Oh, we have an anonymous expert here. So if the frequencies are different, it can never interfere, right? Because everyone knows all band-pass filters are perfect and all electronic components are always perfectly linear... Same thing on a different kind of wave, ever seen these nice ultra-sound speakers that can be used to produce highly-directional audible sounds? They use multiple transducers with very slightly different frequencies... and the non-linearity of the air causes the lower frequencies to appear.
They tested this on Mythbusters and had difficulty getting phones to interfere even in contrived scenarios such as at point blank range, with very old navigation equipment.
The problem with this is that even if you can only repeat the problem once every million flight, it's bound to cause crashes. Oh, and I would expect recent equipment to be more affected, especially stuff like GPS.
I've never heard of any incident where navigation equipment was actually affected by a cell phone in the real world. Wouldn't you think if it were even possible, it would have happened at least once?
There have been many reported incidents, see "Unsafe at any speed?".
It sucks, but why do we need this BS excuse about interference?
How about "because it's not an excuse and we still don't know exactly how bad it is"? Slightly off topic, I became really convinced of how real cellphone interferences are when I realised my (wired) phone at work makes strange noises every time the guy next door receives a call on his mobile.
The two public reasons, interference with other systems, could easily be tested, but neither the FAA nor the FCC manage to do such testing.
Actually, there was an article in IEEE Spectrum about a year ago on interference testing in planes: "Unsafe at any airspeed?". They were measuring what's currently happening when people disobey rules (or just use devices that behave strangely). I don't remember the details, but I remember that the conclusion was along the lines of "this looks a bit scary and you definitely don't want to allow cell phones in planes without doing at least a lot more careful studies". Also, their guesstimate (based on frequency of interference-related incidents and typical accident-to-incidents ratios) was that cell phones (and other electronic devices) might have contributed to about one major accident (expected value) every 12 years. That with the current "ban" (that some people ignore).
If "self help" is available for the university when someone hacked their server, why WOULDN'T the courts allow "investigators" working for the MAFIAA to hack into computers to determine if they were "pirating" music or movies?
Cuts both ways (with the general interpretation you make). If they try to break into your box and you didn't do anything wrong, then *you* would then be allowed to break into their machines... Then again, I'm pretty sure there are lots of restrictions.
The difference with mobile phone conversations is that 1) people talk much louder because the line is crappy and 2) the stupidity of the conversation goes up ("Hey, you know what, I'm on an airplane! I don't have anything interesting to say, but I'll say it anyway.").
The whole point of LGPL is that it can be linked to software with any license.
Well, I know the LGPL will allow me to link with any license, but will the GPL allow me? My understanding of the GPL is that everything you link with it needs to be GPL-compatible. Considering that the GPLv2 software cannot be relicensed as GPLv3 (unless the author agrees) and the GPLv3 software cannot be relicensed as GPLv2, how is it possible to link with a different version of the LGPL?
The license on the C library would be LGPL3, which isn't harmful to anyone but Novell as far as I can tell.
Do you mean GPLv2 software can't link with an LGPLv3 version of libc? I actually think that could be really bad. I'm sure Novell's not the only one with GPLv2 code out there. In general, it's one thing that really annoys me about the new draft process. Nobody's mentioning anything about the LGPL. Is LGPLv3 compatible with GPLv2? Is LGPLv2 compatible with GPLv3? Things like that are *really* important IMO.
Actually, I've also been quite surprised to read that. On the other hand, this is hardly the first time Linus public says he's changing his mind on some issue. In this case, I guess the main reason is not that much him changing his opinion but the FSF coming up with a better (in his opinion at least) draft.
Actually, they won't even bother faking it. They'll just cross the border at one of the many border points that have no officer there. Or they'll cross in the middle of the woods, or by crossing a lake/river, or through an indian reservation... I still haven't figured out why they're pushing these stupid measures. It's bad for the US even economically since Americans can still easily enter Canada, but Canadians have a harder time spending their money in the US. Then again, I'm Canadian, so what do I care...
Although journal referees do not get paid, there are plenty of costs involved in peer review. Articles have to be vetted before being sent out to referee, referees have to be found, and chased, editorial decisions have to made, and the entire process has to be administered.
This is done by associate editors, who get paid very little, if at all.
The peer review process costs, for major journals, millions of dollars a year.
Care to detail where these millions go?
Then there are the costs for editing (good scientist != good writer), proof reading and sub-editing (there are always edits, and these have to be approved etc., which slows down the process and adds to admin. costs as well), and typesetting (which has to be done regardless of the format the document was submitted in). This is not a cheap process by any means.
I haven't seen much editing/typesetting on the papers I've published. Even then, this only applies to journals. Conference proceedings are (in my field) at least as important and yet they are "published" with zero editing. They just take the author's manuscript and add the page number. Yet, they'll charge $10+ to get a copy. Even for journals, I'm willing to live with non-edited papers if it means I can actually have access to it.
Given the speed of all this change, you can see why many publishers are worried about the long-term future, and are hiking up prices now.
If that's indeed the case, I couldn't think of a more stupid move than that -- making sure alternatives appear even more interesting.
Non-commercial open-source-style approaches are being tried, but there are many barriers to their success.
I think the main barrier is journal reputation. Once an "open" journal gets a good reputation in a field (and it has happened at least in physics I'm being told), I don't see how the "old" publishers can compete with that.
There's far too much emphasis placed on fat CVs bulging with papers that no one will ever read.
Actually, there's an increasing emphasis on the number of citations you get on your publications. Making the paper freely available online has been shown (by someone from Google, but can't find the reference) to increase citation rates dramatically.
And seriously, on some academic's web pages the first thing you'll read is about some Prof's 200 or so publications.
These are generally papers written by students. If the prof's been around for a while, it makes sense that he's co-authored hundreds of papers with his students.
Reduce the emphasis on quantity then reviewers will be happier and journals will be less prone to screw around.
Not sure what that would change for journals. What I think would be interesting to emphasise is short (letter-type) papers where researchers can make public minor, but useful results, without the overhead of normal publishing.
Did you know that when an academic writes a paper, to get it published, they have to surrender the copyright to the academic journal? After that, they can't even give copies away. If someone wants to see it, they're supposed to point them to the journal publisher where they can "buy" reprints.
Actually, most publishers (but not all) allow you to publish on your website the accepted version of your paper. What you can't publish is the edited version that appears in the journal. That's what I do for everything I publish (see my web page). The main advantage of doing that for the authors (outside of altruism) is that you get cited more often, which also counts in your record.
On the plus side, there are emerging journals that have an open access policy and I'm considering one of them for the next paper I submit.
(note, I'm talking about scientific journals, like the IEEE ones)
Editing is hard work. Maintaining a consistently high quality of writing, articles that are appropriately in-depth but accessible to the readership, sniffing out the studies that define or redefine the field.
The writing is actually done by authors -- who get no monetary compensation.
Typsesetting can be a misery when working with formulas & like content that has gone through several cycles of review & fine-tuning. Journals shouldn't read like ransom notes.
Most authors submit LaTeX, which is what the journals use I believe.
Reviewers do cost. Finding them, vetting them, coordinating them.
No they don't. I've so far reviewed dozens of paper and still haven't received anything. Not that I'm expecting a compensation, just saying the reviewers aren't being paid (they couldn't afford to pay them anyway).
Illustrations are worth a thousand words, but a consistently good technical illustrator is a rare bird to be treasured.
Except they don't make the illustrations, the authors provide them. Worse, you send them a nice, clean vector figure (eps) and all they do is convert it to a raster image.
Fact-checking, background-reviews, identifying possible conflicts-of-interest, that's a lot of hard-work administrivia that is expected now.
Facts are checked by the reviewers. Conflicts-of-interest are generally not handled, or if they are, it's often post-publication.
Then there are the basic internal administrative costs of keeping the lights on, payroll met, licensing the typefaces, getting the parking lot snowplowed, the PCs virus-free, handling the morass of profit/non-profit taxes & exemptions, all are yet more staff.
That's about the only real cost here, but it can't explain the exorbitant fees for journals.
Subscriber services is everyone's horror. What do you do when a professor or researcher passes out their personal subscription password to everyone, and suddenly you've got 60 sites around the world using that password? Or when Harvard wants a campus-wide subscription, but has several dozen domains folks will be coming in from, not to mention home users?
Maybe the reason people share access is because it's so damn expensive in the first place. My current employer has a subscription to IEEE (and other) journals. If it weren't for that, I'd have to (theoretically) pay 30$ every time there's a journal paper I'd like to look at, not even knowing whether it's useful! It's just ridiculous.
And printing on dead-trees is an expensive proposition, but still the media-of-record. In-house the press is easily a million dollars, not to mention paper, ink, staff, space, insurance, maintenance, distribution, capitol depreciation, etc. Reprints can earn top dollar but those require quality printing and must be accounted for.
In fine if they charge for paper copies. The libraries that want those can pay for that. I just want electronic access, which costs nearly nothing.
Blithely thinking this can all be replaced with a few emails and a database is probably woefully optimistic. Doubtless there is room for journals produced thus, but ones with an active editorial process and rather richer content are probably around for while too; their ecological niche is still a valuable one to their communities.
The most valuable parts of the process (authoring and reviewing) are already done for free. I don't think the associate editors get paid either, so I strongly believe an open process is now possible with just a bit of funding (same kind of funding as many open-source projects get).
Actually, I suspect the system could actually be used to *cool* the passenger compartment in summer. After all, when air comes out of the tank, it's no different than what's happening in a compressor.
Yeah right! Next thing you know people will stop killing each other and have sex (and children). We can't allow that to happen! Won't anybody think of the children???
How secret do you think the steganography algorithm used will be? If not the actual steganography used how about the algorithms used to detect and interpret the hidden data...
The algorithm doesn't need to be secret. If I add a small (inaudible) pseudo-random sequence to a file (for the whole duration), I can use a correlation (with that same pseudo-random sequence that I know) to detect it in the file. Even if the file get encoded/decoded several times, some of it will be left because there's a very small amount of information embedded in a very long signal (it's all covered by Shannon's work and information theory). Now, there are ways to attack this (e.g. screwing up timing and phase), but you need to specifically aim to destroy the watermark and have an idea what you're against.
I don't think it's that robust overall, but this kind of stuff is usually robust to compression, even with heavy losses. The idea is that the message is only a few bits, so it's embedded with a *huge* amount of redundency. To get rid of a watermark, you generally have to specifically attack it, not just throw in noise/compression/whatever blindly.
Come and get me, no DMCA in THIS country. Let's see, which movie should I download tonight?
Do you think they really care if you download a movie? Of course, they pretend to, but in the end it just helps them 1) spread their movies and 2) claim that everyone's a pirate and they're losing 100 trillion dollars due to piracy. Go watch an independent movie instead.
To me, you and the rest here on slashdot that makes perfect sense. To the average user, they'll be upset there were sold, in their minds, a $1000+ paperweight.
If I'm not mistaken, the Dell survey was about "What do *you* want?", not "What do you this the average user wants?". They hoping to know the latter by collecting lots of the former. This is generally more reliable...
Slashdot Mirror
Oh, we have an anonymous expert here. So if the frequencies are different, it can never interfere, right? Because everyone knows all band-pass filters are perfect and all electronic components are always perfectly linear... Same thing on a different kind of wave, ever seen these nice ultra-sound speakers that can be used to produce highly-directional audible sounds? They use multiple transducers with very slightly different frequencies ... and the non-linearity of the air causes the lower frequencies to appear.
They tested this on Mythbusters and had difficulty getting phones to interfere even in contrived scenarios such as at point blank range, with very old navigation equipment.
The problem with this is that even if you can only repeat the problem once every million flight, it's bound to cause crashes. Oh, and I would expect recent equipment to be more affected, especially stuff like GPS.
I've never heard of any incident where navigation equipment was actually affected by a cell phone in the real world. Wouldn't you think if it were even possible, it would have happened at least once?
There have been many reported incidents, see "Unsafe at any speed?".
It sucks, but why do we need this BS excuse about interference?
How about "because it's not an excuse and we still don't know exactly how bad it is"? Slightly off topic, I became really convinced of how real cellphone interferences are when I realised my (wired) phone at work makes strange noises every time the guy next door receives a call on his mobile.
The two public reasons, interference with other systems, could easily be tested, but neither the FAA nor the FCC manage to do such testing.
Actually, there was an article in IEEE Spectrum about a year ago on interference testing in planes: "Unsafe at any airspeed?". They were measuring what's currently happening when people disobey rules (or just use devices that behave strangely). I don't remember the details, but I remember that the conclusion was along the lines of "this looks a bit scary and you definitely don't want to allow cell phones in planes without doing at least a lot more careful studies". Also, their guesstimate (based on frequency of interference-related incidents and typical accident-to-incidents ratios) was that cell phones (and other electronic devices) might have contributed to about one major accident (expected value) every 12 years. That with the current "ban" (that some people ignore).
If "self help" is available for the university when someone hacked their server, why WOULDN'T the courts allow "investigators" working for the MAFIAA to hack into computers to determine if they were "pirating" music or movies?
Cuts both ways (with the general interpretation you make). If they try to break into your box and you didn't do anything wrong, then *you* would then be allowed to break into their machines... Then again, I'm pretty sure there are lots of restrictions.
The difference with mobile phone conversations is that 1) people talk much louder because the line is crappy and 2) the stupidity of the conversation goes up ("Hey, you know what, I'm on an airplane! I don't have anything interesting to say, but I'll say it anyway.").
The OO.o equation editor is shit, but so is Word's. I've been writing all my technical stuff in LyX for the past 10 years now.
The whole point of LGPL is that it can be linked to software with any license.
Well, I know the LGPL will allow me to link with any license, but will the GPL allow me? My understanding of the GPL is that everything you link with it needs to be GPL-compatible. Considering that the GPLv2 software cannot be relicensed as GPLv3 (unless the author agrees) and the GPLv3 software cannot be relicensed as GPLv2, how is it possible to link with a different version of the LGPL?
The license on the C library would be LGPL3, which isn't harmful to anyone but Novell as far as I can tell.
Do you mean GPLv2 software can't link with an LGPLv3 version of libc? I actually think that could be really bad. I'm sure Novell's not the only one with GPLv2 code out there. In general, it's one thing that really annoys me about the new draft process. Nobody's mentioning anything about the LGPL. Is LGPLv3 compatible with GPLv2? Is LGPLv2 compatible with GPLv3? Things like that are *really* important IMO.
Actually, I've also been quite surprised to read that. On the other hand, this is hardly the first time Linus public says he's changing his mind on some issue. In this case, I guess the main reason is not that much him changing his opinion but the FSF coming up with a better (in his opinion at least) draft.
Exactly. And that's why anarchy is a total utopia.
Actually, they won't even bother faking it. They'll just cross the border at one of the many border points that have no officer there. Or they'll cross in the middle of the woods, or by crossing a lake/river, or through an indian reservation... I still haven't figured out why they're pushing these stupid measures. It's bad for the US even economically since Americans can still easily enter Canada, but Canadians have a harder time spending their money in the US. Then again, I'm Canadian, so what do I care...
And your alternative is? Anarchy?
Thanks for the link! I think that's it.
But wasn't Seltzer acting contrary to the law to begin with?
No. Ever heard of fair use?
Although journal referees do not get paid, there are plenty of costs involved in peer review. Articles have to be vetted before being sent out to referee, referees have to be found, and chased, editorial decisions have to made, and the entire process has to be administered.
This is done by associate editors, who get paid very little, if at all.
The peer review process costs, for major journals, millions of dollars a year.
Care to detail where these millions go?
Then there are the costs for editing (good scientist != good writer), proof reading and sub-editing (there are always edits, and these have to be approved etc., which slows down the process and adds to admin. costs as well), and typesetting (which has to be done regardless of the format the document was submitted in). This is not a cheap process by any means.
I haven't seen much editing/typesetting on the papers I've published. Even then, this only applies to journals. Conference proceedings are (in my field) at least as important and yet they are "published" with zero editing. They just take the author's manuscript and add the page number. Yet, they'll charge $10+ to get a copy. Even for journals, I'm willing to live with non-edited papers if it means I can actually have access to it.
Given the speed of all this change, you can see why many publishers are worried about the long-term future, and are hiking up prices now.
If that's indeed the case, I couldn't think of a more stupid move than that -- making sure alternatives appear even more interesting.
Non-commercial open-source-style approaches are being tried, but there are many barriers to their success.
I think the main barrier is journal reputation. Once an "open" journal gets a good reputation in a field (and it has happened at least in physics I'm being told), I don't see how the "old" publishers can compete with that.
There's far too much emphasis placed on fat CVs bulging with papers that no one will ever read.
Actually, there's an increasing emphasis on the number of citations you get on your publications. Making the paper freely available online has been shown (by someone from Google, but can't find the reference) to increase citation rates dramatically.
And seriously, on some academic's web pages the first thing you'll read is about some Prof's 200 or so publications.
These are generally papers written by students. If the prof's been around for a while, it makes sense that he's co-authored hundreds of papers with his students.
Reduce the emphasis on quantity then reviewers will be happier and journals will be less prone to screw around.
Not sure what that would change for journals. What I think would be interesting to emphasise is short (letter-type) papers where researchers can make public minor, but useful results, without the overhead of normal publishing.
Did you know that when an academic writes a paper, to get it published, they have to surrender the copyright to the academic journal? After that, they can't even give copies away. If someone wants to see it, they're supposed to point them to the journal publisher where they can "buy" reprints.
Actually, most publishers (but not all) allow you to publish on your website the accepted version of your paper. What you can't publish is the edited version that appears in the journal. That's what I do for everything I publish (see my web page). The main advantage of doing that for the authors (outside of altruism) is that you get cited more often, which also counts in your record.
On the plus side, there are emerging journals that have an open access policy and I'm considering one of them for the next paper I submit.
(note, I'm talking about scientific journals, like the IEEE ones)
Editing is hard work. Maintaining a consistently high quality of writing, articles that are appropriately in-depth but accessible to the readership, sniffing out the studies that define or redefine the field.
The writing is actually done by authors -- who get no monetary compensation.
Typsesetting can be a misery when working with formulas & like content that has gone through several cycles of review & fine-tuning. Journals shouldn't read like ransom notes.
Most authors submit LaTeX, which is what the journals use I believe.
Reviewers do cost. Finding them, vetting them, coordinating them.
No they don't. I've so far reviewed dozens of paper and still haven't received anything. Not that I'm expecting a compensation, just saying the reviewers aren't being paid (they couldn't afford to pay them anyway).
Illustrations are worth a thousand words, but a consistently good technical illustrator is a rare bird to be treasured.
Except they don't make the illustrations, the authors provide them. Worse, you send them a nice, clean vector figure (eps) and all they do is convert it to a raster image.
Fact-checking, background-reviews, identifying possible conflicts-of-interest, that's a lot of hard-work administrivia that is expected now.
Facts are checked by the reviewers. Conflicts-of-interest are generally not handled, or if they are, it's often post-publication.
Then there are the basic internal administrative costs of keeping the lights on, payroll met, licensing the typefaces, getting the parking lot snowplowed, the PCs virus-free, handling the morass of profit/non-profit taxes & exemptions, all are yet more staff.
That's about the only real cost here, but it can't explain the exorbitant fees for journals.
Subscriber services is everyone's horror. What do you do when a professor or researcher passes out their personal subscription password to everyone, and suddenly you've got 60 sites around the world using that password? Or when Harvard wants a campus-wide subscription, but has several dozen domains folks will be coming in from, not to mention home users?
Maybe the reason people share access is because it's so damn expensive in the first place. My current employer has a subscription to IEEE (and other) journals. If it weren't for that, I'd have to (theoretically) pay 30$ every time there's a journal paper I'd like to look at, not even knowing whether it's useful! It's just ridiculous.
And printing on dead-trees is an expensive proposition, but still the media-of-record. In-house the press is easily a million dollars, not to mention paper, ink, staff, space, insurance, maintenance, distribution, capitol depreciation, etc. Reprints can earn top dollar but those require quality printing and must be accounted for.
In fine if they charge for paper copies. The libraries that want those can pay for that. I just want electronic access, which costs nearly nothing.
Blithely thinking this can all be replaced with a few emails and a database is probably woefully optimistic. Doubtless there is room for journals produced thus, but ones with an active editorial process and rather richer content are probably around for while too; their ecological niche is still a valuable one to their communities.
The most valuable parts of the process (authoring and reviewing) are already done for free. I don't think the associate editors get paid either, so I strongly believe an open process is now possible with just a bit of funding (same kind of funding as many open-source projects get).
would prefer to use an abacus to MS software
On the plus side, when's the last time you heard of an abacus being remotely compromised to start sending spam?
Actually, I suspect the system could actually be used to *cool* the passenger compartment in summer. After all, when air comes out of the tank, it's no different than what's happening in a compressor.
Yeah right! Next thing you know people will stop killing each other and have sex (and children). We can't allow that to happen! Won't anybody think of the children???
How secret do you think the steganography algorithm used will be? If not the actual steganography used how about the algorithms used to detect and interpret the hidden data...
The algorithm doesn't need to be secret. If I add a small (inaudible) pseudo-random sequence to a file (for the whole duration), I can use a correlation (with that same pseudo-random sequence that I know) to detect it in the file. Even if the file get encoded/decoded several times, some of it will be left because there's a very small amount of information embedded in a very long signal (it's all covered by Shannon's work and information theory). Now, there are ways to attack this (e.g. screwing up timing and phase), but you need to specifically aim to destroy the watermark and have an idea what you're against.
I don't think it's that robust overall, but this kind of stuff is usually robust to compression, even with heavy losses. The idea is that the message is only a few bits, so it's embedded with a *huge* amount of redundency. To get rid of a watermark, you generally have to specifically attack it, not just throw in noise/compression/whatever blindly.
Come and get me, no DMCA in THIS country. Let's see, which movie should I download tonight?
Do you think they really care if you download a movie? Of course, they pretend to, but in the end it just helps them 1) spread their movies and 2) claim that everyone's a pirate and they're losing 100 trillion dollars due to piracy. Go watch an independent movie instead.
To me, you and the rest here on slashdot that makes perfect sense. To the average user, they'll be upset there were sold, in their minds, a $1000+ paperweight.
If I'm not mistaken, the Dell survey was about "What do *you* want?", not "What do you this the average user wants?". They hoping to know the latter by collecting lots of the former. This is generally more reliable...