastructure, will be taking it over the barrel on paying for its new infrastructure. So the new operator just shies off from the whole thing. It's really a kind of willful collusion, but there's nothing evil about it. It's just good, obvious business sense.
At best, you can hope for the phone company, the cable company, and maybe some new third leg of wireless operators to form some kind of three way competitive market for delivery services. I don't think this is nearly enough, however, for any thing at all resembling competition. Markets with relatively small numbers of participants tend to engage in huge amounts of tacit collusion. Basically, it's very easy for the various players to watch each other's prices, set similar price points, and become lax about the whole thing. The victim is the consumer.
Real competition occurs in thriving markets where new competitors enter with innovations that lower the fundamental cost basis of their products. This forces competitors to adopt similar innovations or die. This seldom happens in small markets with a static set of competitors, because they're all set in their ways, and know the others are set in their ways. I.e., they can happily never change a thing and GET AWAY WITH IT.
So basically, don't hold your breath on any kind of real competition occurring here. While I'm a big fan of competitive markets, I'm a big cynic on this market. On a bad day, in a bad mood, I think we should just regulate the entire thing.
C//
The most likely explanation is the difference in how controllers handle a marginal drive. For example, 10 years ago it was rare to have a RAID solution that re-wrote a bad block rather than just failing out the whole drive.
This is an interesting insight. I know there is a sea of differences in how enterprise SAN controllers do this. The very best ones will actually REBOOT hard drives, on the ground that most hard disk errors are actually firmware bugs. This was determined by the vendor, DDN, through extensive interviews with disk manufacturers. And you are right on, in observing that most disk drive problems are just failed sectors--the rest of the disk is generally fine. DDN talks this up quite a lot; their architecture is heavily built around real-time recovery (hence allowing drive reboot mid stream) and incremental rebuild (hence not having to rebuild whole drives in response to lost sectors).
Other vendors may do things like this also, but this is one of those cases where this is what the vendor "sells," if you get my drift. I.e., a core competency area.
I'd never considered the matter for DAS raid controllers. You think?
C//
Hmmm. My reading of this is that they total blew the clock cycle predictions as well.
C//
It's possible to have whole bad lots, even from the enterprise lines. I'm hearing this doesn't happen so often now, but when you're the person experiencing the bad lot, this can give one hell of a bad impression. Supposing this case, the phenomenon would be sampling bias.
I still avoid WD like the plague, even though I am aware they have long since fixed the quality problems that led to the bad lots in the 90's... can't help myself. Lost like 3 drives on a personal system in a row...
C//
They said the disk failure rates were lower on the Sun too, but I don't know by how much.
You should be aware that neither Intel nor Sun make disk drives. The most likely explanation for this is manufacturer lot variance, or just luck.
C//
Oh and Thomas Jefferson would kill himself if he saw Bush in action,
You don't know our Founders by reputation very well. To wit:
Thomas Jefferson would kill Bush if he saw Bush in action.
C//
I agree with you. It's not the corporate tax rate. The situation is far more complicated.
Foreign companies have cheap labor.
Foreign countries have lower standards of living.
They don't have minimum wages.
Foreign countries don't have environmental protection.
They don't have labor protection, or worker safety laws.
Foreign countries, even when they have equivalent tax burdens, have tax structures that favor exports, often dramatically.
Some of these things we could do something about. For example, some kind of VAT or sales tax instead of an income tax would seriously fire up our position as an exporter. This isn't just a matter of the corporate tax rate, either. It would remove all the hidden cost on workers from their own personal income taxes from the export supply chain. Once you know that, you wonder why we continue to allow the income tax (or any payroll tax) at all.
Other things we have to wait to fix themselves, such as foreign country standards of living.
A few things, such as lack of worker safety protection or environmental protection as a basis of cheap labor in an outsourced country are cause for moral concern. Arguably, these things will change with increases in the standards of living and the empowerment of people in those counties, but I don't personally care for that answer. I've often thought that we ought to apply a graduated import tax on foreign source countries that is inversely proportional to certain moral/ethical positions of the laws in the source country. I.e., if the country has no child labor protection laws, or the laws are known not to be enforced, you tack on X% export tax as punishment. You'd need to make sure a huge swath of the first world was on the same sheet of music with this, creating external pressure for social improvement in the source country, through partial denial to the world's largest markets.
C//
I understand you clearly, Dale, and agree.
A is not really complying, though. Which is, as you say, splitting hairs. As an aside, the FSF always wants compliance item C, and generally takes an encouraging enlightened approach in their contacts with infringers. Or so I hear.
I wonder how many companies, having found themselves in violation, have selected A, for real? I bet very few.
C//
Citation please. I lurk on slashdot a lot and don't remember seeing such a thing, at least recently.
I don't particularly remember it recently, either. Perhaps the Slashdot public is getting more astute about copyright and contract law? Be that as it may: I've often seen poster or posters positing down this path, and from more than one angle. For example, I recall reading a recommendation once that someone publish some source code, not really theirs to give, under the GPL, because it was irreversible when done. Not so. Only the true and original owner can release rights like that.
But it can't really surprise you that a long time poster like me has seen it all, right? :-)
Anyway, my point is the current reality: a company faced with compliance has its choices. Consider the fringe case. An employee (or even a manager!) inappropriately integrates some GPL code into a much larger 1,000,000 line system. Let's suppose a valuable proprietary work. You don't really think the larger company is legally bound to do anything, right? They're not. They merely have to cease the infringement, and perhaps pay damages, depending.
C//
In addition, the police officer is protected from a mistaken arrest--the citizen is not. If you conduct a citizen's arrest, you MUST be right, or there will be consequences.
C//
Well. It's split hairs because you're familiar with the rules. An oft-repeated slashdot theme is "forcing" a company to distribute its source, or in some cases discussion of hostile access to source and repuplication of their source without permission. I'm not accusing you of this, it's just that I've seen a great deal of misconception of the legal consequences of a GPL violation around here, ...
C//
GPL compliance isn't optional.
This is not correct. An entity finding itself in a position to comply with the GPL or not can indeed choose not; it's just that there will be many consequences. One such consequence could be being forced to rewrite one's software to a non GPL platform, and perhaps, pay damages to the original GPL copyright holder.
C//
Yah. That was stupid. If they'd wanted to do that, they should have done what every other company who wants to do that does, and ripped a BSD kernel.
C//
As resource bases contract and the world goes back to a solar economy, ...
You mean after the nuclear economy takes off and we run out of fissibles? Don't you think making prognosticians about things thousands of years in the future is a little far sighted for your crystal ball? :-)
C//
Yes, but only for a portion of the document. I believe the class of algorithms they use do something along the lines of sliding windows and hashing, which is what they do with the documents when they receive them, and what they store in their databases, that they can determine matches.
C//
They are first party to copies made in entirety to their servers, aren't they? (This question is rhetorical, I don't know the answer) A better way would be to have the hashcode algorithm run by the submitter, and only the hashcodes uploaded. This would be, IMO, a sustainable fair use argument.
C//
One of the biggest I'd go for is proof of harm.
Statutory damages for copyright violation don't require any proof of harm at all. It's coded into the law, per se.
Your idea to attack proof of actual distribution is a good one, however, because the "making available" angle they've been using was made up out of whole cloth: there's nothing in the law for it at all. They're hoping to get it woven into jurisprudence without first weaving it into black letter law, abracadabra, like.
I wouldn't be suprised at all if they're pushing for legislation to make "making available" itself actionable. It would certainly make their life easier, eh?
C//
That does not mean that the software itself is actually sold.
Well, yeah. I know that. :-)
C//