Anyone making a big deal out of this is doing so out of total computer illiteracy or being intellectually dishonest as to their true motive for their outrage.
"The price of freedom is eternal vigilance." - Thomas Jefferson
OMB guidelines are rules on government agencies, and there are organizational penalties for failing to adhere to them. I don't see how we can excuse any agency for failing to follow the rules set out for them, let alone TWO agencies which are involved in espionage. We're not going to see a trial, because the OMB doesn't make laws, but I'm going to write my representatives in Congress and encourage them to issue a new law to codify this OMB guideline - that way, if they DO try it again, the consequences will be much more severe.
If I had evidence, I wouldn't be writing on Slashdot. ^_^ But it worries me that citizens can't observe electronic voting elections in the same way we could have observed paper ballot elections. With paper ballots, there were written instructions on how to count ballots; the equivalent instructions for e-Voting are the source code, which we haven't seen from Diebold. Because of that, I am VERY CONCERNED by a Government contractor which promised to deliver the election to the incumbent President when the election results hinged on the contractor's home state in districts that used the contractor's products.
For a criminal investigation, we would have to prove that Diebold (or someone) acted illegally. I doubt that's going to happen. However, we have the opportunity to make source code availability a condition of ongoing government contracts for e-Voting equipment. This will allow us to have more trust in future elections that use e-Voting equipment. But the only way I will feel convinced that there wasn't foul play in the 2004 Presidential Election is to have Diebold release their source code.
Does the Windows source code matter in this case? No, not really, but it sure gives Diebold a great excuse to stop seeking certification in North Carolina, the only state I am aware of with a code escrow requirement. I do plan, however, on asking my state representatives to enact similar legislation in my home state. By allowing Diebold to slip out of North Carolina, we may be allowing them to slip away with code that could answer questions from so many about the last election.
Because when they want you to submit the source code for "everything" needed to run your app, you can blame Microsoft and avoid any questions about the code that you did get away with using for one election -- an election which was "delivered to Mr. Bush" as promised by Diebold executives.
Did Microsoft finally encounter a state that enacted privacy laws that they find cumbersome, and therefore hope that new Federal laws would be enacted without those same cumbersome issues? Or maybe someone at Microsoft thinks that Federal regulation, through a new set of privacy rules, would serve as an effective barrier against competitors in some of their markets?
And to top it off, the submitter's name links to NotePage, which operates the FeedForAll site as well. And yet, no "conflict of interest" warning from the submitter.
The Fourth Amendment limits the powers of the Government (aka, the Police), not of private individuals. The decision on appeal is that the Police erred in not obtaining a warrant before following up the lead received from the Technician.
This is quite correct, in the limited scope - Police should not be allowed to search through data on your computer without a warrant, as this data is as private as papers in a closed filing cabinet. Even when the cabinet is in transit, a police officer cannot decide to open it except when duly authorized, usually by a warrant. This is a position that needs to be defended, even if the individual at question isn't someone who we feel is entitled to the defense.
On the question of the Tech... well, that's a matter between the customer and the technician. I doubt any laws were broken by the Tech's report to the police - but, then, you would also be open to the "well, prove the data was there BEFORE it came into your hands, Tech" question - the computer didn't come right from the accused's hands, leaving a wide opening for "reasoanble doubt" when/if it came to trial.
How much have you given up to work at your present employer?
Did you agree to random drug tests, at their whim? Perhaps they wanted to see your credit report?
In my experience, companies are constantly trying to gain more concessions from their employees, often without granting anything to the employee in exchange.
It's natural, then, that they're moving on to these genetic test - at least from the company's perspective. Employees, however, are balking at this brand new intrusion for now. But how long until it's just like that drug test that everyone else seems to be OK with, simply because they aren't looking for you?
If you don't speak out for others, no one will be left to speak out for you. This is why Unions are still a good thing - it allows workers to speak up against policies such as this while protecting themselves from direct retribution at work, since the company doesn't know who, exactly, started the complaint.
How is Squid even related here? Squid is a web proxy cache, not a dynamic page generator, so a Squid server would not qualify as a "page server" for this specific patent. If you had a distrubted set of Squid servers, you still would not have the "dynamic generation" capability of the patent.
And if you're supposing that people could un-license GPL'ed code from Epicrealm, remember that such a change could not be done retroactively, and that Epicrealm would be entitled to continue using any GPL-licensed software they already have acquired.
For some reason, I'm left with the feeling that I've missed part of your point...
I was doing this in 1997. The patent is dated 1999.
And the patent application was filed on April 23, 1996. "Prior art" must predate the filing date (04/23/1996), not the date the patent was granted (04/13/1999).
This isn't a patent on dynamic page generation, but about a dynamic server farm where a primary "web server" distributes dynamic page generate requests to one or more "page servers", and where each page server can maintain a cached version of the output of the dynamic page request.
As such, I'm not as concerned about "woe unto all dynamic web sites," but if I managed one that offloads and caches page generation work (i.e., Slashdot, LiveJournal, and probably a lot more) I'd probably be calling my lawyer this morning.
Go do a search for Guardsmark at the NLRB website. I found the full text of this decision very easily, and there is a very important (to me) detail that the linked article omitted: the company in question, which implemented the "no fraternization" rule, provides "security personnel" on contract basis to client companies.
The company, then, has a somewhat reasonable explanation for the need to restrict fraternization - it is a potential security lapse, which would result in a significant loss of reputation for their company.
In this limited role, I cannot help but agree. Especially since the NLRB rules only on the basis of interference with union-organizing activities. As I read the NLRB decision, this wasn't a review for Constitutionality -- only to determine if the rule was designed to curtail or prevent union activity at the company.
Multiple VoIP providers over multiple carriers? Wow, that's a lot of data, but could it lead to bad ratings because of poor carrier performance? Certainly an end-user, espeically a non-technical one, wouldn't care where the fault it, but there could be specific provider/carrier combinations with better reliability. It would be interesting to see if the data identifies any provider/carrier combinations with availability reaching into the 99.9%+ area of POTS.
For instance, I wouldn't be surprised if Verizon VoIP over Verizon DSL was a very reliable combination (assuming, of course, that the DSL installation itself is reliable). But it could be that Verizon VoIP has terrible availability when accessed via any other carrier - because Verizon VoIP might not have an equal-quality interconnect with other IP carriers.
Vertical integration of the service with the carrier - like traditional POTS - allows for integration of the support experience and elimination of interconnect issues which impair reliability. (Just try getting your local loop fixed by the ILEC when there's a fault which causes your non-ILEC DSL service to die frequently.)
Honestly, the only data point which matters to me is what's the reliability of the best provider/carrier combination as compared to POTS, not the "average" reliability of a provider over all carriers.
Since both the NSLU2 and the Mac Mini support USB2, I don't really understand how one would have faster disk I/O than the other. Now, certainly, there can be questions about file system efficiency, but you could also say that the Mac Mini (for, granted, 5x the price) could have much faster I/O, since you have 256MB of RAM instead of 32MB for things like read-caching.
more stable
I'm very interested as to why the NSLU2 would be more stable than a Mac Mini. Personally, I think the short, square design of the Mini is more stable than the tall, tower design of the NSLU2. From a software standpoint, I'd say the two are rather comparable in stability for comparable activities (i.e., no complaining about 3D games crashing on the Mini when the NSLU2 doesn't even have a graphics chip).
more flexible
While the Mini's certainly not a flexibility powerhouse, I don't see the NSLU2 beating it out here, either. Software flexibility? Mac OS X has a lot of software available, both commerical and software libre (GPL/BSD). You can also run GNU/Linux on the Mini, including the Debian distribution with apt-get.
Not to mention the ability to hack it when I want (The warranty is already void!).
Open your Mini and replace the hard drive. Voila, both can be hacked at will because the warranty is void! ^__^
Current iBooks do not have DVI output; they have a "mini-VGA" connector to which an adaptor for Composite or S-Video can be connected. This is the same mini-VGA connector that the iMac G5 provides, so I would not expect to see it replaced with DVI.
While highspeed connections are great, I want to know where this backend bandwidth is coming from and who's paying for it?
You are, of course, by buying the additional services that the phone company wants to sell you for your DSL link, or that the cable company wants to sell you for your cable modem link.
Verizon has thrown a lot of hype into a forthcoming video service. Unlike cable services, though, such Video over DSL services require a fixed amount of bandwidth for each receiver, say 6 Mbps. If you have a 30 Mbps link, you could host 5 receivers (and no other services) or 4 receivers and hae a 6 Mbps downstream ISP connection.
But in this age of DVRs and connected devices, 4 receivers may be insufficient for affluent households with lots of TVs & other video receiving devices. It is because of this that I expect faster connections are going to be deployed, not because we want faster internet.
These are the latest and greatest from Intel and AMD right?
Best performance per watt != Lowest power usage of highest-performing part.
The Pentium M family is much lower power than the Pentium 4, and has reasonably good performance. I don't think AMD really has a chip that competes with the Pentium M, even though AMD's chips are generally less power-hungry than a Pentium 4.
copyright belongs to the photographer I believe. He created the photos.
Subjects in photographs have rights to those images as well, unless they've signed a model release. There are some widely accepted cases where a model release isn't required - such as photographs in the newpaper, if the picture was taken in public (and the individual therefore had no reasonable expectation of privacy).
But there are many places where model releases are required - why else do you think Cops (the TV show) blurs so many faces? It's because they either couldn't find the person to get a model release, or because the individual wouldn't sign it.
I can understand the plantiff's emotional distress, but it is appropriate to put the blame there? Wouldn't suing the ex-boyfriend be a lot more appropriate here - after all, it is the ex-boyfriend who supposedly posted the photos. And if the ex-boyfriend doesn't have a model release, she would certainly have a good case against him.
Of course, if she had just used a DMCA takedown notice - since she appears in the photographs, she has a copyright interest in them - Yahoo would probably have jumped and quickly locked out the offending accounts.
Can you use the iPod to record from satellite radio and play it back later?
And there's the answer of why Sirius is talking with Apple. Delphi introduced the "XM2GO" receiver in November '04 as the first portable satellite radio solution. To get around the "no satellite reception indoors" issue, the XM2GO actually records up to 5 hours of programming internally. Sirius has not yet announced a comparable device for their service.
The iPod has a very strong market at the moment. It's a good thought for Sirius to try and leverage the iPod into an "XM2GO" killer which uses the Sirius service. XM2GO retails for $250 (just checked on Amazon), so an integrated iPod/Sirius unit could probably come in at a reasonable price.
I must admit that Prof. Dannenberg actually did rather little to counter Sherman's arguments
I think the Prof. did a good job here, primarily in showing that there's a historical record of established companies either (a) using their technology to control who gets published, or (b) using their position in an attempt to prevent other companies from competing with them.
Internet2 is all about research - should that research be restricted because an organization known for strong-arm tactics is attempting to strong-arm the research institutions? What about all those independent acts out there who can't get their creative works published, and who want to use P2P to build the fan base that could get them a record deal? What about alternatively-licensed content that needs to be effectively distributed? Should we listen to an organization that has historically shown itself as an impediment to new technology?
I think the Professor did a fine job in discrediting the article's author. But, alas, it's in today's paper, which is read less (ok... of which fewer copies are sold) than the Sunday edition, where the article was published. Think the RIAA cares about the letter?
s/are thinking of switching/have switched already/
I purchased my first iPod about a year and a half ago. After the terrible experience of MusicMatch & the iPod on Windows, I longed for the Mac iPod experience. I also lusted after OS X and the Dock - which works the way I like my computer to work.
A few months later, my web-surfing laptop died on me. In looking for a new laptop, the iBooks definitely stood out to me, and I went for it. Safari is good, Firefox works well, and most other things that I really wanted on my web laptop were available.
It's worked out for me. I've even replaced my desktop with a 20" iMac G5, and I don't miss my noisy, tons-of-compatibility-problems PC.
I was always taught that it was bad form to depend on
if ( !ptr )
being correct as a NULL-pointer test. Not that it's a violation of the C standard, but it's converting a pointer type to a boolean type, which is a bad habit - especially in a permissive language such as C.
Let's assume for a minute that some engineer, buried deep in the guts of FacelessMegaCo, Inc., manages to design a piece of hardware where bridging two unmarked terminals will bypass all security restrictions on the broadcase. Since this is an assumption, we'll further assume that this modification is undetectable and cannot be protected against except by replacing hardware in the field.
Why did the designer do it? Well, I think we would assume that it's for the desiner's own use. Necessity is the mother of invention, and there won't be much invention going around. Since it's a hardware kludge, it's very unlikely that it will ever be detected during testing. Even if an external body must approve of the design, it's unlikely this specific quirk will be detected. The designer will be able to buy one of these devices and modify it.
But would the designer share their knowledge with a friend? This is what The Public (tm) would want. The designer may even realize that their company could take a larger share of the market by "leaking" information about the kludge. The designer could certainly help their company.
But, then, the sudden swing in sales would alert someone that something's not right. Eventually they would find those same instructions online, trace down the people who could have inserted such a backdoor, fire them, and probably sue them for breach of contract.
So, even if such a backdoor were to exist, the only way it would be known is if it's a product that's already obsolete (therefore, cannot create a new demand that will trigger alarms) or if the secret kludge is very well guarded - in which case we call it an "engineering prototype."
Who says virtualization needs to be used to run different/independent operating systems? I think it'd be nice to be able to run multiple copies of the same OS, each in an independent virtual machine, so that programs - or, more importantly, virues/malware - cannot affect other software running on the same system, even if the OS itself is compromised.
Until, of course, a flaw is found in the virtualization layer itself, at which point it would be possible to hijack a computer at the CPU level and run a new, independent, trojan OS to do who-knows-what. Thankfully IBM has some experience with this, which means that such a vulnerability is less likely... right? ^^;
Slashdot Mirror
"The price of freedom is eternal vigilance." - Thomas Jefferson
OMB guidelines are rules on government agencies, and there are organizational penalties for failing to adhere to them. I don't see how we can excuse any agency for failing to follow the rules set out for them, let alone TWO agencies which are involved in espionage. We're not going to see a trial, because the OMB doesn't make laws, but I'm going to write my representatives in Congress and encourage them to issue a new law to codify this OMB guideline - that way, if they DO try it again, the consequences will be much more severe.
For a criminal investigation, we would have to prove that Diebold (or someone) acted illegally. I doubt that's going to happen. However, we have the opportunity to make source code availability a condition of ongoing government contracts for e-Voting equipment. This will allow us to have more trust in future elections that use e-Voting equipment. But the only way I will feel convinced that there wasn't foul play in the 2004 Presidential Election is to have Diebold release their source code.
Does the Windows source code matter in this case? No, not really, but it sure gives Diebold a great excuse to stop seeking certification in North Carolina, the only state I am aware of with a code escrow requirement. I do plan, however, on asking my state representatives to enact similar legislation in my home state. By allowing Diebold to slip out of North Carolina, we may be allowing them to slip away with code that could answer questions from so many about the last election.
If the Windows tagline is "Where Do You Want To Go Today?", then is the Xbox360's "Where Have You Tried Looking Today?"
Because when they want you to submit the source code for "everything" needed to run your app, you can blame Microsoft and avoid any questions about the code that you did get away with using for one election -- an election which was "delivered to Mr. Bush" as promised by Diebold executives.
Heck, even better - maybe it's both!
And to top it off, the submitter's name links to NotePage, which operates the FeedForAll site as well. And yet, no "conflict of interest" warning from the submitter.
This is quite correct, in the limited scope - Police should not be allowed to search through data on your computer without a warrant, as this data is as private as papers in a closed filing cabinet. Even when the cabinet is in transit, a police officer cannot decide to open it except when duly authorized, usually by a warrant. This is a position that needs to be defended, even if the individual at question isn't someone who we feel is entitled to the defense.
On the question of the Tech... well, that's a matter between the customer and the technician. I doubt any laws were broken by the Tech's report to the police - but, then, you would also be open to the "well, prove the data was there BEFORE it came into your hands, Tech" question - the computer didn't come right from the accused's hands, leaving a wide opening for "reasoanble doubt" when/if it came to trial.
Did you agree to random drug tests, at their whim? Perhaps they wanted to see your credit report?
In my experience, companies are constantly trying to gain more concessions from their employees, often without granting anything to the employee in exchange.
It's natural, then, that they're moving on to these genetic test - at least from the company's perspective. Employees, however, are balking at this brand new intrusion for now. But how long until it's just like that drug test that everyone else seems to be OK with, simply because they aren't looking for you?
If you don't speak out for others, no one will be left to speak out for you. This is why Unions are still a good thing - it allows workers to speak up against policies such as this while protecting themselves from direct retribution at work, since the company doesn't know who, exactly, started the complaint.
And if you're supposing that people could un-license GPL'ed code from Epicrealm, remember that such a change could not be done retroactively, and that Epicrealm would be entitled to continue using any GPL-licensed software they already have acquired.
For some reason, I'm left with the feeling that I've missed part of your point...
And the patent application was filed on April 23, 1996. "Prior art" must predate the filing date (04/23/1996), not the date the patent was granted (04/13/1999).
As such, I'm not as concerned about "woe unto all dynamic web sites," but if I managed one that offloads and caches page generation work (i.e., Slashdot, LiveJournal, and probably a lot more) I'd probably be calling my lawyer this morning.
The company, then, has a somewhat reasonable explanation for the need to restrict fraternization - it is a potential security lapse, which would result in a significant loss of reputation for their company.
In this limited role, I cannot help but agree. Especially since the NLRB rules only on the basis of interference with union-organizing activities. As I read the NLRB decision, this wasn't a review for Constitutionality -- only to determine if the rule was designed to curtail or prevent union activity at the company.
For instance, I wouldn't be surprised if Verizon VoIP over Verizon DSL was a very reliable combination (assuming, of course, that the DSL installation itself is reliable). But it could be that Verizon VoIP has terrible availability when accessed via any other carrier - because Verizon VoIP might not have an equal-quality interconnect with other IP carriers.
Vertical integration of the service with the carrier - like traditional POTS - allows for integration of the support experience and elimination of interconnect issues which impair reliability. (Just try getting your local loop fixed by the ILEC when there's a fault which causes your non-ILEC DSL service to die frequently.)
Honestly, the only data point which matters to me is what's the reliability of the best provider/carrier combination as compared to POTS, not the "average" reliability of a provider over all carriers.
Since both the NSLU2 and the Mac Mini support USB2, I don't really understand how one would have faster disk I/O than the other. Now, certainly, there can be questions about file system efficiency, but you could also say that the Mac Mini (for, granted, 5x the price) could have much faster I/O, since you have 256MB of RAM instead of 32MB for things like read-caching.
more stable
I'm very interested as to why the NSLU2 would be more stable than a Mac Mini. Personally, I think the short, square design of the Mini is more stable than the tall, tower design of the NSLU2. From a software standpoint, I'd say the two are rather comparable in stability for comparable activities (i.e., no complaining about 3D games crashing on the Mini when the NSLU2 doesn't even have a graphics chip).
more flexible
While the Mini's certainly not a flexibility powerhouse, I don't see the NSLU2 beating it out here, either. Software flexibility? Mac OS X has a lot of software available, both commerical and software libre (GPL/BSD). You can also run GNU/Linux on the Mini, including the Debian distribution with apt-get.
Not to mention the ability to hack it when I want (The warranty is already void!).
Open your Mini and replace the hard drive. Voila, both can be hacked at will because the warranty is void! ^__^
Current iBooks do not have DVI output; they have a "mini-VGA" connector to which an adaptor for Composite or S-Video can be connected. This is the same mini-VGA connector that the iMac G5 provides, so I would not expect to see it replaced with DVI.
You are, of course, by buying the additional services that the phone company wants to sell you for your DSL link, or that the cable company wants to sell you for your cable modem link.
Verizon has thrown a lot of hype into a forthcoming video service. Unlike cable services, though, such Video over DSL services require a fixed amount of bandwidth for each receiver, say 6 Mbps. If you have a 30 Mbps link, you could host 5 receivers (and no other services) or 4 receivers and hae a 6 Mbps downstream ISP connection.
But in this age of DVRs and connected devices, 4 receivers may be insufficient for affluent households with lots of TVs & other video receiving devices. It is because of this that I expect faster connections are going to be deployed, not because we want faster internet.
Best performance per watt != Lowest power usage of highest-performing part.
The Pentium M family is much lower power than the Pentium 4, and has reasonably good performance. I don't think AMD really has a chip that competes with the Pentium M, even though AMD's chips are generally less power-hungry than a Pentium 4.
Subjects in photographs have rights to those images as well, unless they've signed a model release. There are some widely accepted cases where a model release isn't required - such as photographs in the newpaper, if the picture was taken in public (and the individual therefore had no reasonable expectation of privacy).
But there are many places where model releases are required - why else do you think Cops (the TV show) blurs so many faces? It's because they either couldn't find the person to get a model release, or because the individual wouldn't sign it.
Of course, if she had just used a DMCA takedown notice - since she appears in the photographs, she has a copyright interest in them - Yahoo would probably have jumped and quickly locked out the offending accounts.
And there's the answer of why Sirius is talking with Apple. Delphi introduced the "XM2GO" receiver in November '04 as the first portable satellite radio solution. To get around the "no satellite reception indoors" issue, the XM2GO actually records up to 5 hours of programming internally. Sirius has not yet announced a comparable device for their service.
The iPod has a very strong market at the moment. It's a good thought for Sirius to try and leverage the iPod into an "XM2GO" killer which uses the Sirius service. XM2GO retails for $250 (just checked on Amazon), so an integrated iPod/Sirius unit could probably come in at a reasonable price.
I think the Prof. did a good job here, primarily in showing that there's a historical record of established companies either (a) using their technology to control who gets published, or (b) using their position in an attempt to prevent other companies from competing with them.
Internet2 is all about research - should that research be restricted because an organization known for strong-arm tactics is attempting to strong-arm the research institutions? What about all those independent acts out there who can't get their creative works published, and who want to use P2P to build the fan base that could get them a record deal? What about alternatively-licensed content that needs to be effectively distributed? Should we listen to an organization that has historically shown itself as an impediment to new technology?
I think the Professor did a fine job in discrediting the article's author. But, alas, it's in today's paper, which is read less (ok... of which fewer copies are sold) than the Sunday edition, where the article was published. Think the RIAA cares about the letter?
I purchased my first iPod about a year and a half ago. After the terrible experience of MusicMatch & the iPod on Windows, I longed for the Mac iPod experience. I also lusted after OS X and the Dock - which works the way I like my computer to work.
A few months later, my web-surfing laptop died on me. In looking for a new laptop, the iBooks definitely stood out to me, and I went for it. Safari is good, Firefox works well, and most other things that I really wanted on my web laptop were available.
It's worked out for me. I've even replaced my desktop with a 20" iMac G5, and I don't miss my noisy, tons-of-compatibility-problems PC.
Why did the designer do it? Well, I think we would assume that it's for the desiner's own use. Necessity is the mother of invention, and there won't be much invention going around. Since it's a hardware kludge, it's very unlikely that it will ever be detected during testing. Even if an external body must approve of the design, it's unlikely this specific quirk will be detected. The designer will be able to buy one of these devices and modify it.
But would the designer share their knowledge with a friend? This is what The Public (tm) would want. The designer may even realize that their company could take a larger share of the market by "leaking" information about the kludge. The designer could certainly help their company.
But, then, the sudden swing in sales would alert someone that something's not right. Eventually they would find those same instructions online, trace down the people who could have inserted such a backdoor, fire them, and probably sue them for breach of contract.
So, even if such a backdoor were to exist, the only way it would be known is if it's a product that's already obsolete (therefore, cannot create a new demand that will trigger alarms) or if the secret kludge is very well guarded - in which case we call it an "engineering prototype."
Until, of course, a flaw is found in the virtualization layer itself, at which point it would be possible to hijack a computer at the CPU level and run a new, independent, trojan OS to do who-knows-what. Thankfully IBM has some experience with this, which means that such a vulnerability is less likely... right? ^^;