Well, it probably does have more money than a few countries, like say, the one that sould the rights to.tv
On the other hand, the article is about "per-captia" income. So what it really says is that people sink more money on average into Everquest than many people have to spend. It doesn't mean they have more money total--most countries have populations larger than Everquest.
On the other hand, it is kind of sad that you can earn more from playing Everquest all day than the people in a lot of countries earn in a week. I wonder if "virtual sweatshops" could actually come into being--people come into work, log on to computers, and make virtual artifacts all day. Hey--probably beats farming.
The idea of UWB making GPS "obsolete" is pretty laughable. The article gives no details on how it works, but since the range is limited, I'm imagining is uses triangulation from a bunch of ground stations. So, OK, all those little GPS receivers to find your way around a city will become obsolete.
But for wilderness and nautical applications, what good is a limited range signal going to do you?
I really doubt that even the majority of RISC chips produced are 64 bit, since RISC chips are very popular for embedded computing. Most ARM chips are 32-bit in Arm mode and 16-bit in Thumb mode. The PowerPC 601, 603, 750, 7400 (Altivec) etc architectures are all 32 bit. MIPS chips using ISA I through ISA III are 32 bit, I believe only in ISA IV did any 64 bit instructions get added.
I know the Alpha is 64 bit, but that is an insane chip. I just found out at lunch (can you tell I'm a geek) that it has PAL units on it, too, among other things, emulate certain really usefull VAX instructions. Basically, you can do a sort of "make your own instruction" thing. Many high-end RISC chips produced now are 64 bit, but, as I said, pretty sure its not the majority. Even if it were, it's a dumb thing to have in the article. RISC design in and of itself doesn't make it easier or harder to add 64bit support, although the if you make all your instructions 64 bit you're going to take a MASSIVE code size hit. (And you better at least double the size of your I-cache).
Well, if you don't have to tell a third party to use Plucker, it looks from a reading of the patent that it wouldn't even be infringing. Having state stored by a provider seems to be a key part of the patent.
I'm guessing you read that and came back here to post in indignation. Probably a good thing, if you have high blood pressure. Among other "gems" from the article:
"RISC chips not only process multiple instructions at the same time but also run at 64 bits"
The story says VeriSign bought.tv, but the prime minister seems to be under the impression that they are taking over management, or at least that's the impression I got from the article...
Well, the Prime Minister is correct, the poster is wrong. Tuvalu sold the rights to.tv
two years ago to a company called Idealab for $50 million. So Idealab lost money. Now Verisign is buying it for less than that. Big deal.
"Where exactly did the poster you're replying to bitch about stamp costs?"
One of the things he was "f***ed on" was having to pay for stamps.
"As far as it goes, if I could get reliably delivered mail that wasn't shredded half the time, I'd be glad to pay additional postage. But since the USPS is effectively another government granted monopoly (or is it a goverment agency? Hm, wish they could make up their minds), there's no real competition for letter carrying to force them to be worth a damn."
Well, you could send the documents FedEx. They are competition for the high end, fast, reliable delivery market. The charge a _lot_ more, because they can't just lose money.
"You subsidize spammy junk mail catalogs because companies get favorable bulk mail rates"
Companies get favorable bulk rates because bulk prebarcoded mail is easier to process. I don't think you "subsidize" it.
"And now you have to veiw adds while you stand in line or check your box"
OK, I hate ads too. But you're already bitching about your stamp costs. How do you want them to make enough money to continue their operations? I barely ever go into a post office--if it bothers you that much, you could probably avoid it too.
Of course, since it was just an encoded mathematical problem, I doubt they even bothered to copyright it. And it was the poster, not the editor, who made the stupid comment. I still would argue that the flambait line should of have been removed, but hey. Too bad you can't mod stories.
The idea sounds interesting, but kinda gimmicky. Especially with a scholarship for speed--with a problem it took 30 minute for _New_Scientist_ to solve (or did I misread something) that seems a bit silly. Now, waiving the application fee for anyone who solves it, that seems a more commensurate prize.
I don't know if it would be all that bad. Remember, this isn't some random plane, this is one specifically retrofitted for high-level Chinese government officials. It's a matter of respect for our spooks to bug it--shows we're taking them seriously. Then their spooks comb the plane, trying to find _all_ the bugs we planted. The Russians did the same thing with the embassy they built for us, I'm sure we did the same for them.
Parent got modded up as funny, but it actually makes a serious good point. One of the people interviewed for the article said that without this hardware copy prevention, "music could become a cottage industry in a few years." Guess what: tough shit. How would the world be a worse place if music was a "cottage industry" run by many small independant companies? Would we really lose anything?
I'm in favor of copyright, generally. I think it's worth it to have people who can spend their entire lives producing entertainment. But that's not what this battle is about. If piracy _actually_ started making it too hard to produce new content, then there would be a public backlash that would fix things with either a cultural or technical solution
This battle is about maintaining record companies and big studios place in the revenue stream. And they are becoming obsolete. This is like professional letter writers (yes, they actually used to exist) lobbying against public education because it would doom there buisness.
It's not that "real time processing does not belong in a macrokernel architechture" it's that "macrokernel architechture does not belong in a hard real time system"
I don't see any problem with making the linux kernel preemptible to be able to make better real time garuntees. Sure, I don't think you'll ever get hard realtime, but that doesn't mean that you won't get benefits from being able to respond to interrupts even when the system is running the kernel.
("hard realtime" -- maximum interrupt latency of xxx nanoseconds. "soft realtime" -- runs fast enough, usually)
Where are all the people from the debate on making all government-funded software open source? Why aren't they in this debate howling because he was let off easy? After all, if state money means "the people" own everything you do, he was stealing from "the people" with this--especially since he wasn't going to share the money if he won.
I think the prosecution of McOwen went overboard. However, if you don't own machines, you shouldn't run software on them without permission. Increasing system load means the system spends less time in power saving modes. It does consume bandwith. Also, and I don't know if this was the case here, it can be damned annoying when not set up properly. When I was working to help administer the computers at a Math department at my college, the sysadmin for general computing stayed logged in and run Seti@home through scripts. Problem was, he didn't do a very good job, and sometimes two or more copies would run at once. They also seemed to take a perceptible amount of time to get off the CPU.
"It would be a logistical hassle to decide what's just 'incidental' and what's primary to the research, too. I imagine that while we allow 'incidental' work to be closed and exploited for private gain, most of it will miraculously be 'incidental.'"
There is a difference between not allowing commercial release and requiring open release. The petition would require researches to openly publish their code as a condition of the grant, just as they are required to publish their research. That is the focus of the principal objections by the author of the opposing paper.
Not allowing closed-source, commercial releases of software developed as part of a government contract is much more limited in scope and more reasonable. It solves the NDA problems--the modification to the propreitary program can't be sold by the original company, but the distrubution can be restricted. Some researchers would object to this, but it does put things on a footing similar to that of most employees. If I, for instance, develop any software using company resources such as computers, I can't distrubute it without a written disclaimer of interest from my company. I think, also, that it would satisfy the opposing essay author.
That's exactly what I meant by "constrain" Why shouold research in which the software is incidental not be given a grant just because some modification they make to a propreitary software package won't be released? Qui bono?
"software patents are a perversion the framers would never have assented to"
Along with granting the vote to women, but that's another thread. My example didn't refer to software patents, but to algorithmic patents. I think these fit perfectly with the intent of the framers. An algorithm takes research to develope. It is often hard to duplicate, even if you see the results (unlike most software patents). By making them patentable, the technique used to develope the algorithm is available immediately and the algorithm itself is garunteed to enter the public domain in a limited amount of time. By the by, I agree with you completely on the issue of the long duration of software copyright.
"The right answer is eminently simple. That unclassified research and its fruits (in this case, software) paid for by the people belong to the people. If we're to buy into the intellectual "property" rhetoric of the software industry, then I guess that means that not releasing government funded software into the public domain is piracy and theft!
The right answer is eminently simple. That unclassified research and its fruits (in this case, software) paid for by the people belong to the people. If we're to buy into the intellectual "property" rhetoric of the software industry, then I guess that means that not releasing government funded software into the public domain is piracy and theft!
"The right answer is eminently simple. That unclassified research and its fruits (in this case, software) paid for by the people belong to the people. If we're to buy into the intellectual "property" rhetoric of the software industry, then I guess that means that not releasing government funded software into the public domain is piracy and theft"
The right answer is seldom simple. If we were talking about software developed as one of the primary objectives of a government grant, I might agree with you. But we're not. The issue is software produced in the course of doing other government research. Should the public be granted access to the spare cycles of any government funded machines? Of course not, the logistical hassle far outways the benefits. The same is true in many cases with software, which is why I disagree with petition as currently written.
"I am sure that one can find other examples of this, and I am sure they are equally outrageous"
I'm sure you can "find" other examples too, since you just make them up out of whole cloth. Please state the drug and manufacturer that you are thinking of?
Bullshit. Your and many other posts seem to miss a key point of this debate: The software produced is often incidental to the main purpose of the research. And forcing the public release of the source code does have its downsides.
First, as the opposing essay points out, no software can be liscenced from a commercial vendor and customized to suit the needs of the research project, now this can be done with an NDA.
Second, this produces a "tainting" effect from public money vis a vis research rights. Say, while working on a partly government-funded climate modeling project, I come up with an improved algorithm. If this proposal became law, I would not use that algorithm in the final software because then I would lose the ability to patent it. If lawsuits started to arise, it would be even worse for everyone but the lawyers: Have you heard from people working on stem cell research how hard it is to completely seperate private/public funding?
I understand and agree with the goals of the supporters of this petition. Publication of software under open-source or public domain liscences would be for the public benefit. However, this blanket approach ignores cases where this benefit would be outweighed by slowing the research, the results of which would benefit the public. Instead, government agencies should be encouraged to put language in grants that requires the release of software produced in the course of the project. However, they should be given the discretion to leave out the software provision in cases where they would constrain the research.
Yes, UDP doesn't resend packets. It doesn't need a concept of packet reassembly since IP does it for you. But you've pointed out the hitch " If the UDP-based application doesn't implement [error correction, retries, etc], then you're not going to have much luck transferring large files".
For something like multiplayer gaming, or most voice conferencing apps, retries are not needed. As you pointed out, " If you don't receive a packet on time, it is usually better to just drop it than keep retrying."
The problem comes when you implement an application where latency doesn't matter as much, but not dropping packets does, like file transfer. TCP has a carefully tuned retransmission scheme to prevent it from saturating networks. The IETF is legitimately concerned whenever a protocol that needs reliable delivery uses UDP.
I do think that the article generally was a bit alarmist. However, the concerns it presented were legitimate and I would call it particularly ignorant.
It's not that the energy would be from particles from higher dimensions, but rather that without the extra dimensions the cosmic rays would have insufficent energy to form the black holes in the first place. This is made pretty clear by even a cursory reading of the article, but that seems to doesn't seem to be the norm around here.
Now, a slightly more interesting question is why the extra dimensions would lower the threshold of singularity formation. Do they make space more compressible by giving it more room to flex in or what?
The reason the author's of the article are worried is that SIMPLE, the new instant messaging interconnect protocol, supports UDP as well as TCP. UDP has no congestion control. It is horrible to transfer large files with as most applications end up resending too many packets. This could, in fact, bring backbone segments down.
Of course "Its not the end of the world" The end of the article even said that it wasn't a "showstopper" for SIMPLE. But it is a genuine problem.
The last proposal I remember reading (admitedly, this was Spring '00 when the IMPP group list degenerated into flamewars and I stopped following it) was simply to tag things with username@service or impp://service/username. There's several solutions, I'm not sure which one they're going with but it shouldn't be a big issue.
I know that one company I used to work for had a military contract and one of the items was to garuntee that the data containing parts of the system would be completely melted by X degress at temperature Y. I believe they did something like put thermite and a triggering unit in the case with it.
That, of course, does much more damage than just a chip explosion. I've only seen one exploded chip--the SCSI controller on an MVME2700 (Motorola) blew about a quarter of the area and half the depth out. I was kinda reassured when I looked at my own board and saw that it had a different model chip in that spot. I don't think it could have killed anyone put it certianly could have done some serious damage to an eye if you leaned over to plug the chassis in. (I wish I had the picture handy, put I don't)
"when people click there from a Microsoft Exchange email message, Exchange helpfully gives us the subject line and username."
This bothers me far more than the "poll fixing" Do people realize that this kind of information is leaking all over the net? Is there a way to disable this "feature"?
Slashdot Mirror
Well, it probably does have more money than a few countries, like say, the one that sould the rights to .tv
On the other hand, the article is about "per-captia" income. So what it really says is that people sink more money on average into Everquest than many people have to spend. It doesn't mean they have more money total--most countries have populations larger than Everquest.
On the other hand, it is kind of sad that you can earn more from playing Everquest all day than the people in a lot of countries earn in a week. I wonder if "virtual sweatshops" could actually come into being--people come into work, log on to computers, and make virtual artifacts all day. Hey--probably beats farming.
The idea of UWB making GPS "obsolete" is pretty laughable. The article gives no details on how it works, but since the range is limited, I'm imagining is uses triangulation from a bunch of ground stations. So, OK, all those little GPS receivers to find your way around a city will become obsolete.
But for wilderness and nautical applications, what good is a limited range signal going to do you?
I really doubt that even the majority of RISC chips produced are 64 bit, since RISC chips are very popular for embedded computing. Most ARM chips are 32-bit in Arm mode and 16-bit in Thumb mode. The PowerPC 601, 603, 750, 7400 (Altivec) etc architectures are all 32 bit. MIPS chips using ISA I through ISA III are 32 bit, I believe only in ISA IV did any 64 bit instructions get added.
I know the Alpha is 64 bit, but that is an insane chip. I just found out at lunch (can you tell I'm a geek) that it has PAL units on it, too, among other things, emulate certain really usefull VAX instructions. Basically, you can do a sort of "make your own instruction" thing. Many high-end RISC chips produced now are 64 bit, but, as I said, pretty sure its not the majority. Even if it were, it's a dumb thing to have in the article. RISC design in and of itself doesn't make it easier or harder to add 64bit support, although the if you make all your instructions 64 bit you're going to take a MASSIVE code size hit. (And you better at least double the size of your I-cache).
Well, if you don't have to tell a third party to use Plucker, it looks from a reading of the patent that it wouldn't even be infringing. Having state stored by a provider seems to be a key part of the patent.
I'm guessing you read that and came back here to post in indignation. Probably a good thing, if you have high blood pressure. Among other "gems" from the article:
"RISC chips not only process multiple instructions at the same time but also run at 64 bits"
So, which is worse:
Making a few errors and not featuring corrections so prominently
Making so many errors that are so basic that people find out within minutes, but at least you post corrections fast too.
Just a thought, vis a vis michaels article yesterday.
Well, the Prime Minister is correct, the poster is wrong. Tuvalu sold the rights to .tv
two years ago to a company called Idealab for $50 million. So Idealab lost money. Now Verisign is buying it for less than that. Big deal.
"Where exactly did the poster you're replying to bitch about stamp costs?"
One of the things he was "f***ed on" was having to pay for stamps.
"As far as it goes, if I could get reliably delivered mail that wasn't shredded half the time, I'd be glad to pay additional postage. But since the USPS is effectively another government granted monopoly (or is it a goverment agency? Hm, wish they could make up their minds), there's no real competition for letter carrying to force them to be worth a damn."
Well, you could send the documents FedEx. They are competition for the high end, fast, reliable delivery market. The charge a _lot_ more, because they can't just lose money.
"You pay for a stamp to send mail"
And the post office loses money on it.
"You pay for a box to receive mail"
Or receive all your mail at a friends house.
"You subsidize spammy junk mail catalogs because companies get favorable bulk mail rates"
Companies get favorable bulk rates because bulk prebarcoded mail is easier to process. I don't think you "subsidize" it.
"And now you have to veiw adds while you stand in line or check your box"
OK, I hate ads too. But you're already bitching about your stamp costs. How do you want them to make enough money to continue their operations? I barely ever go into a post office--if it bothers you that much, you could probably avoid it too.
Of course, since it was just an encoded mathematical problem, I doubt they even bothered to copyright it. And it was the poster, not the editor, who made the stupid comment. I still would argue that the flambait line should of have been removed, but hey. Too bad you can't mod stories.
The idea sounds interesting, but kinda gimmicky. Especially with a scholarship for speed--with a problem it took 30 minute for _New_Scientist_ to solve (or did I misread something) that seems a bit silly. Now, waiving the application fee for anyone who solves it, that seems a more commensurate prize.
I don't know if it would be all that bad. Remember, this isn't some random plane, this is one specifically retrofitted for high-level Chinese government officials. It's a matter of respect for our spooks to bug it--shows we're taking them seriously. Then their spooks comb the plane, trying to find _all_ the bugs we planted. The Russians did the same thing with the embassy they built for us, I'm sure we did the same for them.
Parent got modded up as funny, but it actually makes a serious good point. One of the people interviewed for the article said that without this hardware copy prevention, "music could become a cottage industry in a few years." Guess what: tough shit. How would the world be a worse place if music was a "cottage industry" run by many small independant companies? Would we really lose anything?
I'm in favor of copyright, generally. I think it's worth it to have people who can spend their entire lives producing entertainment. But that's not what this battle is about. If piracy _actually_ started making it too hard to produce new content, then there would be a public backlash that would fix things with either a cultural or technical solution
This battle is about maintaining record companies and big studios place in the revenue stream. And they are becoming obsolete. This is like professional letter writers (yes, they actually used to exist) lobbying against public education because it would doom there buisness.
It's not that "real time processing does not belong in a macrokernel architechture" it's that "macrokernel architechture does not belong in a hard real time system"
I don't see any problem with making the linux kernel preemptible to be able to make better real time garuntees. Sure, I don't think you'll ever get hard realtime, but that doesn't mean that you won't get benefits from being able to respond to interrupts even when the system is running the kernel.
("hard realtime" -- maximum interrupt latency of xxx nanoseconds. "soft realtime" -- runs fast enough, usually)
Where are all the people from the debate on making all government-funded software open source? Why aren't they in this debate howling because he was let off easy? After all, if state money means "the people" own everything you do, he was stealing from "the people" with this--especially since he wasn't going to share the money if he won.
I think the prosecution of McOwen went overboard. However, if you don't own machines, you shouldn't run software on them without permission. Increasing system load means the system spends less time in power saving modes. It does consume bandwith. Also, and I don't know if this was the case here, it can be damned annoying when not set up properly. When I was working to help administer the computers at a Math department at my college, the sysadmin for general computing stayed logged in and run Seti@home through scripts. Problem was, he didn't do a very good job, and sometimes two or more copies would run at once. They also seemed to take a perceptible amount of time to get off the CPU.
"It would be a logistical hassle to decide what's just 'incidental' and what's primary to the research, too. I imagine that while we allow 'incidental' work to be closed and exploited for private gain, most of it will miraculously be 'incidental.'"
There is a difference between not allowing commercial release and requiring open release. The petition would require researches to openly publish their code as a condition of the grant, just as they are required to publish their research. That is the focus of the principal objections by the author of the opposing paper.
Not allowing closed-source, commercial releases of software developed as part of a government contract is much more limited in scope and more reasonable. It solves the NDA problems--the modification to the propreitary program can't be sold by the original company, but the distrubution can be restricted. Some researchers would object to this, but it does put things on a footing similar to that of most employees. If I, for instance, develop any software using company resources such as computers, I can't distrubute it without a written disclaimer of interest from my company. I think, also, that it would satisfy the opposing essay author.
"If there is no competition, then no grant"
That's exactly what I meant by "constrain" Why shouold research in which the software is incidental not be given a grant just because some modification they make to a propreitary software package won't be released? Qui bono?
"software patents are a perversion the framers would never have assented to"
Along with granting the vote to women, but that's another thread. My example didn't refer to software patents, but to algorithmic patents. I think these fit perfectly with the intent of the framers. An algorithm takes research to develope. It is often hard to duplicate, even if you see the results (unlike most software patents). By making them patentable, the technique used to develope the algorithm is available immediately and the algorithm itself is garunteed to enter the public domain in a limited amount of time. By the by, I agree with you completely on the issue of the long duration of software copyright.
"The right answer is eminently simple. That unclassified research and its fruits (in this case, software) paid for by the people belong to the people. If we're to buy into the intellectual "property" rhetoric of the software industry, then I guess that means that not releasing government funded software into the public domain is piracy and theft!
The right answer is eminently simple. That unclassified research and its fruits (in this case, software) paid for by the people belong to the people. If we're to buy into the intellectual "property" rhetoric of the software industry, then I guess that means that not releasing government funded software into the public domain is piracy and theft!
"The right answer is eminently simple. That unclassified research and its fruits (in this case, software) paid for by the people belong to the people. If we're to buy into the intellectual "property" rhetoric of the software industry, then I guess that means that not releasing government funded software into the public domain is piracy and theft"
The right answer is seldom simple. If we were talking about software developed as one of the primary objectives of a government grant, I might agree with you. But we're not. The issue is software produced in the course of doing other government research. Should the public be granted access to the spare cycles of any government funded machines? Of course not, the logistical hassle far outways the benefits. The same is true in many cases with software, which is why I disagree with petition as currently written.
"I am sure that one can find other examples of this, and I am sure they are equally outrageous"
I'm sure you can "find" other examples too, since you just make them up out of whole cloth. Please state the drug and manufacturer that you are thinking of?
Bullshit. Your and many other posts seem to miss a key point of this debate: The software produced is often incidental to the main purpose of the research. And forcing the public release of the source code does have its downsides.
First, as the opposing essay points out, no software can be liscenced from a commercial vendor and customized to suit the needs of the research project, now this can be done with an NDA.
Second, this produces a "tainting" effect from public money vis a vis research rights. Say, while working on a partly government-funded climate modeling project, I come up with an improved algorithm. If this proposal became law, I would not use that algorithm in the final software because then I would lose the ability to patent it. If lawsuits started to arise, it would be even worse for everyone but the lawyers: Have you heard from people working on stem cell research how hard it is to completely seperate private/public funding?
I understand and agree with the goals of the supporters of this petition. Publication of software under open-source or public domain liscences would be for the public benefit. However, this blanket approach ignores cases where this benefit would be outweighed by slowing the research, the results of which would benefit the public. Instead, government agencies should be encouraged to put language in grants that requires the release of software produced in the course of the project. However, they should be given the discretion to leave out the software provision in cases where they would constrain the research.
Yes, UDP doesn't resend packets. It doesn't need a concept of packet reassembly since IP does it for you. But you've pointed out the hitch " If the UDP-based application doesn't implement [error correction, retries, etc], then you're not going to have much luck transferring large files".
For something like multiplayer gaming, or most voice conferencing apps, retries are not needed. As you pointed out, " If you don't receive a packet on time, it is usually better to just drop it than keep retrying."
The problem comes when you implement an application where latency doesn't matter as much, but not dropping packets does, like file transfer. TCP has a carefully tuned retransmission scheme to prevent it from saturating networks. The IETF is legitimately concerned whenever a protocol that needs reliable delivery uses UDP.
I do think that the article generally was a bit alarmist. However, the concerns it presented were legitimate and I would call it particularly ignorant.
It's not that the energy would be from particles from higher dimensions, but rather that without the extra dimensions the cosmic rays would have insufficent energy to form the black holes in the first place. This is made pretty clear by even a cursory reading of the article, but that seems to doesn't seem to be the norm around here.
Now, a slightly more interesting question is why the extra dimensions would lower the threshold of singularity formation. Do they make space more compressible by giving it more room to flex in or what?
The reason the author's of the article are worried is that SIMPLE, the new instant messaging interconnect protocol, supports UDP as well as TCP. UDP has no congestion control. It is horrible to transfer large files with as most applications end up resending too many packets. This could, in fact, bring backbone segments down.
Of course "Its not the end of the world" The end of the article even said that it wasn't a "showstopper" for SIMPLE. But it is a genuine problem.
The last proposal I remember reading (admitedly, this was Spring '00 when the IMPP group list degenerated into flamewars and I stopped following it) was simply to tag things with username@service or impp://service/username. There's several solutions, I'm not sure which one they're going with but it shouldn't be a big issue.
I know that one company I used to work for had a military contract and one of the items was to garuntee that the data containing parts of the system would be completely melted by X degress at temperature Y. I believe they did something like put thermite and a triggering unit in the case with it.
That, of course, does much more damage than just a chip explosion. I've only seen one exploded chip--the SCSI controller on an MVME2700 (Motorola) blew about a quarter of the area and half the depth out. I was kinda reassured when I looked at my own board and saw that it had a different model chip in that spot. I don't think it could have killed anyone put it certianly could have done some serious damage to an eye if you leaned over to plug the chassis in. (I wish I had the picture handy, put I don't)
"when people click there from a Microsoft Exchange email message, Exchange helpfully gives us the subject line and username."
This bothers me far more than the "poll fixing" Do people realize that this kind of information is leaking all over the net? Is there a way to disable this "feature"?