The last time they took comments, they implemented two exceptions to the DMCA anti-circumvention statute. Based on comments in 2000, they decide to allow the circumvention of content protection in two cases:
1. Compilations consisting of lists of websites blocked by filtering software applications; and
2. Literary works, including computer programs and databases, protected by access control mechanisms that fail to permit access because of malfunction, damage or obsolescence.
These exemptions are in effect from October 28, 2000 to October 28, 2003.
In other words, this is relevant, and not a waste of dissenter's time. However, it is not a replacement for judicial or legislative attempts to overcome a bad law.
This argument is strongly flawed. First, the preservation of art form has little to do with profitability and everything to do with art lover's willingness to preserve those forms. History is full of examples of obscure books, art, and music that have been preserved while more popular (profitable) works fell by the wayside.
Many of these works have been saved by individuals who were not the original producers or copyright holders of the material. When CD players that support a particular flavor of DRM become rare, it will be illegal for me, as an individual to use a device to convert the files to a new format that I can listen to.
Furthermore, the proper way to preserve musical recordings like 78 rpm records is to preserve the means of playing those records. For example. 78 rpm record players are still readily available, they just take some work to find.
If I had a collection of rare audio recordings that happened to be stored on betamax or old 8" floppy disks, I shouldn't have to keep searching for working beta and floppy drives to be able to listen to my recordings over the next 50 years. As an individual, I should be allowed to make copies for my personal use (under Fair Use statutes) so that I can continue to listen to them after the format dies.
As another example, Circuit City's DiVX users had an option that allowed them to "purchase" a DiVX disc -- that is, they could pay a price for unlimited use of a disc. However, in 1999 or 2000 (I don't remember the exact date) a year or so after the service was terminated, the servers that were in charge of permissions were turned off. That means that any DiVX discs that you "purchased" were compeletely unusable. It wouldn't matter if you had 100 working DiVX players, you still can't access the content. Now, imagine that you had some rare, unprofitable music recording that had been made using a similar, failed technology...
Also, while it may be best to listen to analog 78s using 78rpm turntables (as a non-audiophile I'm not as wedded to this, but I'll take your word for it), digitally encoded data won't lose their quality if they are converted, in a lossless manner from technology X to technology Y.
Putting these recordings on P2P networks for anyone to download just denies descendants of the original artists of those recordings their rightful royalties.
That argument is a bit disingenuous in this context. Klawan's comments don't say anything about
A. putting the recordings on P2P networks, or
B. Denying royalties.
His examples focus on personal use and sharing them with his daughter's history class -- cases where Fair Use would traditionally be supported. In addition, his comments also raise concerns that recordings whose copyrights have expired may be rendered inaccessible.
I think what they're actually talking about is digitally signing the original or computing some type of checksum to maintain an auditable chain of custody and to track versions of the data.
So, ideally, an officer takes a photo and it is automatically digitally signed in some form of read only storage. The image and the signature are then transferred to an electronic "vault." Any 'enhancements' would also be signed and stored in the vault. When the case goes to court, the defense is given access to all versions of the picture, and all the images are matched with their signatures to ensure that they haven't been tampered with. This way, the defense can have their own experts evaluate the 'enhancements' that the police made.
In this scenario, you never deal with concerns that encrypted images may be decrypted. You have to have confidence in the vault -- I'm guessing that a physically secured, tamper evident device with easily auditable features could be implemented (e.g, in the same manner as the FBI carnivore machines are secured at ISP sites).
I agree with you. I've propogated a citation error myself. I was reading article A which cited article B (but it didn't include the subtitle of article B in the reference section). I copied down the details and went to the library to get article B. After reading article B, I wrote article C which cited article B. In writing the references, I used the citation as it was written in article A (sans subtitle). Thus, I read article B, found it relevant, and cited it. However, I propogated the error.
These days, there are a number of programs like procite and endnote that manage your citations. If you were to (as I do) type in new references as soon as you hear about them, you could propogate errors inadvertantly.
This doesn't mean that the authors didn't read the papers they cited. It is an interesting finding, as it does show the spread of ideas -- but it doesn't indicate intellectual dishonesty.
That's not what the article says. It points out that wireless insecurities, particularly on corporate networks, pose a security threat -- no surprise there.
Because of all the hoopla about homeland security, people are pointing out that *any* insecurity that allows people to access networks in unauthorized ways can be a vector for Bad People who want to do Bad Things.
The same could be said about critical security problems in networked computers that may be exploited to attack critical networks. I'm sure that federal cybersecurity czar Richard Clarke would say that any insecurity that enables unauthorized network access *may* be a national security threat.
It doesn't say that you can't have a home or office wifi network. It doesn't even say that freely available wifi is a tool of the terrorists. It says, that systems should be secured, and that responsibility lies at many levels (manufacturers, corporate users, etc).
This isn't to say that the government doesn't engage in FUD or that civil rights aren't under attack. But it makes mare sense to fight the real threats to individual liberties.
The problem is a bit more complex than you make it sound. People doing text-to-speech development are smart, and would have jumped on this idea years ago if it were as easy at it sounds at first.
To sound natural, speech has to incorporate prosody and intonation as well as being able to support coarticulation.
Coarticulation refers to the fact that the sound of a phoneme (the smallest unit of linguistic sound) is affected by those that come before and after it.
It is not an easy problem, but there have been some nice advances in concatenative text-to-speech systems. For example here is a pdf about IBM's approach to the problem.
Unfortunately for you, discreet speech is seen as passe by the major players (IBM, L&H, MS). For a long time, continuous speech was seen as the major boundry to widespread acceptance of general purpose dictation software (another boundry was the support of large vocabularies). Eventually, processor power and algorithms evolved to a point that both barriers were overcome and discrete speech (and small vocabs) were left by the wayside.
One byproduct of this was a decrease in voice error correction performance -- Most verbal corrections are single words (e.g., the user selects the misrecognized word, "foo" and repeats the intended word "bar" without any of the coarticulation cues that the continuous recognition engine relies on). The recognition of isolated words by a continuous speech recognizer is inferior to the performance of a discrete system, yet the major software companies removed the discrete recognition engines from their products. (for more on speech errors, see this or this pdf).
Anyway, the use of discrete recognition engines has been essentially abandoned by the major players, and seems to have been relegated to the specialty shops that cater to disabled users. One outcome of this is that there is very little innovation related to discrete speech because it was one of (many) historical barriers to the use of desktop speech reco. I can certainly understand the resistence by the big companies to go back to an "inferior" recognition engine for handheld devices. Most likely, speech reco on the handheld will emerge in a client-server environment with the speech signal (maybe somewhat processed) being sent from the handheld to a server for recognition, and the text being returned to the handheld. We probably won't see a general purpose speech recognition application (as opposed to a limited vocab application) that runs solely on a handheld until continuous processing can be done entirely on the device.
1. This stops working when all new components have DRM built into them.
But let's say you save your old equipment and can access the data, then:
2. Just because *you* may be able to come up with a solution, it doesn't mean that the problem goes away. The fact is, if this allows content providers to hinder the ability of law abiding citizens to exercise their fair use rights, then that's a Bad Thing.
Think of DRM circumvention as if it were spam blocking. Which would be better, a) you block your incoming spam, or b) there is no more spam. From your perspective, option a is fine, but spammers don't care. As long as option b doesn't happen, they are happy.
Content providers don't care if a few techies manage to watch their DVDs on Linux boxes or listen to protected music on unapproved devices. If most people are subjected to the imposed constraints, then they're happy. Just because you can avoid the problem doesn't mean the problem isn't there.
We shouldn't focus solely on avoiding the problems, we should be working on making the problems go away (e.g., supporting legislation that returns our fair use rights).
From the article: After a 5% decline in the sales of recorded music in 2001, the first fall in living memory,.
That statement would be correct if nobody could remember way back to 1997. In those heady days of the Clinton presidency and the dot com boom, the folks at the RIAA reported a 6.5% decrease in annual sales. Back then they didn't have the p2p bogeyman to blame so they laid the blame on retailers streamlining their inventories.
On the whole 'who to blame' angle, I'm amazed that nobody is talking about the role of Clearchannel's radio monopoly on decreased music sales. Before one company dictated that there would be only a handful of radio formats across most major cities, stations were more likely to expand their playlists to include local acts, independent musicians, and songs that local programming personnel liked. Now, playlists are sent down from the home office, and there is more homogeneity among playlists. What does that mean? Fewer new songs get any real airplay, thus giving the listeners of Big Radio fewer unique albums to consider buying...
Back to EMI: The description of their system has so many vague statements that I seriously doubt that this will take off (and we know that EMI never tries to mislead listeners). What listeners want is ease and freedom.
Here's what needs to happen for online music to be profitable for the labels:
1. Record companies have to realize that consumers really don't care who produces or distributes an album. When I go to a record store to by an album, I don't have to know whether it's a BMG or Sony album, I just go to the store and buy it. With these disperate online music services, each with their own catalogs, consumers are supposed to care about these things.
2. Give me the freedom to listen to my music how I want and when I want. Too many of these services offer limited ability to burn CDs or copy to mp3 players. Stop that. I bought the damn music, let me listen to it the way I want. Stop treating your customers like crooks.
It's not that hard. Record executives have a hard time realizing that the music industry is about the artists. Yes, Mr. Exec I'm sure you're a really neat guy, and I know you spend a lot of time doing important things like Bribing radio stations to play your music and engaging in $480,000,000 in price fixing, and I can only imagine how difficult it is to threaten academic researchers. But seriously, you may be getting just a teansy bit greedy and irrational.
There are 250 Million blank CDRs and tapes bought and used this year for copying music in comparison to 213 Million prerecorded audio media. This means the owners are only being paid for 46 per cent of the musical content. For a comparison: In 1998 almost 90% of all audio media was paid for.
Let's take a look at those numbers. I'll use US sales figures, supplied by the RIAA (The European data, provided by IFPI are not freely available to the public).
If we look at US sales in 1998, we see that CD sales had just increased to 847 million units. The folks at EMI claim that this number is 90% of all audio media paid for, so there were 941 million CDs listened to (941=847/.9).
Now, lets look at the numbers for 2002. In the first half of 2002, there have been 398.1 Million units sold (which is down 10% from last year). Let's assume that Christina Agulara and Eminem's new albums do well, and that the second half of 2002 is as good as the first half. This gives us a total of 796.2 million units for 2002. Now, if we assume that these represent only 46% of all purchased audio media, that means that music fans are listening to 1.73 Billion CDs worth of music (1731=796.2/.46).
So, to sum up,according to EMI, music demand has skyrocketed 183% over the last three years (183=1731/941*100).
Who is making all this great music? All I hear is crap on the radio. Why is everyone complaining about the quality of music these days when in fact the public demand for music has nearly doubled in just a few years?
No wonder I'm swarmed by homeless record executives trying to wash my windshield at every intersection. Won't someone think of the Record Executives?!?
Depending on how this is implemented, lossy audo compression techniques used in MP3 or OGG may strip the info. One of the reasons that these formats can get such good compression rates is that they strip a lot of the acoustic information that people can't hear -- which is exactly where these guys are looking to put their signals.
My guess is that someone could come up with a lower bandwidth approach that would remain in the signal after compression. But they'd probably have to tailor their approach to the specific algorithms in a particular MP3/OGG encoder. If someone used a different encoder or the encoder was upgraded, a new solution would need to be created.
Of course if this technology is primarily used for advertising, then people would want to strip the information. If, on the other hand, the data were truly useful to users, there might be an effort to preserve it after compression.
While spammers will undoubtedly continue to refine the content of their messages, one of the strengths of using a Bayesian filter like this is that it uses the user's own spam and non-spam (ham) as the basis for its calculations. This means that messages are categorized not only by whether they contain spammy words, but also whether they contain the hammy words from your own messages. So, even if spammers could refrain from using words like "free" "mortgage" "sluts" and "spam", they probably wouldn't use words that discriminate your own ham from others (e.g., if you are a computer scientist, your mail may include hammy words like "algorithm" "compile" "project" or "stargate" that would help distinguish ham from spam. The challenge to the spammer would then be to target you with spam that looks like *your* ham (which is probably different from the ham of others).
Future systems (assuming faster processors and more HD space) could include semantic analysis (e.g., Latent Semantic Analysis) to do an even better job and go beyond the word level.
As others have mentioned, DivX 5.x isn't the same as the first hacked version.
What worries me is that they're trying to patent it. Have there been any statements about whether they'll start to charge for licensing to open source projects?
It seems that this is much ado about nothing. It seems that it was a case of overzealous cutting and pasting.
According to this response by the alleged spammers: The TRAC site did indeed for a short period of time state "You are currently signed up for news and information." - but that statement was inaccurate and was just a template response that was generated when a person's story was submitted (standard language from a script that was copied).
Don't ascribe to malice that which can be better explained by incompetance...
This story was most likely generated by someone doing PR for LivePerson. Evidence:
It does not link to a real story, instead just says nice things about the company.
It was submitted to plastic.com in nearly the exact same format. Registered users can check the submission cue there. Here's how it is listed:
The voice on the other end of the phone does not tell you to "please listen carefully, as our menu options have changed". E-mail inquiries do not pretend as though they were never sent. More and more companies are finding out that people actually want to interact with other real people, not pseudo-intelligent machines that can't respond to simple things like "Hi, how are you?" Did pseudo-intelligent humans forget something so obvious? Companies like LivePerson help clients from large ISPs to small libraries communicate one-on-one with people via the web. Softroad takes the concept of live help one step further by allowing anyone, anywhere access to their live Internet surfers via SMS, 2-way pager, or other mobile device. There's nothing like human ingenuity when it comes to questions more complex than "what's my balance?" or "what's the weather in Miami?". But are more companies going to listen?
This looks like an example of stealth marketing. It's the latest thing for marketing scum. Check out this article in Time Magazine.
I don't mind a well designed interactive voice respones (IVR) system, it can be much better than some of the mouthbreathers that man customer service lines. The key here is well designed.
If I want to know what time a flight is landing, I would much rather deal with a well designed system (IIRC continental is a good example). As I recall, it has a decent speech reco system with some natural language understanding. You can say "When does flight 541 land in baltimore" and it tells you. This can be better than waiting-for-the-next-available-representative, while being reminded that my call-is-important-and-will-be-answered-in-the-orde r-it-was-received, only to have them complain about their computer being slow, or be difficult to understand.
Give me cold efficiency over warm incompetance any day.
On the other hand, I can't stand poorly designed IVR systems. For example, when I call my credit card company, they ask for my 16-20 digit credit card number (although the GM card only asks for the last 4 digits), and the last 4 digits of my SSN or my zip code. Then, they tell me all about my balance and recent payments (which I didn't ask for). When I finally get transferred to a live rep, the first question they ask.... "May I have your credit card number?"
The technology exists to match the data with the caller. They shouldn't have to ask for my CC# again. Stupid (or cheap) IVR programming (or purchasing) is the culprit.
On the other hand, I'm always pleasantly surprised when they get it right...
From what I can see, you are concerned that corporate funding will prevent web publishing primarily because of the timing issues associated with the two different mediums. You point out that reviewers of journal articles must maintain secrecy during the review process, thus delaying the disclosure date of patentable findings. You also believe that web publishing wouldn't have the same time lag, and instead disclosure date would be when the paper was submitted, rather than accepted.
While I don't necessarily disagree with your conclusion that corporate funding changes the goals of research, and may stifle some types of collaboration, I'm not sure I agree with your assessment of the issues associated with web vs. traditional publishing.
First, just because the publication medium is the internet, it does not mean that uploading is equivalent to publicly disclosing. The peer review process rests heavily on maintaining a strict level of confidentiality. Whether the publication was ultimately eletronic or paper, the level of secrecy can be maintained. That is, a journal (online or not) can maintain a level of secrecy it is comfortable with. Publishing on the web does not have to mean putting a manuscript in a publicly available place for all to see -- you can submit a secure copy to be reviewed. In fact that is what many journals and conferences already do. So, I don't think the timing objection is really an issue.
Secondly, I don't think that corporations are concerned about delaying disclosure before the patent is filed. In fact, it is likely that the corporation will file the initial disclosure before the article is accepted. Especially in competitive arenas, corporations want to demonstrate that they had ideas first. Having worked for a large corporation, I've seen the pressure of filing disclosure papers as soon as possible when it is likely that others are mere weeks or months behind.
While the issues of fame and fortune certainly exist -- it is definitely better to be first in science. I don't know how much that has changed in recent years. Certainly, there is more emphasis on publication and reputation, and these pressures have increased as sciences have grown. But the fact is there are many grand old men of science who are still riding discoveries made in the 50s and 60s.
Yes, a good deal of university research is funded by industry. Yes, some of that research is conducted for the exclusive use of the funders (i.e., not published). However, I would argue that a vast majority of all university research does not suffer from such publication restrictions (in whole or in part).
A good deal of corporate funding for university research does not carry with it any NDA.
The biggest hurdles to net publishing (as others have mentioned) are entrenched publishers and the reputations of journals.
If the problem were corporate gag orders, we wouldn't have the tremendous academic output we have today. The fact is, that there are more articles being submitted than are (or should be) published.
Publishers manage the peer review process by providing a limited number of outlets, each with a perceived level of quality. If you could ever read some of the manuscripts that get rejected by even the best journals, you'd recognize the importance of the peer review filter.
Now, if a researcher finishes a study, and wants to publicize her results (for personal gain, e.g., tenure, and to advance The Cause of Science), she will look to present it in a respected forum. Sure, she could make a bunch of photocopies and mail them out, or she could post them on her personal web site, but to do so would give the appearance that the quality of the work is unknown (See Wolfram for an exception). However, if a bunch of respected peers in her field agree that her work is worthy of publication in a respected journal, people may take notice.
A peer reviewed internet publisher can succeed, but it would have to develop a positive reputation, maintain quality editors and reviewers, and be at least as accessible as paper journals (e.g., be listed in library databases).
One approach is to build such a system from scratch (see this article), another is to transition from existing journals (many of which now offer online versions to subscribers). Each have their strengths and weaknesses.
It is important to note that publishing is pretty expensive, even without the costs of printing and delivering journals.
At a recent conference there was a paper discussing one interesting aspect of this work. They are working on defining a standard remote control markup language to allow clients to render appropriate layouts. The pdf is here.
This reminds me of an off-the-cuff theory I developed when asked why the DOS PC was more prevelant in business than the old Mac.
One common claim is that there were more business apps written for the PC. While I have no doubt that this is true, I think the explanation is more complex.
My theory was that nobody wanted to computerize their company with a plastic box that had a smiley face when you turned it on...
Computers were supposed to be BUSINESS MACHINES. A business machine is a big gray metal box with confusing text display, not a plastic toy with a smiley and pretty windows you can click on...
While business decision makers would love to save money, they are EXTREMELY failure-averse. PLUS, a super low bid can indicate that you don't have the operating expenses associated with a successful, long term service provider. Would you feel as compelled to support their product after only a 5k investment?
Plus, it's possible that the 15k proposal was better, and the 5k one appeared amateurish or didn't impress the decision-makers.
Slashdot Mirror
IANAL, but this exercise is not fruitless.
The last time they took comments, they implemented two exceptions to the DMCA anti-circumvention statute. Based on comments in 2000, they decide to allow the circumvention of content protection in two cases:
1. Compilations consisting of lists of websites blocked by filtering software applications; and
2. Literary works, including computer programs and databases, protected by access control mechanisms that fail to permit access because of malfunction, damage or obsolescence.
These exemptions are in effect from October 28, 2000 to October 28, 2003.
In other words, this is relevant, and not a waste of dissenter's time. However, it is not a replacement for judicial or legislative attempts to overcome a bad law.
Many of these works have been saved by individuals who were not the original producers or copyright holders of the material. When CD players that support a particular flavor of DRM become rare, it will be illegal for me, as an individual to use a device to convert the files to a new format that I can listen to.
Furthermore, the proper way to preserve musical recordings like 78 rpm records is to preserve the means of playing those records. For example. 78 rpm record players are still readily available, they just take some work to find.
If I had a collection of rare audio recordings that happened to be stored on betamax or old 8" floppy disks, I shouldn't have to keep searching for working beta and floppy drives to be able to listen to my recordings over the next 50 years. As an individual, I should be allowed to make copies for my personal use (under Fair Use statutes) so that I can continue to listen to them after the format dies.
As another example, Circuit City's DiVX users had an option that allowed them to "purchase" a DiVX disc -- that is, they could pay a price for unlimited use of a disc. However, in 1999 or 2000 (I don't remember the exact date) a year or so after the service was terminated, the servers that were in charge of permissions were turned off. That means that any DiVX discs that you "purchased" were compeletely unusable. It wouldn't matter if you had 100 working DiVX players, you still can't access the content. Now, imagine that you had some rare, unprofitable music recording that had been made using a similar, failed technology...
Also, while it may be best to listen to analog 78s using 78rpm turntables (as a non-audiophile I'm not as wedded to this, but I'll take your word for it), digitally encoded data won't lose their quality if they are converted, in a lossless manner from technology X to technology Y.
Putting these recordings on P2P networks for anyone to download just denies descendants of the original artists of those recordings their rightful royalties.
That argument is a bit disingenuous in this context. Klawan's comments don't say anything about
- A. putting the recordings on P2P networks, or
- B. Denying royalties.
His examples focus on personal use and sharing them with his daughter's history class -- cases where Fair Use would traditionally be supported. In addition, his comments also raise concerns that recordings whose copyrights have expired may be rendered inaccessible.I think what they're actually talking about is digitally signing the original or computing some type of checksum to maintain an auditable chain of custody and to track versions of the data.
So, ideally, an officer takes a photo and it is automatically digitally signed in some form of read only storage. The image and the signature are then transferred to an electronic "vault." Any 'enhancements' would also be signed and stored in the vault. When the case goes to court, the defense is given access to all versions of the picture, and all the images are matched with their signatures to ensure that they haven't been tampered with. This way, the defense can have their own experts evaluate the 'enhancements' that the police made.
In this scenario, you never deal with concerns that encrypted images may be decrypted. You have to have confidence in the vault -- I'm guessing that a physically secured, tamper evident device with easily auditable features could be implemented (e.g, in the same manner as the FBI carnivore machines are secured at ISP sites).
dupe
I agree with you. I've propogated a citation error myself. I was reading article A which cited article B (but it didn't include the subtitle of article B in the reference section). I copied down the details and went to the library to get article B. After reading article B, I wrote article C which cited article B. In writing the references, I used the citation as it was written in article A (sans subtitle). Thus, I read article B, found it relevant, and cited it. However, I propogated the error.
These days, there are a number of programs like procite and endnote that manage your citations. If you were to (as I do) type in new references as soon as you hear about them, you could propogate errors inadvertantly.
This doesn't mean that the authors didn't read the papers they cited. It is an interesting finding, as it does show the spread of ideas -- but it doesn't indicate intellectual dishonesty.
That's not what the article says. It points out that wireless insecurities, particularly on corporate networks, pose a security threat -- no surprise there.
Because of all the hoopla about homeland security, people are pointing out that *any* insecurity that allows people to access networks in unauthorized ways can be a vector for Bad People who want to do Bad Things.
The same could be said about critical security problems in networked computers that may be exploited to attack critical networks. I'm sure that federal cybersecurity czar Richard Clarke would say that any insecurity that enables unauthorized network access *may* be a national security threat.
It doesn't say that you can't have a home or office wifi network. It doesn't even say that freely available wifi is a tool of the terrorists. It says, that systems should be secured, and that responsibility lies at many levels (manufacturers, corporate users, etc).
This isn't to say that the government doesn't engage in FUD or that civil rights aren't under attack. But it makes mare sense to fight the real threats to individual liberties.
What, there's something beside A*?
The problem is a bit more complex than you make it sound. People doing text-to-speech development are smart, and would have jumped on this idea years ago if it were as easy at it sounds at first.
To sound natural, speech has to incorporate prosody and intonation as well as being able to support coarticulation.
Coarticulation refers to the fact that the sound of a phoneme (the smallest unit of linguistic sound) is affected by those that come before and after it.
It is not an easy problem, but there have been some nice advances in concatenative text-to-speech systems. For example here is a pdf about IBM's approach to the problem.
We're not there yet, but things are improving.
Unfortunately for you, discreet speech is seen as passe by the major players (IBM, L&H, MS). For a long time, continuous speech was seen as the major boundry to widespread acceptance of general purpose dictation software (another boundry was the support of large vocabularies). Eventually, processor power and algorithms evolved to a point that both barriers were overcome and discrete speech (and small vocabs) were left by the wayside.
One byproduct of this was a decrease in voice error correction performance -- Most verbal corrections are single words (e.g., the user selects the misrecognized word, "foo" and repeats the intended word "bar" without any of the coarticulation cues that the continuous recognition engine relies on). The recognition of isolated words by a continuous speech recognizer is inferior to the performance of a discrete system, yet the major software companies removed the discrete recognition engines from their products. (for more on speech errors, see this or this pdf).
Anyway, the use of discrete recognition engines has been essentially abandoned by the major players, and seems to have been relegated to the specialty shops that cater to disabled users. One outcome of this is that there is very little innovation related to discrete speech because it was one of (many) historical barriers to the use of desktop speech reco. I can certainly understand the resistence by the big companies to go back to an "inferior" recognition engine for handheld devices. Most likely, speech reco on the handheld will emerge in a client-server environment with the speech signal (maybe somewhat processed) being sent from the handheld to a server for recognition, and the text being returned to the handheld. We probably won't see a general purpose speech recognition application (as opposed to a limited vocab application) that runs solely on a handheld until continuous processing can be done entirely on the device.
Two problems:
1. This stops working when all new components have DRM built into them.
But let's say you save your old equipment and can access the data, then:
2. Just because *you* may be able to come up with a solution, it doesn't mean that the problem goes away. The fact is, if this allows content providers to hinder the ability of law abiding citizens to exercise their fair use rights, then that's a Bad Thing.
Think of DRM circumvention as if it were spam blocking. Which would be better, a) you block your incoming spam, or b) there is no more spam. From your perspective, option a is fine, but spammers don't care. As long as option b doesn't happen, they are happy.
Content providers don't care if a few techies manage to watch their DVDs on Linux boxes or listen to protected music on unapproved devices. If most people are subjected to the imposed constraints, then they're happy. Just because you can avoid the problem doesn't mean the problem isn't there.
We shouldn't focus solely on avoiding the problems, we should be working on making the problems go away (e.g., supporting legislation that returns our fair use rights).
From the article: After a 5% decline in the sales of recorded music in 2001, the first fall in living memory,.
That statement would be correct if nobody could remember way back to 1997. In those heady days of the Clinton presidency and the dot com boom, the folks at the RIAA reported a 6.5% decrease in annual sales. Back then they didn't have the p2p bogeyman to blame so they laid the blame on retailers streamlining their inventories.
On the whole 'who to blame' angle, I'm amazed that nobody is talking about the role of Clearchannel's radio monopoly on decreased music sales. Before one company dictated that there would be only a handful of radio formats across most major cities, stations were more likely to expand their playlists to include local acts, independent musicians, and songs that local programming personnel liked. Now, playlists are sent down from the home office, and there is more homogeneity among playlists. What does that mean? Fewer new songs get any real airplay, thus giving the listeners of Big Radio fewer unique albums to consider buying...
Back to EMI: The description of their system has so many vague statements that I seriously doubt that this will take off (and we know that EMI never tries to mislead listeners). What listeners want is ease and freedom.
Here's what needs to happen for online music to be profitable for the labels:
1. Record companies have to realize that consumers really don't care who produces or distributes an album. When I go to a record store to by an album, I don't have to know whether it's a BMG or Sony album, I just go to the store and buy it. With these disperate online music services, each with their own catalogs, consumers are supposed to care about these things.
2. Give me the freedom to listen to my music how I want and when I want. Too many of these services offer limited ability to burn CDs or copy to mp3 players. Stop that. I bought the damn music, let me listen to it the way I want. Stop treating your customers like crooks.
It's not that hard. Record executives have a hard time realizing that the music industry is about the artists. Yes, Mr. Exec I'm sure you're a really neat guy, and I know you spend a lot of time doing important things like Bribing radio stations to play your music and engaging in $480,000,000 in price fixing, and I can only imagine how difficult it is to threaten academic researchers. But seriously, you may be getting just a teansy bit greedy and irrational.
Man, I need some sleep...
There are 250 Million blank CDRs and tapes bought and used this year for copying music in comparison to 213 Million prerecorded audio media. This means the owners are only being paid for 46 per cent of the musical content. For a comparison: In 1998 almost 90% of all audio media was paid for.
Let's take a look at those numbers. I'll use US sales figures, supplied by the RIAA (The European data, provided by IFPI are not freely available to the public).
If we look at US sales in 1998, we see that CD sales had just increased to 847 million units. The folks at EMI claim that this number is 90% of all audio media paid for, so there were 941 million CDs listened to (941=847/.9).
Now, lets look at the numbers for 2002. In the first half of 2002, there have been 398.1 Million units sold (which is down 10% from last year). Let's assume that Christina Agulara and Eminem's new albums do well, and that the second half of 2002 is as good as the first half. This gives us a total of 796.2 million units for 2002. Now, if we assume that these represent only 46% of all purchased audio media, that means that music fans are listening to 1.73 Billion CDs worth of music (1731=796.2/.46).
So, to sum up,according to EMI, music demand has skyrocketed 183% over the last three years (183=1731/941*100).
Who is making all this great music? All I hear is crap on the radio. Why is everyone complaining about the quality of music these days when in fact the public demand for music has nearly doubled in just a few years?
No wonder I'm swarmed by homeless record executives trying to wash my windshield at every intersection. Won't someone think of the Record Executives?!?
how will converting it into MP3s affect it?
Depending on how this is implemented, lossy audo compression techniques used in MP3 or OGG may strip the info. One of the reasons that these formats can get such good compression rates is that they strip a lot of the acoustic information that people can't hear -- which is exactly where these guys are looking to put their signals.
My guess is that someone could come up with a lower bandwidth approach that would remain in the signal after compression. But they'd probably have to tailor their approach to the specific algorithms in a particular MP3/OGG encoder. If someone used a different encoder or the encoder was upgraded, a new solution would need to be created.
Of course if this technology is primarily used for advertising, then people would want to strip the information. If, on the other hand, the data were truly useful to users, there might be an effort to preserve it after compression.
While spammers will undoubtedly continue to refine the content of their messages, one of the strengths of using a Bayesian filter like this is that it uses the user's own spam and non-spam (ham) as the basis for its calculations. This means that messages are categorized not only by whether they contain spammy words, but also whether they contain the hammy words from your own messages. So, even if spammers could refrain from using words like "free" "mortgage" "sluts" and "spam", they probably wouldn't use words that discriminate your own ham from others (e.g., if you are a computer scientist, your mail may include hammy words like "algorithm" "compile" "project" or "stargate" that would help distinguish ham from spam. The challenge to the spammer would then be to target you with spam that looks like *your* ham (which is probably different from the ham of others).
Future systems (assuming faster processors and more HD space) could include semantic analysis (e.g., Latent Semantic Analysis) to do an even better job and go beyond the word level.
As others have mentioned, DivX 5.x isn't the same as the first hacked version.
What worries me is that they're trying to patent it. Have there been any statements about whether they'll start to charge for licensing to open source projects?
It seems that the disease that is PanIP has been spreading...
That's why I mail myself empty envelopes once a month...
It seems that this is much ado about nothing. It seems that it was a case of overzealous cutting and pasting.
According to this response by the alleged spammers:
The TRAC site did indeed for a short period of time state "You are currently signed up for news and information." - but that statement was inaccurate and was just a template response that was generated when a person's story was submitted (standard language from a script that was copied).
Don't ascribe to malice that which can be better explained by incompetance...
Evidence:
This looks like an example of stealth marketing. It's the latest thing for marketing scum. Check out this article in Time Magazine.
Marketing bitches...
I don't mind a well designed interactive voice respones (IVR) system, it can be much better than some of the mouthbreathers that man customer service lines. The key here is well designed.
e r-it-was-received, only to have them complain about their computer being slow, or be difficult to understand.
If I want to know what time a flight is landing, I would much rather deal with a well designed system (IIRC continental is a good example). As I recall, it has a decent speech reco system with some natural language understanding. You can say "When does flight 541 land in baltimore" and it tells you. This can be better than waiting-for-the-next-available-representative, while being reminded that my call-is-important-and-will-be-answered-in-the-ord
Give me cold efficiency over warm incompetance any day.
On the other hand, I can't stand poorly designed IVR systems. For example, when I call my credit card company, they ask for my 16-20 digit credit card number (although the GM card only asks for the last 4 digits), and the last 4 digits of my SSN or my zip code. Then, they tell me all about my balance and recent payments (which I didn't ask for). When I finally get transferred to a live rep, the first question they ask.... "May I have your credit card number?"
The technology exists to match the data with the caller. They shouldn't have to ask for my CC# again. Stupid (or cheap) IVR programming (or purchasing) is the culprit.
On the other hand, I'm always pleasantly surprised when they get it right...
From what I can see, you are concerned that corporate funding will prevent web publishing primarily because of the timing issues associated with the two different mediums. You point out that reviewers of journal articles must maintain secrecy during the review process, thus delaying the disclosure date of patentable findings. You also believe that web publishing wouldn't have the same time lag, and instead disclosure date would be when the paper was submitted, rather than accepted.
While I don't necessarily disagree with your conclusion that corporate funding changes the goals of research, and may stifle some types of collaboration, I'm not sure I agree with your assessment of the issues associated with web vs. traditional publishing.
First, just because the publication medium is the internet, it does not mean that uploading is equivalent to publicly disclosing. The peer review process rests heavily on maintaining a strict level of confidentiality. Whether the publication was ultimately eletronic or paper, the level of secrecy can be maintained. That is, a journal (online or not) can maintain a level of secrecy it is comfortable with. Publishing on the web does not have to mean putting a manuscript in a publicly available place for all to see -- you can submit a secure copy to be reviewed. In fact that is what many journals and conferences already do. So, I don't think the timing objection is really an issue.
Secondly, I don't think that corporations are concerned about delaying disclosure before the patent is filed. In fact, it is likely that the corporation will file the initial disclosure before the article is accepted. Especially in competitive arenas, corporations want to demonstrate that they had ideas first. Having worked for a large corporation, I've seen the pressure of filing disclosure papers as soon as possible when it is likely that others are mere weeks or months behind.
While the issues of fame and fortune certainly exist -- it is definitely better to be first in science. I don't know how much that has changed in recent years. Certainly, there is more emphasis on publication and reputation, and these pressures have increased as sciences have grown. But the fact is there are many grand old men of science who are still riding discoveries made in the 50s and 60s.
Yes, a good deal of university research is funded by industry. Yes, some of that research is conducted for the exclusive use of the funders (i.e., not published). However, I would argue that a vast majority of all university research does not suffer from such publication restrictions (in whole or in part).
A good deal of corporate funding for university research does not carry with it any NDA.
The biggest hurdles to net publishing (as others have mentioned) are entrenched publishers and the reputations of journals.
If the problem were corporate gag orders, we wouldn't have the tremendous academic output we have today. The fact is, that there are more articles being submitted than are (or should be) published.
Publishers manage the peer review process by providing a limited number of outlets, each with a perceived level of quality. If you could ever read some of the manuscripts that get rejected by even the best journals, you'd recognize the importance of the peer review filter.
Now, if a researcher finishes a study, and wants to publicize her results (for personal gain, e.g., tenure, and to advance The Cause of Science), she will look to present it in a respected forum. Sure, she could make a bunch of photocopies and mail them out, or she could post them on her personal web site, but to do so would give the appearance that the quality of the work is unknown (See Wolfram for an exception). However, if a bunch of respected peers in her field agree that her work is worthy of publication in a respected journal, people may take notice.
A peer reviewed internet publisher can succeed, but it would have to develop a positive reputation, maintain quality editors and reviewers, and be at least as accessible as paper journals (e.g., be listed in library databases).
One approach is to build such a system from scratch (see this article), another is to transition from existing journals (many of which now offer online versions to subscribers). Each have their strengths and weaknesses.
It is important to note that publishing is pretty expensive, even without the costs of printing and delivering journals.
At a recent conference there was a paper discussing one interesting aspect of this work. They are working on defining a standard remote control markup language to allow clients to render appropriate layouts. The pdf is here.
This could be pretty cool...
This reminds me of an off-the-cuff theory I developed when asked why the DOS PC was more prevelant in business than the old Mac.
One common claim is that there were more business apps written for the PC. While I have no doubt that this is true, I think the explanation is more complex.
My theory was that nobody wanted to computerize their company with a plastic box that had a smiley face when you turned it on...
Computers were supposed to be BUSINESS MACHINES. A business machine is a big gray metal box with confusing text display, not a plastic toy with a smiley and pretty windows you can click on...
While business decision makers would love to save money, they are EXTREMELY failure-averse. PLUS, a super low bid can indicate that you don't have the operating expenses associated with a successful, long term service provider. Would you feel as compelled to support their product after only a 5k investment?
Plus, it's possible that the 15k proposal was better, and the 5k one appeared amateurish or didn't impress the decision-makers.
Who would have thought a question about diamonds of all things would beat out The star fraction
I'm shocked.