People should keep in mind that Mr. Kircaali really doesn't have the option of fully admitting and apologizing for anything. That would just open up him and his company to a giant lawsuit.
He is open to giant lawsuits already, and this sort of behavior doesn't look good to a jury. If he comes crawling on his knees to PJ, perhaps they can reach a settlement.
Do not replace or add any software to the DISH 921 DVR with items compiled from these source trees. Doing so will void all warranties and cause the unit to fail.'"
Almost the whole point of the GPL is that you can do exactly this: you should be able to change the behavior of GPL'ed software components and replace the existing versions of it.
I suspect future versions of the GPL are going to try to limit these kinds of abuses: if you distribute systems containing GPL-derived binaries, you must ensure that people can reasonably replace your GPL'ed software components with components they recompiled. You should not be permitted to use either cryptographic means, warranties, patents, or proprietary development tools to prevent that.
Think about it: how would you feel if you saw your hard work being passed off as the property of someone else? What would you do?
Well, that's not difficult to determine for a computer scientist: all major computer companies pass off the hard work of other people as their own. Microsoft keeps doing it, and so do Apple and Sun. Of course, such acts of intellectual dishonesty and plagiarism do not automatically amount to a crime, as the term "thought thieves" erroneously suggest.
There is a basic flaw in their reasoning, which is to assume that everybody likes the "booth babe" type. Here's a hint: some guys like mature women, some guys like full women. Some guys even like guys, from androgynous boyish looks to furry bears. And some of the audience is women, straight women even, who would be physically more attracted by any male booth staff (although might still generally prefer chatting with female booth staff).
Hire people for personality, not looks, and you'll do well. And most people with the right personality for customer contacts will also know how to dress appropriately for the event and their body type. Of course, if you force everybody into some ridiculous booth outfit (skin tight pink T-shirts, etc.), then many people will look bad.
one would basically have to make sure that the other virtual CPU is idle before entering ring 0 at any system call
Probably not. For example, under UNIX, the bug changes nothing for two threads within the same program, or for two threads under the same uids. On systems with messy and unnecessarily complex privilege systems (like Windows), all bets are off, however.
What Intel might hope for is that this is fixable in microcode. If not, well, then there's real trouble.
More importantly, there are few multiuser systems around where this would matter much anymore. An exploit of this requires being able to execute arbitrary code on the machine handling the data. Almost anything that handles sensitive data is firewalled and allows only admin logins and specific services.
Well, I didn't see an explanation on the site... here is my guess what happened.
With HT, registers for multiple threads have to reside within the same CPU, whereas with SMP, they are physically in two different chips. Perhaps the registers used by HT are not properly protected against reading by other threads. The fact that SMP code was probably reused may be the cause of the problem: the code developed for SMP didn't have to deal with this situation before.
Guessing even a bit more, this may be hard to fix in software. If the hardware simply doesn't protect registers properly, then the kernel may have to clobber them to protect the information, but that may impose far too much overhead for HT.
However, a workaround might be to permit HT only for multiple threads within the same process. That would still give some speedups to compute-intensive processes that are written to take advantage of threading.
It's completely scientific?? Care to run that naive thing by me It's a biased, self-selecting sample with biased, pre-selected choices.
That describes a lot of psychiatric, psychological, and medical research.
Your problem is that you are using "scientific poll" as synonymous with "unbiased, randomly sampled poll", and that's just wrong. A biased, self-selected poll can yield plenty of useful, valid information; in fact, most polls are biased and self-selected, and companies still pay billions for them every year because they are useful.
The semantics of XA transactions say that at the end of the "prepare" step, the data is already on the disc (or whatever other medium), just not yet made visible. That, basically all that could possibly fail, has in fact had its chance to fail. And if you got an OK, then it didn't.
Well, and the drives still guarantee that, as long as you don't interrupt their power. If you put the computer on a UPS (or the external enclosure), then you pretty much guarantee it.
The small remaining probability that the drive loses power because of some other screwup is comparable to the probability that the disk block you wrote goes bad because of some other screwup. Whehther you want to deal with that probability, too, is up to you.
Well, it's unlikely this is going to change. The real solution is to give power long enough to the disk drive to let it complete its writes no matter what, and/or to add non-volatile or flash memory to the disk drive so that it can complete its writes after coming back up.
There is a fairly simple external solution for that: a UPS. They're good. Get one.
And even then it is not guaranteed that just because you write a block, you can read it again, because nothing can guarantee it. So, file systems need to deal, one way or another, with the possibility that this case occurs.
Most systems can identify and patch out bad sectors so that they aren't used. What surprised me is that the manufacturers have their own bad sector table, so when you get the disk it's fairly likely that there are already bad areas which have been mapped out.
Well, the thing that's disturbing about that is that it potentially messes up scheduling algorithms in the OS, but apparently not enough to matter.
It's been quite a few years but it was somewhere between on error in every 10E5 to 10E6 bits. So it's not unusual to find a mistake in every megabyte read. Of course CRC picks up this error and hides that from you too.
The people who design these things calculate those error rates and make them part of the design.
A JFS needs to be sure the journal has been flushed out to disk (and resides safely on the platters) before continuing to write the actual (meta)data.
That won't happen reliably even if the disk behaves properly: you still get bad sectors and other problems (like this one, for example).
A file system must take into account that arbitrary blocks can disappear at arbitrary times. Building a JFS assuming that the disk is going to work as advertised is simply insane.
Unfortunately, there is a lot of insanity like that in software.
I believe Windows implements this in the "kernel" or as part of the operating system libraries. The Borland patent pretty specifically refers to a "software development system".
So, if this were incorporated into a separate library, independent of a development system, or if it were implemented via a kernel call, it might not infringe.
The basic ideas, in any case, are not patentable because they have been in use much longer than Borland. There are also plenty of ways of implementing something functionally equivalent that work completely differently; only Windows programs that rely on the exact layout of the stack would fail then.
Have you heard of people saying that in order to get a patent there should be an implementation?
That isn't the case anymore. When it was the case, it was referring to your own implementation. You could not and cannot patent something you did not invent.
This is one such patent that does exactly that. Referring to one implementation does not invalidate the concepts being patented.
Sure it does: if it's someone else's implementation that predates the patent filing by enough time.
They check to make sure that it doesn't run on P4 and Athlon, and the design (=resource hogging) of the operating system and common applications ensures that users wouldn't want to run it on anything else.
I haven't had problems. However, if USB doesn't work for you, you can always hotsync through Bluetooth or 802.11. For Bluetooth, you can either use the Bluez stack and have a choice between serial or network sync, or you can use a Bluetooth-to-Ethernet interface and use network sync.
Social filtering of email is already widely implemented. It's not implemented at the level of the end user email client, but at the level of email servers, which compare messages to users and blackhole addresses.
A P2P approach and querying of other people's address books has huge privacy and compatibility problems without any obvious advantages.
No, what I'm saying is that the members of the Open Source community who want to guide the development of the Java platform do so by working with Sun and the JCP
Sun owns the output of the JCP; why should people work for free to contribute to Sun's proprietary interests? Do you think open source developers are complete dopes? Sun should bloody well pay for the development of their own proprietary software themselves. Even Microsoft is at least that honest.
just simply bitching, whining, and moaning about Sun having the extreme audacity to maintain some level of control of their own intellectual property
The bitching and moaning isn't about Sun maintaining their intellectual property, it's about Sun lying about what they are doing. Java is proprietary through and through, yet Sun pretends it's some open alternative to Microsoft.
If you take off your "if-it-isn't-open-source-it-sucks" blinders for a moment, you'd see it instantly. But I'm not gonna hold my breath...
I used to be exclusively a Sun user: their software was the best and their machines were inexpensive and powerful. Then their software started deteriorating, their prices started skyrocketing, and alternatives got better and better. Porting software was painful, and it was one of the reasons why I don't want to get locked into a vendor anymore. And given Sun's recent behavior, I particularly don't want to have anything to do with that company anymore.
For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.
This is a stated goal of the GPL:
For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.
There is more, but you get the idea. Read the GPL.
Winelib is covered by the LGPL. If Borland has shipped Winelib in Kylix, they may already have given people a license to use the code. Perhaps they can wiggle out of that, but they can't continue to ship Winelib if they claim a patent on something in it.
So, to summarize, you are saying that there is no good open source implementation of Java because nobody in the open source community is capable of producing such a beast.
There are high-quality open source implementations of SMB, of UNIX, of X11, of C++, of CommonLisp, of HTTP, of OpenGL, of Eiffel. But you are saying that Java (a failed attempt at creating a language for set-top boxes by a bunch of part-time language designers) is so special and so advanced that no open source developer is capable of producing it.
Yet you have us believe that Sun can? A company that failed with NeWS and SunView? A company with bug lists longer than the Bible? A company incapable of designing a decent network file system? A company whose success is based on BSD UNIX and GNU user mode tools?
Give me a break. The only reason Sun still controls Java is because they rigged the process: they own the output from the JCP, and they control the so-called "compatibility testing process". Sun is scared stiff of allowing competition in the Java market because they know they'd lose control within a few months because they are doing such a piss-poor job.
The currency of science is citations: the more you are cited, the more you are worth. Academics therefore have a natural incentive to have their work be more accessible.
That is partly balanced by the fact that papers published in well-marketed journals with recognizable brand names will be cited more frequently. But they still have to be well-known, which is why even expensive journals tolerate "illegal" copies of scientific papers (this is similar to software companies tolerating some piracy and low-cost versions in order to keep low-cost competitors from entering the market).
On balance, I think academic publishers are going to lose this one for the most part. In the end, they don't offer any value, since all the hard work is already volunteer work. All the academic publishers do is marketing, printing, type setting, and mailing to libraries, and none of those are essential for academic journals anymore. Some journals will probably continue to be proprietary and expensive, but most will probably not be.
So is threatening your customers with a base-ball bat in order to extract more money. Trouble is: your customers have a choice, and they may choose to do business with people without baseball bats and without lock-in.
If you found a company based on some idealistic notion that lock-in is bad, then you are going to fail just like any other two-bit company with no business sense.
And if you found a company based on the mistaken notion that you can lock in your customers into whatever piece of crap you produce, you are also going to fail, just like any other two-bit company with delusions of grandeur.
The results of poll on a web site are more than less than scientific, they're pretty much meaningless.
A poll on a website is completely scientific, and it has a well-defined meaning; it simply happens to be a meaning that differs from what you might naively expect; it is not an unbiased sample of a population. But it doesn't have to be an unbiased sample of a population in order to be useful, since we aren't interested in (say) predicting the outcome of an election or sales figures for a new product.
Attempting to draw a number of conclusions from said polls is downright silly.
Every poll, no matter how it has been conducted, allows both silly and sensible conclusions.
Lots of visitors participate in those polls because they hope to see a specific result, or to prevent one, or to annoy someone in the room with them
That is an implausible hypothesis in this case. Pushing "less vendor dependence" as a hypothesis over "lower cost" is not exactly a pet peeve of a lot of people. The result of this poll doesn't "prove" anything (few polls ever do), but it does fulfill a primary function of polls: to give you ideas about where things might stand or where they might be headed.
Slashdot Mirror
People should keep in mind that Mr. Kircaali really doesn't have the option of fully admitting and apologizing for anything. That would just open up him and his company to a giant lawsuit.
He is open to giant lawsuits already, and this sort of behavior doesn't look good to a jury. If he comes crawling on his knees to PJ, perhaps they can reach a settlement.
Do not replace or add any software to the DISH 921 DVR with items compiled from these source trees. Doing so will void all warranties and cause the unit to fail.'"
Almost the whole point of the GPL is that you can do exactly this: you should be able to change the behavior of GPL'ed software components and replace the existing versions of it.
I suspect future versions of the GPL are going to try to limit these kinds of abuses: if you distribute systems containing GPL-derived binaries, you must ensure that people can reasonably replace your GPL'ed software components with components they recompiled. You should not be permitted to use either cryptographic means, warranties, patents, or proprietary development tools to prevent that.
Think about it: how would you feel if you saw your hard work being passed off as the property of someone else? What would you do?
Well, that's not difficult to determine for a computer scientist: all major computer companies pass off the hard work of other people as their own. Microsoft keeps doing it, and so do Apple and Sun. Of course, such acts of intellectual dishonesty and plagiarism do not automatically amount to a crime, as the term "thought thieves" erroneously suggest.
There is a basic flaw in their reasoning, which is to assume that everybody likes the "booth babe" type. Here's a hint: some guys like mature women, some guys like full women. Some guys even like guys, from androgynous boyish looks to furry bears. And some of the audience is women, straight women even, who would be physically more attracted by any male booth staff (although might still generally prefer chatting with female booth staff).
Hire people for personality, not looks, and you'll do well. And most people with the right personality for customer contacts will also know how to dress appropriately for the event and their body type. Of course, if you force everybody into some ridiculous booth outfit (skin tight pink T-shirts, etc.), then many people will look bad.
one would basically have to make sure that the other virtual CPU is idle before entering ring 0 at any system call
Probably not. For example, under UNIX, the bug changes nothing for two threads within the same program, or for two threads under the same uids. On systems with messy and unnecessarily complex privilege systems (like Windows), all bets are off, however.
What Intel might hope for is that this is fixable in microcode. If not, well, then there's real trouble.
If people care about HT at all...
More importantly, there are few multiuser systems around where this would matter much anymore. An exploit of this requires being able to execute arbitrary code on the machine handling the data. Almost anything that handles sensitive data is firewalled and allows only admin logins and specific services.
Well, I didn't see an explanation on the site... here is my guess what happened.
With HT, registers for multiple threads have to reside within the same CPU, whereas with SMP, they are physically in two different chips. Perhaps the registers used by HT are not properly protected against reading by other threads. The fact that SMP code was probably reused may be the cause of the problem: the code developed for SMP didn't have to deal with this situation before.
Guessing even a bit more, this may be hard to fix in software. If the hardware simply doesn't protect registers properly, then the kernel may have to clobber them to protect the information, but that may impose far too much overhead for HT.
However, a workaround might be to permit HT only for multiple threads within the same process. That would still give some speedups to compute-intensive processes that are written to take advantage of threading.
It's completely scientific?? Care to run that naive thing by me It's a biased, self-selecting sample with biased, pre-selected choices.
That describes a lot of psychiatric, psychological, and medical research.
Your problem is that you are using "scientific poll" as synonymous with "unbiased, randomly sampled poll", and that's just wrong. A biased, self-selected poll can yield plenty of useful, valid information; in fact, most polls are biased and self-selected, and companies still pay billions for them every year because they are useful.
UPSs can still fail you. Something as simple as tripping on a power cord.
Yes, and so can disks: even a disk that is constructed only to acknowledge writes only after they have completed still cannot guarantee things.
Flash is too slow.
Not necessarily--depends on how parallel you make it.
What most raid cards worth their salt do is have a large DRAM cache which is battery backed.
Yes, as I was saying.
The semantics of XA transactions say that at the end of the "prepare" step, the data is already on the disc (or whatever other medium), just not yet made visible. That, basically all that could possibly fail, has in fact had its chance to fail. And if you got an OK, then it didn't.
Well, and the drives still guarantee that, as long as you don't interrupt their power. If you put the computer on a UPS (or the external enclosure), then you pretty much guarantee it.
The small remaining probability that the drive loses power because of some other screwup is comparable to the probability that the disk block you wrote goes bad because of some other screwup. Whehther you want to deal with that probability, too, is up to you.
Well, it's unlikely this is going to change. The real solution is to give power long enough to the disk drive to let it complete its writes no matter what, and/or to add non-volatile or flash memory to the disk drive so that it can complete its writes after coming back up.
There is a fairly simple external solution for that: a UPS. They're good. Get one.
And even then it is not guaranteed that just because you write a block, you can read it again, because nothing can guarantee it. So, file systems need to deal, one way or another, with the possibility that this case occurs.
Most systems can identify and patch out bad sectors so that they aren't used. What surprised me is that the manufacturers have their own bad sector table, so when you get the disk it's fairly likely that there are already bad areas which have been mapped out.
Well, the thing that's disturbing about that is that it potentially messes up scheduling algorithms in the OS, but apparently not enough to matter.
It's been quite a few years but it was somewhere between on error in every 10E5 to 10E6 bits. So it's not unusual to find a mistake in every megabyte read. Of course CRC picks up this error and hides that from you too.
The people who design these things calculate those error rates and make them part of the design.
A JFS needs to be sure the journal has been flushed out to disk (and resides safely on the platters) before continuing to write the actual (meta)data.
That won't happen reliably even if the disk behaves properly: you still get bad sectors and other problems (like this one, for example).
A file system must take into account that arbitrary blocks can disappear at arbitrary times. Building a JFS assuming that the disk is going to work as advertised is simply insane.
Unfortunately, there is a lot of insanity like that in software.
I believe Windows implements this in the "kernel" or as part of the operating system libraries. The Borland patent pretty specifically refers to a "software development system".
So, if this were incorporated into a separate library, independent of a development system, or if it were implemented via a kernel call, it might not infringe.
The basic ideas, in any case, are not patentable because they have been in use much longer than Borland. There are also plenty of ways of implementing something functionally equivalent that work completely differently; only Windows programs that rely on the exact layout of the stack would fail then.
Have you heard of people saying that in order to get a patent there should be an implementation?
That isn't the case anymore. When it was the case, it was referring to your own implementation. You could not and cannot patent something you did not invent.
This is one such patent that does exactly that. Referring to one implementation does not invalidate the concepts being patented.
Sure it does: if it's someone else's implementation that predates the patent filing by enough time.
They check to make sure that it doesn't run on P4 and Athlon, and the design (=resource hogging) of the operating system and common applications ensures that users wouldn't want to run it on anything else.
I haven't had problems. However, if USB doesn't work for you, you can always hotsync through Bluetooth or 802.11. For Bluetooth, you can either use the Bluez stack and have a choice between serial or network sync, or you can use a Bluetooth-to-Ethernet interface and use network sync.
Social filtering of email is already widely implemented. It's not implemented at the level of the end user email client, but at the level of email servers, which compare messages to users and blackhole addresses.
A P2P approach and querying of other people's address books has huge privacy and compatibility problems without any obvious advantages.
No, what I'm saying is that the members of the Open Source community who want to guide the development of the Java platform do so by working with Sun and the JCP
Sun owns the output of the JCP; why should people work for free to contribute to Sun's proprietary interests? Do you think open source developers are complete dopes? Sun should bloody well pay for the development of their own proprietary software themselves. Even Microsoft is at least that honest.
just simply bitching, whining, and moaning about Sun having the extreme audacity to maintain some level of control of their own intellectual property
The bitching and moaning isn't about Sun maintaining their intellectual property, it's about Sun lying about what they are doing. Java is proprietary through and through, yet Sun pretends it's some open alternative to Microsoft.
If you take off your "if-it-isn't-open-source-it-sucks" blinders for a moment, you'd see it instantly. But I'm not gonna hold my breath...
I used to be exclusively a Sun user: their software was the best and their machines were inexpensive and powerful. Then their software started deteriorating, their prices started skyrocketing, and alternatives got better and better. Porting software was painful, and it was one of the reasons why I don't want to get locked into a vendor anymore. And given Sun's recent behavior, I particularly don't want to have anything to do with that company anymore.
Sure it does:
For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.
This is a stated goal of the GPL:
For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.
There is more, but you get the idea. Read the GPL.
Winelib is covered by the LGPL. If Borland has shipped Winelib in Kylix, they may already have given people a license to use the code. Perhaps they can wiggle out of that, but they can't continue to ship Winelib if they claim a patent on something in it.
So, to summarize, you are saying that there is no good open source implementation of Java because nobody in the open source community is capable of producing such a beast.
There are high-quality open source implementations of SMB, of UNIX, of X11, of C++, of CommonLisp, of HTTP, of OpenGL, of Eiffel. But you are saying that Java (a failed attempt at creating a language for set-top boxes by a bunch of part-time language designers) is so special and so advanced that no open source developer is capable of producing it.
Yet you have us believe that Sun can? A company that failed with NeWS and SunView? A company with bug lists longer than the Bible? A company incapable of designing a decent network file system? A company whose success is based on BSD UNIX and GNU user mode tools?
Give me a break. The only reason Sun still controls Java is because they rigged the process: they own the output from the JCP, and they control the so-called "compatibility testing process". Sun is scared stiff of allowing competition in the Java market because they know they'd lose control within a few months because they are doing such a piss-poor job.
The currency of science is citations: the more you are cited, the more you are worth. Academics therefore have a natural incentive to have their work be more accessible.
That is partly balanced by the fact that papers published in well-marketed journals with recognizable brand names will be cited more frequently. But they still have to be well-known, which is why even expensive journals tolerate "illegal" copies of scientific papers (this is similar to software companies tolerating some piracy and low-cost versions in order to keep low-cost competitors from entering the market).
On balance, I think academic publishers are going to lose this one for the most part. In the end, they don't offer any value, since all the hard work is already volunteer work. All the academic publishers do is marketing, printing, type setting, and mailing to libraries, and none of those are essential for academic journals anymore. Some journals will probably continue to be proprietary and expensive, but most will probably not be.
Let's face it: lock in is just smart business
So is threatening your customers with a base-ball bat in order to extract more money. Trouble is: your customers have a choice, and they may choose to do business with people without baseball bats and without lock-in.
If you found a company based on some idealistic notion that lock-in is bad, then you are going to fail just like any other two-bit company with no business sense.
And if you found a company based on the mistaken notion that you can lock in your customers into whatever piece of crap you produce, you are also going to fail, just like any other two-bit company with delusions of grandeur.
The results of poll on a web site are more than less than scientific, they're pretty much meaningless.
A poll on a website is completely scientific, and it has a well-defined meaning; it simply happens to be a meaning that differs from what you might naively expect; it is not an unbiased sample of a population. But it doesn't have to be an unbiased sample of a population in order to be useful, since we aren't interested in (say) predicting the outcome of an election or sales figures for a new product.
Attempting to draw a number of conclusions from said polls is downright silly.
Every poll, no matter how it has been conducted, allows both silly and sensible conclusions.
Lots of visitors participate in those polls because they hope to see a specific result, or to prevent one, or to annoy someone in the room with them
That is an implausible hypothesis in this case. Pushing "less vendor dependence" as a hypothesis over "lower cost" is not exactly a pet peeve of a lot of people. The result of this poll doesn't "prove" anything (few polls ever do), but it does fulfill a primary function of polls: to give you ideas about where things might stand or where they might be headed.