This may be an example of some junior attorney attempting to justify his or her retainer to Comedy Central. I'm sure CC has a small army of lawyers for defending IP concerns. But as Adobe learned recently, your overzealous laywers sometime can get you in public relation deep shit. Comedy Central may not even be aware of this exchange.
Unlikely, not just because the creative director is issuing the correspondence. Any trademark/domain name lawyer worth their salt would have grabbed all the domains themselves in the first place.
People need to understand that priority in the DNS registration system in no way provides priority over a US Trademark filing unless you can clearly demonstrate a legitimate use and that there will be no confusion in the marketplace
Not completely accurate. If the guy had been using the name 'in trade' prior to the use by the trademark holder then the trademark holder cannot prevent the prior use.
But the priority is based on the filling date, not the issue date. So the fact that the trademark was filled before the domain name was registered means that the corporation wins.
You must know that most long distance telephone is digital these days and that the standard bandwidth for a voice line is 64Kbits/sec. That is why 56Kb modems are the end of the road.
If these folks really did have a way to send more than 28.8Kb/sec over a modem connection they would be selling improved modems.
I simply don't believe your claims as to how modems work. All the modem drivers I have used involve a mapping to a serial port. In the old days they used to be a separate box connected to the com port. The PC is quite definitely not doing processing for the modem. Modems that are Windows only are very unusual.
It is possible that the way the demo is cooked is to use a programmable modem of some sort. But there certainly isn't any way to get more than 28.8 kbits/sec of data to the vast majority users of legacy modems.
It strikes me that the cable companies will have a great opportunity o take over @Home's infrastructure for cents on the dolar.
The major cost of network infrastructure is equipment which depreciates over months not years and deployment which is a sunk cost. Having laid out several billion to build the network @Home's creditors would probably be lucky to get a few tens of million back in a liquidation.
Only companies that can use the infrastructure where it is are the existing cable cos. So my guess is that @Home will end up being bought out by the cable cos and split up amongst the cable companies with Excite being flogged off to one of the Internet companies that makes money.
I have a feeling NASA doesn't use outlook for email, or run unpatched systems, even if they did use IIS.
A professional company doesn't use microsoft to run everything, they know better.
"I believe it must be true so I'll post it to slashdot even though I have no idea what the truth of the matter is"
NASA has for many years been failing to deploy desktop security from a Canadian company called Entrust. That means that their security infrastructure has to be running on W2K machines 'cos thats all Entrust support.
I find it interesting that the company I work for strips out Sircam virus using a plug in to their Exchange server while the MIT AI lab where I have a courtesy account still hasn't put a patch into their sendmail running on Slowlaris.
I still get about 200 Sircam messages a day on my AI account. Not a problem if I have a high bandwidth connection but my account is now unusable from a dialup modem.
The reason that most viruses attack Outlook is that it is much easier to access the outlook address book with a couple of lines of VB script than to parse the headers in the mail spool.
It HAS to have a larger HD in it... MPEG2 CAN'T be 4x more efficient than MPEG1
Actually it can because the MPEG2 stream from dish networks has several advantages. First the input signal is much cleaner than the signal you get after transaling the pictures into NTSC, bouncing them through the ionosphere and demodulating them with a cheap circuit. Second the compression hardware they are using is very much more powerful. Third they are using a better algorithm
What that amounts to is much higher compression for at least the same end quality.
I Just wonder when I can get me DishNetwork version of the same.
Just because he's using a modem doesn't mean that he's actually transmitting digital data over the phone line. What sort of video compression can be achieved when you don't need (or get) bit-perfect transmission, but rather encode video properties directly in the analog signal?
Several reasons
The article stated it was a 28.8 modem which by definition takes a digital input from one end (the computer) and an analogue one at the other (the phone line).
The phone system is digital from the point where the line to the house hits the exchange.
Even when the system was analog there were multiplexing schemes in place to get the maximum out of the available signal cable. Phase division, time division, frequency division you name it. Long distance lines have not been simple wire connections since the earliest days of the telegraph.
Either you misunderstand or it's a subtle troll...
Actually latency does throttle bandwidth in modems and the effect can reduce throughput by several kbits/sec. There is an interaction between the various layers of the protocol stack.
The most extreeme example of this effect is the Mobitex network used by RIM and PalmVI devices. Although the claimed bandwidth is 9.5Kb/sec the actual is no more than 50 bits/sec because there is a long delay between sending one packet and getting the ack that allows you to send the next.
TCP/IP over PPP is nowhere near as bad as Mobitex but there is a significant effect.
MPEG works by sending a stream of key frames interspersed with a number of delta frames.
Persistence of vision becomes really flakey at under 25 frames per second. With the overhead of stop bits, start bits, PPP protocol etc 28.8Kbits/sec is actually more like 22,000 bits/sec. That means that there are less than 900 bits to encode the delta between one frame and the next.
There might be something to be had out of using second order derivatives, a delta encoding of the delta encodings. There might be something to be had out of more powerfull delta encoding techniques, more complex transformations from one piece of screen to the next.
However the law of diminishing returns applies here and however good the delta encoding is, there is still the need to send key frames from time to time. At the very minimum once per scene change. In practice very much more often. It is quite likely that a scheme substantially better than MPEG is possible, but the scheme claimed is just too close to the fundamental limits.
There are two ways to cook a compression demo. The first is to pre-load the cached data, the second is to chose the content to be compressed very carefully. For example Larry King Live compresses quite well because the video shows only two talking heads from fixed camera angles. Star Trek TNG would be much harder because the camera is often moving.
Einstein reported that he was often acosted by people who would say something like 'how do we get to the next solar system if we can't go faster than the speed of light?', to which he would reply 'I don't set the laws of physics, I am just telling you what they are'.
Seems to me that the reason that so many people invested so much in Pixelon was that they believed that because they needed the solution so baddly, it had to exist, even if Shanon's law dictated otherwise.
Similar thinking runs rampant in the GOP mania for ABM technology. There has not been a single successful test that has not been cooked, in their last test the target had a radio beacon sending out its GPS measured position to the interceptor. But because they want to believe in the technology they will believe their own cooked figures and threaten MIT Professors that try to tell them they are being had with jail.
That's the scary part of the Hague treaty. If it's passed, the foreign verdict CAN be enforced in your home country.
If you are living in another Common law jurisdiction (Canada, UK, etc.) then getting the judgement enforced is possible but not exactly easy. The exception is the US where the courts routinely refuse enforcement of foreign libel judgements, particularly those originating in the UK.
Not necessarily. We deposed Noriega, and took him to within our borders, tried, convicted, and sentenced him to prison.
Noriega was actually 'extradited'. The extradition process was somewhat suspect since the US had invaded the country in question but the forms of legal behavior were followed even though the spirit was completely ignored.
Is that really what developers had at NCSA had in mind?
No, making a quick buck out of the work done at CERN was Marc's game.
The absolute level of counter-cultural and off the wall stuff on the Web has never been higher. The fact that the Web now has a vast mainstream audience as well does not dilute the subversive use.
The Web was about political transformations, but not necessarily of the type the trite journalists of the Sunday NYT think. We had a Web service in Sarajevo in 1992 during the seige.
Have you ever read the DMCA? Why do you think that instead of being forced to pay millions of dollars to the RIAA Napster has only been forced to remove copyright infringing content?
You clearly haven't. Napster has not been forced to pay anything yet because the case has not yet come to trial. The judge ruled that the RIAA was likely to win on the merits and was entitled to a temporary injunction pending judgement.
If people cannot sue gun companies for what people do with guns, then I sincerely believe this lawsuit will be fighting an uphill battle trying to sue mp3.com for what other people did with their downloaded files.
The reason that people cannot sue makers of guns over the use to which they are put is that the gun lobby got bills passed in several states to block them.
This is not a manufacturing defect case however. It is a claim over the consequences of a scheme that a court has already found to be a breach of copyright law. Even though the action was eventually settled out of court the precedent has been established because the judge issued a rulling on the case. In this case it is pretty difficult for MP3.com to escape the precedent because it directly refered to their own conduct.
The difficulty here will be in assesing the extent to which the MP3.com scheme helped people put tracks on Napster. I suspect that it is very difficult to quantify, but that does not help Mp3.com all that much much because they set up the idiot service.
I am reposting this 'cos the slashcode appears to have lost the original.
Modulo the perhaps-conscious reference to St. Anselm, your story starts late.
I was posting an explanation of why philosophers regard the AI use of the term 'Ontology' incredibly pretentious. I did not claim to give an entire history of the field. If you want that you can go to Amazon and buy my history of hermeneutics when I have finished it.
Ontology has a history that goes back through Hegel and Husserl,
Getting confused about Heidegger is excusable, the guy himself got confused. He thought that the third reich had appeared to embody his philosophy (they thought different).
The phenomonological is by no means either the exclusive nor the canonical version of the term.
Well it is not any more, but only because the AI field has misappropriated the term. There is no circumstance in which I have read the term 'ontology' in an AI paper where another word would not convery the concept better.
The AI field has arrived at an intersubjective definition of the term 'ontology', it means 'I have read more obscure german philosophers than you have' (the term obscure binding to their litterary style and not their prominence).
In AI terms, "ontology" is simply
Well if it is simple then use the simple language instead. Stop borrowing terms from another field for pretentious purposes.
The "ontology" of a hammer emerges from the act of hammering. We become aware of it when we miss the nail and hit the thumb.
So does your thumb emerge into the being of the hammer?
When I was at the AI lab this sort of argument struck me as an example of the type of behaviour a late friend wrote of concerning a moral dilema involving a cloning machine 'first the lawyers tried redefining murder, re-evaluating it and finaly re-spelling it'.
I still think they would have done much better throwing out all the copies of any german philosopher that begins with an H into a large skip outside the lab and using them to incinerate every last Lisp machine (including the Whitehouse Publications servers).
An Ontology is a system of being, as in 'the ontological proof of the existence of God.
It got wedged into AI theory when a bunch of guys started reading the Hermeneutics litterature and got real, real confused about Heidegger.
In 'Being and Time' there is a hopelessly confused attempt to define being in terms of communication. Until recently the english translation was even more confused because German words for the two types of 'being' Heidegger makes a crucial distinction between are both translated using the same word in English!
To cut a long story short later but for later chaps (Satre, Gadamer, Ricoeur, Habbermas) who rescued the ideas Heidegger would probably have been written off as just another Nazi (the party didn't much like him though, at the end of the war they tried their best to get him shot). Heidegger's radical revision of the theological field of hermeneutics created a new field of philosophy of communications, a key part of which is the concept of a 'shared vocabulary' being essential to communication and hence 'being' and hence an 'ontology'.
So various AI researchers have attempted to apply the gradiose title 'ontology' to a mish mash of concepts in an attempt to convince people that something deep is going on.
Bluetooth is "secure" in that its range is relatively short.
Sounds like security through obscurity to me.
As with all things Internet-Aged, if you want to assuage your paranoia you need to encrypt your data before you send it and not rely on the network to be secure.
My concern is that they are providing an encryption scheme that appears to be based on DIY crypto and I don't know any of the people involved. If it was a Ron Rivest, Phil Rogaway or the like behind the encryption or a Paul Kocher, Matt Blaze or the like behind the protocol I might be less skeptical about the claim 'it ain't been broken it must be secure'.
If they are using a DIY cipher (the specs are hard to search) then the chances are they are doing it because they are concerned about performance which probably means that they aren't doing enough encryption work for safety.
Ciphers like AES candidates represent the 'bleeding edge' in terms of performance, Adi Shamir suggested adding a couple of extra rounds because of that (I disagree but thats because adding rounds could introduce a compromise, I would rather use one of the already defined and tested modes with a larger keysize which uses more rounds with a tested key schedule). So if someone comes along with a cipher that is markedly faster than AES one tends to be concerned about the security.
Lots of folk are focussing on the 802.11b WEP break as a reason to switch to Bluetooth. I wonder how many of those people have looked at the security Bluetooth provides?
It is possible to secure an 802.11b network, just get somebody competent to wrap an IPSEC VPN arround it.
I am just scanning through the Bluetooth documents, I do not see the tern 'AES' or 'RC4' or any other cipher I am familiar with in the acronyms, I do see the acronym LFSR however. Looks to me like they are using a Linear feedback shift register. If so my guess is that it will be lucky to survive three months of serious analysis.
I don't see the type of security architecture in Bluetooth that would be needed to support their applications securely. The 'Security Architecture' document appears to be one long explanation of why they are not providing any.
People should not take the lack of exploits of Bluetooth to indicate that it must be secure. People only started to look at 802.11b security after the devices went on sale branded 'secure'. If somebody wants my input at the design stage they have to pay me for it. If I am going to work for free I want to at least get publicity in return. Breaking a prototype specification does not create publcity and generate consulting gigs.
I don't buy the argument that Bluetooth is designed to serve a different market to 802.11b. A general purpose LAN will serve any general purpose, end of story.
The best idea the Bluetooth types have come up with for a killer application to date is allowing my laptop to talk to my cell phone. If I want my laptop making G3 wireless data calls I will get it a PCMCIA card to do just that. I don't want to buy a $300 bluetooth card and a new $500 cell phone. In Europe the standard cellphone contracts now allow multiple phones per household by default. That pricing model will apear in the US if G3 or GPRS are to take off.
If my wireless keyboard or mouse offendeth my 802.11b network I will cut them off.
Isn't it striking that people who claim to be members of a group advocating free thought and speech would be so anal and vitrolic about everyone who doesn't call Linux GNU/Linux?
The term 'Thought Police' implies that RMS is engaged in monitoring peoples behavior and punishing people for deviation. This is odd since the distinctive fact about RMS is that he manages to be out of touch with the hacker community beyond the Switzerland floor of 545 tech square to breathtaking degree.
RMS does not use any non-free software and is quite likely to reject pieces of software where he does not like the license conditions. Back in 1997 he still had not used the Web.
Linux uses a lot of GNU code, however it makes no attempt to hide the fact. It is not like NCSA's attempt to hijack the Web where CERN's libwww was used without attribution and the original documentation had no mention of CERN or the phrase World Wide Web.
The point about Linus is that he delivered more than just the kernel. He got together a complete bootable image that people could install. GNU did not lack a kernel, they lacked a loader. They had the CMU MACH kernel but they could not get it together to write a loader.
At the time the GNU libraries and applications were considerably less robust than they are today. The hacker base was much smaller and fewer patches were submitted. With the exception of emacs you would not use the GNU tools for additional features although you would use gcc because it saved you $1000+. Many of the other tools simply did not get enough use for the bugs to be eliminated. The only reason to use them was religion since the UNIX workstation you used already included them.
Until Linus came along and provided a bootable image that would run on PC hardware that is. Once you had an O/S that did not come with a native version of make it was pretty important that gmake worked well. People had always used gmake (if you wanted to build emacs it was advisable) but there was no reason to run the gnu version of most of the UNIX utilities.
What it comes down to is vision. Linus understood that if you put the basic framework in place unstructured community action would complete it. RMS wanted to be in control of everything and there was no reason why the kernel was any more objectionable than the applications he was using. In fact it was probably less objectionable because he wasn't so aware of it.
I used OpenGL and PHIGS about ten years ago, there were pretty obvious problems with both but they did the job for their intended market - presenting 3D visualization data.
The idea that games programs are 3d therefore the game designers should be flocking to "Open" GL is ridiculous. In the first place OpenGL is still effectively controlled by SGI so the only difference between 'Open'GL and DirectX is that one is developed by a dying company and the other is developed by a highly profitable one.
Giving OpenGL over to a committee is not a solution either. The hardware vendors are not going to wait for a standard to be established before they sell the hardware to expoit it.
Benchmarks of OpenGL vs DirectX are almost certainly meaningless. Even if someone was to implement a large complex application in both the performance is going to be dominated by the performance of the drivers. The idea that the hardware manufacturers are going to spend more time and effort optimizing their OpenGL drivers than the DirectX drivers most of the games will be played on is pretty naive.
There are some MacLoonies out there who believe that their machines are still faster than PCs, it was true five years ago so it must still be true, after all they have these pretty case colors.
Slashdot Mirror
Unlikely, not just because the creative director is issuing the correspondence. Any trademark/domain name lawyer worth their salt would have grabbed all the domains themselves in the first place.
That does not make it true. The registration was granted after the DNS registration, it was filed much earlier in 1999.
READ PEOPLE!
Not a bad strategy to try yourself sometime.
Not completely accurate. If the guy had been using the name 'in trade' prior to the use by the trademark holder then the trademark holder cannot prevent the prior use.
But the priority is based on the filling date, not the issue date. So the fact that the trademark was filled before the domain name was registered means that the corporation wins.
You must know that most long distance telephone is digital these days and that the standard bandwidth for a voice line is 64Kbits/sec. That is why 56Kb modems are the end of the road.
If these folks really did have a way to send more than 28.8Kb/sec over a modem connection they would be selling improved modems.
I simply don't believe your claims as to how modems work. All the modem drivers I have used involve a mapping to a serial port. In the old days they used to be a separate box connected to the com port. The PC is quite definitely not doing processing for the modem. Modems that are Windows only are very unusual.
It is possible that the way the demo is cooked is to use a programmable modem of some sort. But there certainly isn't any way to get more than 28.8 kbits/sec of data to the vast majority users of legacy modems.
The major cost of network infrastructure is equipment which depreciates over months not years and deployment which is a sunk cost. Having laid out several billion to build the network @Home's creditors would probably be lucky to get a few tens of million back in a liquidation.
Only companies that can use the infrastructure where it is are the existing cable cos. So my guess is that @Home will end up being bought out by the cable cos and split up amongst the cable companies with Excite being flogged off to one of the Internet companies that makes money.
"I believe it must be true so I'll post it to slashdot even though I have no idea what the truth of the matter is"
NASA has for many years been failing to deploy desktop security from a Canadian company called Entrust. That means that their security infrastructure has to be running on W2K machines 'cos thats all Entrust support.
I find it interesting that the company I work for strips out Sircam virus using a plug in to their Exchange server while the MIT AI lab where I have a courtesy account still hasn't put a patch into their sendmail running on Slowlaris.
I still get about 200 Sircam messages a day on my AI account. Not a problem if I have a high bandwidth connection but my account is now unusable from a dialup modem.
The reason that most viruses attack Outlook is that it is much easier to access the outlook address book with a couple of lines of VB script than to parse the headers in the mail spool.
Actually it can because the MPEG2 stream from dish networks has several advantages. First the input signal is much cleaner than the signal you get after transaling the pictures into NTSC, bouncing them through the ionosphere and demodulating them with a cheap circuit. Second the compression hardware they are using is very much more powerful. Third they are using a better algorithm
What that amounts to is much higher compression for at least the same end quality.
I Just wonder when I can get me DishNetwork version of the same.
Several reasons
Actually latency does throttle bandwidth in modems and the effect can reduce throughput by several kbits/sec. There is an interaction between the various layers of the protocol stack.
The most extreeme example of this effect is the Mobitex network used by RIM and PalmVI devices. Although the claimed bandwidth is 9.5Kb/sec the actual is no more than 50 bits/sec because there is a long delay between sending one packet and getting the ack that allows you to send the next.
TCP/IP over PPP is nowhere near as bad as Mobitex but there is a significant effect.
Persistence of vision becomes really flakey at under 25 frames per second. With the overhead of stop bits, start bits, PPP protocol etc 28.8Kbits/sec is actually more like 22,000 bits/sec. That means that there are less than 900 bits to encode the delta between one frame and the next.
There might be something to be had out of using second order derivatives, a delta encoding of the delta encodings. There might be something to be had out of more powerfull delta encoding techniques, more complex transformations from one piece of screen to the next.
However the law of diminishing returns applies here and however good the delta encoding is, there is still the need to send key frames from time to time. At the very minimum once per scene change. In practice very much more often. It is quite likely that a scheme substantially better than MPEG is possible, but the scheme claimed is just too close to the fundamental limits.
There are two ways to cook a compression demo. The first is to pre-load the cached data, the second is to chose the content to be compressed very carefully. For example Larry King Live compresses quite well because the video shows only two talking heads from fixed camera angles. Star Trek TNG would be much harder because the camera is often moving.
Einstein reported that he was often acosted by people who would say something like 'how do we get to the next solar system if we can't go faster than the speed of light?', to which he would reply 'I don't set the laws of physics, I am just telling you what they are'.
Seems to me that the reason that so many people invested so much in Pixelon was that they believed that because they needed the solution so baddly, it had to exist, even if Shanon's law dictated otherwise.
Similar thinking runs rampant in the GOP mania for ABM technology. There has not been a single successful test that has not been cooked, in their last test the target had a radio beacon sending out its GPS measured position to the interceptor. But because they want to believe in the technology they will believe their own cooked figures and threaten MIT Professors that try to tell them they are being had with jail.
Nah, use a Powerbook or one of the newer laptops with 802.11b built in. A PCMCIA card with an antennae hanging out he side would be a dead giveaway.
If you are living in another Common law jurisdiction (Canada, UK, etc.) then getting the judgement enforced is possible but not exactly easy. The exception is the US where the courts routinely refuse enforcement of foreign libel judgements, particularly those originating in the UK.
Noriega was actually 'extradited'. The extradition process was somewhat suspect since the US had invaded the country in question but the forms of legal behavior were followed even though the spirit was completely ignored.
Mind you that is a bit different because there were only nine people with votes that were counted.
No, making a quick buck out of the work done at CERN was Marc's game.
The absolute level of counter-cultural and off the wall stuff on the Web has never been higher. The fact that the Web now has a vast mainstream audience as well does not dilute the subversive use.
The Web was about political transformations, but not necessarily of the type the trite journalists of the Sunday NYT think. We had a Web service in Sarajevo in 1992 during the seige.
You clearly haven't. Napster has not been forced to pay anything yet because the case has not yet come to trial. The judge ruled that the RIAA was likely to win on the merits and was entitled to a temporary injunction pending judgement.
The reason that people cannot sue makers of guns over the use to which they are put is that the gun lobby got bills passed in several states to block them.
This is not a manufacturing defect case however. It is a claim over the consequences of a scheme that a court has already found to be a breach of copyright law. Even though the action was eventually settled out of court the precedent has been established because the judge issued a rulling on the case. In this case it is pretty difficult for MP3.com to escape the precedent because it directly refered to their own conduct.
The difficulty here will be in assesing the extent to which the MP3.com scheme helped people put tracks on Napster. I suspect that it is very difficult to quantify, but that does not help Mp3.com all that much much because they set up the idiot service.
Modulo the perhaps-conscious reference to St. Anselm, your story starts late.
I was posting an explanation of why philosophers regard the AI use of the term 'Ontology' incredibly pretentious. I did not claim to give an entire history of the field. If you want that you can go to Amazon and buy my history of hermeneutics when I have finished it.
Ontology has a history that goes back through Hegel and Husserl,
Getting confused about Heidegger is excusable, the guy himself got confused. He thought that the third reich had appeared to embody his philosophy (they thought different).
The phenomonological is by no means either the exclusive nor the canonical version of the term.
Well it is not any more, but only because the AI field has misappropriated the term. There is no circumstance in which I have read the term 'ontology' in an AI paper where another word would not convery the concept better.
The AI field has arrived at an intersubjective definition of the term 'ontology', it means 'I have read more obscure german philosophers than you have' (the term obscure binding to their litterary style and not their prominence).
In AI terms, "ontology" is simply
Well if it is simple then use the simple language instead. Stop borrowing terms from another field for pretentious purposes.
The "ontology" of a hammer emerges from the act of hammering. We become aware of it when we miss the nail and hit the thumb.
So does your thumb emerge into the being of the hammer?
When I was at the AI lab this sort of argument struck me as an example of the type of behaviour a late friend wrote of concerning a moral dilema involving a cloning machine 'first the lawyers tried redefining murder, re-evaluating it and finaly re-spelling it'.
I still think they would have done much better throwing out all the copies of any german philosopher that begins with an H into a large skip outside the lab and using them to incinerate every last Lisp machine (including the Whitehouse Publications servers).
It got wedged into AI theory when a bunch of guys started reading the Hermeneutics litterature and got real, real confused about Heidegger.
In 'Being and Time' there is a hopelessly confused attempt to define being in terms of communication. Until recently the english translation was even more confused because German words for the two types of 'being' Heidegger makes a crucial distinction between are both translated using the same word in English!
To cut a long story short later but for later chaps (Satre, Gadamer, Ricoeur, Habbermas) who rescued the ideas Heidegger would probably have been written off as just another Nazi (the party didn't much like him though, at the end of the war they tried their best to get him shot). Heidegger's radical revision of the theological field of hermeneutics created a new field of philosophy of communications, a key part of which is the concept of a 'shared vocabulary' being essential to communication and hence 'being' and hence an 'ontology'.
So various AI researchers have attempted to apply the gradiose title 'ontology' to a mish mash of concepts in an attempt to convince people that something deep is going on.
Sounds like security through obscurity to me.
As with all things Internet-Aged, if you want to assuage your paranoia you need to encrypt your data before you send it and not rely on the network to be secure.
My concern is that they are providing an encryption scheme that appears to be based on DIY crypto and I don't know any of the people involved. If it was a Ron Rivest, Phil Rogaway or the like behind the encryption or a Paul Kocher, Matt Blaze or the like behind the protocol I might be less skeptical about the claim 'it ain't been broken it must be secure'.
If they are using a DIY cipher (the specs are hard to search) then the chances are they are doing it because they are concerned about performance which probably means that they aren't doing enough encryption work for safety.
Ciphers like AES candidates represent the 'bleeding edge' in terms of performance, Adi Shamir suggested adding a couple of extra rounds because of that (I disagree but thats because adding rounds could introduce a compromise, I would rather use one of the already defined and tested modes with a larger keysize which uses more rounds with a tested key schedule). So if someone comes along with a cipher that is markedly faster than AES one tends to be concerned about the security.
It is possible to secure an 802.11b network, just get somebody competent to wrap an IPSEC VPN arround it.
I am just scanning through the Bluetooth documents, I do not see the tern 'AES' or 'RC4' or any other cipher I am familiar with in the acronyms, I do see the acronym LFSR however. Looks to me like they are using a Linear feedback shift register. If so my guess is that it will be lucky to survive three months of serious analysis.
I don't see the type of security architecture in Bluetooth that would be needed to support their applications securely. The 'Security Architecture' document appears to be one long explanation of why they are not providing any.
People should not take the lack of exploits of Bluetooth to indicate that it must be secure. People only started to look at 802.11b security after the devices went on sale branded 'secure'. If somebody wants my input at the design stage they have to pay me for it. If I am going to work for free I want to at least get publicity in return. Breaking a prototype specification does not create publcity and generate consulting gigs.
I don't buy the argument that Bluetooth is designed to serve a different market to 802.11b. A general purpose LAN will serve any general purpose, end of story.
The best idea the Bluetooth types have come up with for a killer application to date is allowing my laptop to talk to my cell phone. If I want my laptop making G3 wireless data calls I will get it a PCMCIA card to do just that. I don't want to buy a $300 bluetooth card and a new $500 cell phone. In Europe the standard cellphone contracts now allow multiple phones per household by default. That pricing model will apear in the US if G3 or GPRS are to take off.
If my wireless keyboard or mouse offendeth my 802.11b network I will cut them off.
The term 'Thought Police' implies that RMS is engaged in monitoring peoples behavior and punishing people for deviation. This is odd since the distinctive fact about RMS is that he manages to be out of touch with the hacker community beyond the Switzerland floor of 545 tech square to breathtaking degree.
RMS does not use any non-free software and is quite likely to reject pieces of software where he does not like the license conditions. Back in 1997 he still had not used the Web.
Linux uses a lot of GNU code, however it makes no attempt to hide the fact. It is not like NCSA's attempt to hijack the Web where CERN's libwww was used without attribution and the original documentation had no mention of CERN or the phrase World Wide Web.
The point about Linus is that he delivered more than just the kernel. He got together a complete bootable image that people could install. GNU did not lack a kernel, they lacked a loader. They had the CMU MACH kernel but they could not get it together to write a loader.
At the time the GNU libraries and applications were considerably less robust than they are today. The hacker base was much smaller and fewer patches were submitted. With the exception of emacs you would not use the GNU tools for additional features although you would use gcc because it saved you $1000+. Many of the other tools simply did not get enough use for the bugs to be eliminated. The only reason to use them was religion since the UNIX workstation you used already included them.
Until Linus came along and provided a bootable image that would run on PC hardware that is. Once you had an O/S that did not come with a native version of make it was pretty important that gmake worked well. People had always used gmake (if you wanted to build emacs it was advisable) but there was no reason to run the gnu version of most of the UNIX utilities.
What it comes down to is vision. Linus understood that if you put the basic framework in place unstructured community action would complete it. RMS wanted to be in control of everything and there was no reason why the kernel was any more objectionable than the applications he was using. In fact it was probably less objectionable because he wasn't so aware of it.
Clearly you don't know RMS, he has a genuine psychological aversion to water. Or at least that was the gossip in the MIT AI lab.
My theory is that the AI/LCS types keep him arround because they know that these days they are being seriously out-wierded by the Media lab.
RMS tried to do just that a couple of years back. People told him the name sucked and he changed tack to GNU/Linux.
If the FSF were to actually put together a GNU distribution of Linux they would have a point.
The idea that games programs are 3d therefore the game designers should be flocking to "Open" GL is ridiculous. In the first place OpenGL is still effectively controlled by SGI so the only difference between 'Open'GL and DirectX is that one is developed by a dying company and the other is developed by a highly profitable one.
Giving OpenGL over to a committee is not a solution either. The hardware vendors are not going to wait for a standard to be established before they sell the hardware to expoit it.
Benchmarks of OpenGL vs DirectX are almost certainly meaningless. Even if someone was to implement a large complex application in both the performance is going to be dominated by the performance of the drivers. The idea that the hardware manufacturers are going to spend more time and effort optimizing their OpenGL drivers than the DirectX drivers most of the games will be played on is pretty naive.
There are some MacLoonies out there who believe that their machines are still faster than PCs, it was true five years ago so it must still be true, after all they have these pretty case colors.