I'm upgrading to 4.1 because now the generic kernel allows my PowerMac G4 router/server to restart automatically in the event of a power failure. But frankly, I probably would have upgraded anyway: otherwise, it would be difficult for me to justify buying the CD and supporting the project, wouldn't it?:P
Just a correction to one of your points: RSS is not push; it's pull.
I just stopped in here to make that point (RSS was branded as "push" in TFA as well), but it's good to see that somebody beat me to it. Anybody who has to pay the bandwidth for a popular Atom or RSS feed can tell you that RSS is most definitely a "pull" protocol.
Like RSS, AJAX is really just another application of Good Old HTTP. AJAXish web sites can indeed yield more efficient bandwidth utilization than traditional designs, but from the network's perspective, AJAX is a "pull" protocol as well. (*Not that I am implying AJAX to actually be a protocol, or anything at all more than an overly-hyped term referring to an amorphous set of web design techniques, but I digress...)
I don't know, I sort of saw it the other way around:
"Hey all you guys, listen up. I know some of you were thinking of switching to Linux or the Mac or something for improved security, but really, you're better off staying put with Windows. And by the way, did I mention that our products run on Windows?"
2. HP makes a clone that costs $100 and gouges you for a hefty $4 per ml for ink.
Fortunately for Silverbrook, it sounds like they have several patents on their technology. HP won't be able to sink Memjet by cloning this printer, because HP would have to pay them royalties for each clone sold. Silverbrook could even prevent HP from copying it altogether if they desired.
But this is all assuming that Silverbrook actually wants to sell these things itself. If their core business is indeed licensing patents, then it's possible that they just wanted to come up with a prototype to scare the pants off of the big inkjet manufacturers. Make a nice press release with a cool video, and stir up coverage with promises of inexpensive ink, and soon HP, Epson, Canon, and all the others will be knocking at the door, asking how they can license this for their own use.
If Silverbrook genuinely wants to sell us cheap Memjet ink, then HP won't be able to stop them. But it's entirely possible that they would prefer to license Memjet to would-be competitors, in which case your prediction comes true; everyone carries on as before.
You're missing the point. It's not that the current "presentation" of the Semantic Web is too complex; the problem is that actually creating the Semantic Web is too complex a task for most Web content creators to be interested in.
Essentially, the Semantic Web asks users to explicitly state relations between concepts and ideas to make up for our current lack of an AI capable of discerning such things for itself from natural human language. But let's face it, the average Joe writing his weblog or LiveJournal entries - or even a more technical user such as myself - would generally not be interested in performing this time-consuming task, even with the aid of a fancy WordPress plugin or other automated process. This is what the parent meant by saying it's just "too complicated".
The way to realize the Semantic Web is to advance AI technology to the point where it becomes an automated process. Anything less would require too much manual labor to take off.
The other problem with running DD-WRT on an off-the-shelf router, aside from the comparative lack of flexibility, is the distinct possibility of bricking the thing with a bad firmware update, even if you're careful. My Linksys WRT-54G died when the power browned out during a DD-WRT upgrade. I couldn't even fix it by soldering on a JTAG header and trying to load the software manually.
This kind of thing is fun to try if you have an old Linksys lying around, but if you really want to set up your own Über-Router you should find something that has a bit more RAM, and is a bit less fragile.
In a competitive market like that of the PC, what these companies can charge for their products is determined by market pressures, to the effect that (botique brands aside) sellers don't have the liberty of toying around with their profit margins. And when the PC's profit margin is restricted by a competitive market, customers are, in effect, paying for what their computers are worth, plus a predictable percentage profit.
You say that it might be absorbed by the company's bottom line instead, but where does that bottom line come from? It's meaningless to say "the company had to pay for the increase in cost due to the Windows licenses, whereas the customer paid for the increase due to the addition of a graphics card". Money is money, and either way it comes from the customer. Since playing around with semantics doesn't change the profit that a PC seller is able to demand for its machines, the customer will pay for any and all added costs of manufacture.
And more importantly, the guy who made this doesn't have to walk in front of the screen in the middle of the football / basketball / synchronized swimming game, either. Now we just need a bottle-tossing version...
There would be no NAT boxes required to share your connection amongst several computers, meaning all those worms would have affected just about every Windows computer on the Internet (instead of just the ones that were directly connected).
Whether you have a sufficient address space to hand an address to each computer is irrelevant – providing Internet access to a number of machines behind a single Internet connection is still going to require a router. And there's no reason that, just because a consumer-grade IPv6 router no longer has to kludge around with NAT, it should spontaneously cease to function as a firewall.
They're doing a great job porting OpenOffice to OSX, a job that the OpenOffice people seem unwilling to do, and I hope they get the suppor they need.
And the way to ensure they get that support is to go to NeoOffice.org and make a donation. I did two weeks ago, and I encourage everyone else interested in ensuring the continued development of this crucial OS X office suite to do the same. Yes, I'm talking to you!
The existence of vigilante software like this is, in my mind, one of the strongest arguments for capabilities-based security. In traditional systems with ACL-based security (i.e., every popular PC operating system today), we really don't have a way to say "I trust this program to record video from my screen, but not to delete all of my documents." A properly-implemented capabilities system, on the other hand, could give us just that.
"Fairly comprehensive" and "The Register" never, ever belong in the same sentence together.
This is one of those few times I've found myself wishing Slashdot had Digg's "Bury Story" feature – this article serves neither to enlighten nor to persuade. It's not aimed at the kind of intelligent, informed people at the center of the open source community who would genuinely be interested in how Vista's release affects Windows security; it only preaches to the choir of those poor and confused souls who hate Microsoft because it feels good to hate Microsoft.
This quote was particularly enlightening:
And there's the catch: "Windows needs your permission to install this cleverly-disguised Trojan / nifty program. Click Yes to get rooted / continue."... So you see that, here again, MS's security strategy involves shifting responsibility to the user.
Honestly, how is this any different from the state of affairs on Linux, BSD, OS X, Solaris, or any other operating system? Thankfully Windows now does what it can to ensure you're fully aware when software is being installed on your system, but within the realm of current technology, it will always be ultimately up to the administrator (i.e., end user) to differentiate between trustworthy and untrustworthy software. That's just the nature of the game. To try to play this off as some particular flaw in Windows is idiotic, and completely wrecks the author's credibility.
To those who seem to be enjoying this article so much: If you just like the adrenaline rush, consider playing a game of racquetball instead. For the sake of the rest of us, please leave Slashdot for actual, honest-to-goodness news and analysis.
You're correct in that even this is better than no security at all. However, what I (and the summary) meant by "increase security" was security with respect to the traditional method of locking one's phone, which is with a PIN or a password of some kind. In that sense, this system will not increase security, as fingerprint authentication systems are demonstrably less secure than a well-chosen password.
Frighteningly, I don't think there's that great a difference, in terms of technical feasibility, cost of implementation, or legal dubiousness, between the NSA clandestinely spying on the private conversations of U.S. citizens by the aid of AT&T and others; and that same agency, hypothetically, collecting fingerprint data from consumers by the aid of whichever cellular carriers will offer this phone.
It may seem improbable, but we've already seen equally grievous government intrusions into personal privacy. Such a scenario would honestly just be more of the same at this point.
Of course it also means that someone is more likely to chop your hand off if they desperately want your data.
More realistically, you'd also have to worry about somebody lifting your fingerprint from, say, the phone itself, then using that to log in. The MythBusters did a segment showing how easy it is to lift somebody's fingerprint, then use that print to defeat a scanner.
This thing isn't going to increase security, it's only going to increase convenience.
That could, of course, be fixed by moving all of the encryption to the drive itself (or perhaps to the drive controller), but that's a completely different architecture than what we're talking about here.
Where possible, Vista's BitLocker uses a Trusted Platform Module to encrypt and decrypt data to and from the disk. The article was light on details (needless to say; it's The Register), but if this hypothetical cryptographic back door were implemented in the TPM rather than in the OS, there wouldn't be much the end user could do to get around it.
"I personally hate Scientology but they are a religion and must be respected as one."
I disagree. Actually, this reminds me of something that Douglas Adams once said:
If somebody votes for a party that you don't agree with, you're free to argue about it as much as you like; everybody will have an argument but nobody feels aggrieved by it. If somebody thinks taxes should go up or down you are free to have an argument about it. But on the other hand if somebody says 'I musn't move a light switch on a Saturday', you say, 'I respect that'.
Why should it be that it's perfectly legitimate to support the Labour party or the Conservative party, Republicans or Democrats, this model of economics versus that, Macintosh instead of Windows - but to have an opinion about how the Universe began, about who created the Universe... no, that's holy?... We are used to challenging religious ideas but it's very interesting how much of a furore Richard [Dawkins] creates when he does it! Everybody gets absolutely frantic about it because you're not allowed to say these things. Yet when you look at it rationally there is no reason why those ideas shouldn't be as open to debate as any other, except that we have agreed somehow between us that they shouldn't be.
The BSD guys are (clearly) amazing, but you're terribly misguided if you think they had anything to do with the fact that 32-bit IOKit drivers can be loaded into the 64-bit kernel, or that 64-bit Cocoa is 32-bit safe.
I agree completely. Looking forward, the best way to add value to The Children's Machine is to increase the amount and diversity of software available for it. And the best way to do that is to sell the machine to those who want to tinker with, and possibly program, it.
I agree. My kneejerk reaction to the news that Apple TV only plays h.264 video was that this was a weird attempt at pseudo- vendor lock-in. While that still may have been a small factor in this design choice, it seems far more likely that the decision was due to h.264's high compression ratio, which is a major advantage not only when you're storing movies on that 40GB hard drive, but also when trying to stream 720p video over 802.11.
No, it's not DRM. It's a watermark. It does not and cannot prevent you from doing anything at all with the video; all it does is provide the content owners with a means of identifying copyright violators, should the video show up in a torrent somewhere.
I agree with shirizaki - this is the proper way for media to be distributed online. This watermark will never be an issue for you until after you've already, publicly violated someone else's copyright.
If you forget the password you'll have to contact Apple, but the Apple ID of your account is stored within your purchased music files. Attempting to play a FairPlay DRMed track for which your computer is not authorized will cause iTunes to prompt you for the password to the appropriate Apple account.
No, he's right on the point. The reason our old Apple ][s are easier to program than a modern computer is that the Apple ][ didn't have an operating system with nearly as much functionality as a modern version of Windows, OS X, Linux, or BSD. Today's more sophisticated (yes, I would argue sophisticated) systems necessitate more complex APIs.
The large number of functions and libraries that a modern programmer has to dig through while writing software for today's systems is an unfortunate – but necessary – byproduct of these more advanced OSes. You generalize that these APIs are a waste of time: "Not worth the effort: just do it yourself and get it over with." That may be fine for some very small tasks, but when it comes to the more complex aspects of writing software, that stance is completely unfeasible. Would you like to implement that TCP stack you're using all by yourself? How about that windowing system which your customers seem to love so much? Maybe the filesystem, too? Where do we draw the line?
A similar case can be made for less fundamental system APIs, such as Apple's Core Data API in Tiger. You might think of it as a waste of time, and if so you're free to roll your own data persistence framework in each of your applications; by merely providing the API, Apple isn't forcing you to waste your time learning Core Data if you don't want to use it. But when I need to roll out an OS X application on a deadline, having something like Core Data to save me development time can be a real lifesaver.
If you know what you're looking for, having a variety of pre-built solutions to build on top will save you way more time than you will "waste" looking through the docs. Sure, that shifts some of the difficulty of software development from the skill of laying down code to the skill of figuring out which pieces to glue together under certain circumstances, but that's just the art of programming. As developers, it is our job to solve problems in the most efficient way feasible.
I think one of the single most important factors holding back the Macintosh in the corporate arena is the lack of a clearly defined product lifecycle for OS X. Correct me if I'm wrong, but nobody outside Apple seems to know, on authority, how much longer we'll receive security updates for 10.3 or 10.4.
It's difficult to justify widely deploying any given platform, even one as nice as OS X, if you don't know when the product will be forcefully obsoleted.
Slashdot Mirror
I'm upgrading to 4.1 because now the generic kernel allows my PowerMac G4 router/server to restart automatically in the event of a power failure. But frankly, I probably would have upgraded anyway: otherwise, it would be difficult for me to justify buying the CD and supporting the project, wouldn't it? :P
Just a correction to one of your points: RSS is not push; it's pull.
I just stopped in here to make that point (RSS was branded as "push" in TFA as well), but it's good to see that somebody beat me to it. Anybody who has to pay the bandwidth for a popular Atom or RSS feed can tell you that RSS is most definitely a "pull" protocol.
Like RSS, AJAX is really just another application of Good Old HTTP. AJAXish web sites can indeed yield more efficient bandwidth utilization than traditional designs, but from the network's perspective, AJAX is a "pull" protocol as well. (*Not that I am implying AJAX to actually be a protocol, or anything at all more than an overly-hyped term referring to an amorphous set of web design techniques, but I digress...)
I don't know, I sort of saw it the other way around:
"Hey all you guys, listen up. I know some of you were thinking of switching to Linux or the Mac or something for improved security, but really, you're better off staying put with Windows. And by the way, did I mention that our products run on Windows?"
Maybe I'm just cynical today...
Fortunately for Silverbrook, it sounds like they have several patents on their technology. HP won't be able to sink Memjet by cloning this printer, because HP would have to pay them royalties for each clone sold. Silverbrook could even prevent HP from copying it altogether if they desired.
But this is all assuming that Silverbrook actually wants to sell these things itself. If their core business is indeed licensing patents, then it's possible that they just wanted to come up with a prototype to scare the pants off of the big inkjet manufacturers. Make a nice press release with a cool video, and stir up coverage with promises of inexpensive ink, and soon HP, Epson, Canon, and all the others will be knocking at the door, asking how they can license this for their own use.
If Silverbrook genuinely wants to sell us cheap Memjet ink, then HP won't be able to stop them. But it's entirely possible that they would prefer to license Memjet to would-be competitors, in which case your prediction comes true; everyone carries on as before.
Well it seems only fitting, since they're the ones who invented Ajax in the first place...
You're missing the point. It's not that the current "presentation" of the Semantic Web is too complex; the problem is that actually creating the Semantic Web is too complex a task for most Web content creators to be interested in.
Essentially, the Semantic Web asks users to explicitly state relations between concepts and ideas to make up for our current lack of an AI capable of discerning such things for itself from natural human language. But let's face it, the average Joe writing his weblog or LiveJournal entries - or even a more technical user such as myself - would generally not be interested in performing this time-consuming task, even with the aid of a fancy WordPress plugin or other automated process. This is what the parent meant by saying it's just "too complicated".
The way to realize the Semantic Web is to advance AI technology to the point where it becomes an automated process. Anything less would require too much manual labor to take off.
The other problem with running DD-WRT on an off-the-shelf router, aside from the comparative lack of flexibility, is the distinct possibility of bricking the thing with a bad firmware update, even if you're careful. My Linksys WRT-54G died when the power browned out during a DD-WRT upgrade. I couldn't even fix it by soldering on a JTAG header and trying to load the software manually.
This kind of thing is fun to try if you have an old Linksys lying around, but if you really want to set up your own Über-Router you should find something that has a bit more RAM, and is a bit less fragile.
In a competitive market like that of the PC, what these companies can charge for their products is determined by market pressures, to the effect that (botique brands aside) sellers don't have the liberty of toying around with their profit margins. And when the PC's profit margin is restricted by a competitive market, customers are, in effect, paying for what their computers are worth, plus a predictable percentage profit.
You say that it might be absorbed by the company's bottom line instead, but where does that bottom line come from? It's meaningless to say "the company had to pay for the increase in cost due to the Windows licenses, whereas the customer paid for the increase due to the addition of a graphics card". Money is money, and either way it comes from the customer. Since playing around with semantics doesn't change the profit that a PC seller is able to demand for its machines, the customer will pay for any and all added costs of manufacture.
And more importantly, the guy who made this doesn't have to walk in front of the screen in the middle of the football / basketball / synchronized swimming game, either. Now we just need a bottle-tossing version...
Whether you have a sufficient address space to hand an address to each computer is irrelevant – providing Internet access to a number of machines behind a single Internet connection is still going to require a router. And there's no reason that, just because a consumer-grade IPv6 router no longer has to kludge around with NAT, it should spontaneously cease to function as a firewall.
And the way to ensure they get that support is to go to NeoOffice.org and make a donation. I did two weeks ago, and I encourage everyone else interested in ensuring the continued development of this crucial OS X office suite to do the same. Yes, I'm talking to you!
The existence of vigilante software like this is, in my mind, one of the strongest arguments for capabilities-based security. In traditional systems with ACL-based security (i.e., every popular PC operating system today), we really don't have a way to say "I trust this program to record video from my screen, but not to delete all of my documents." A properly-implemented capabilities system, on the other hand, could give us just that.
See http://www.eros-os.org/essays/capintro.html for a better introduction to capability systems than I could possibly provide here.
"Fairly comprehensive" and "The Register" never, ever belong in the same sentence together.
This is one of those few times I've found myself wishing Slashdot had Digg's "Bury Story" feature – this article serves neither to enlighten nor to persuade. It's not aimed at the kind of intelligent, informed people at the center of the open source community who would genuinely be interested in how Vista's release affects Windows security; it only preaches to the choir of those poor and confused souls who hate Microsoft because it feels good to hate Microsoft.
This quote was particularly enlightening:
Honestly, how is this any different from the state of affairs on Linux, BSD, OS X, Solaris, or any other operating system? Thankfully Windows now does what it can to ensure you're fully aware when software is being installed on your system, but within the realm of current technology, it will always be ultimately up to the administrator (i.e., end user) to differentiate between trustworthy and untrustworthy software. That's just the nature of the game. To try to play this off as some particular flaw in Windows is idiotic, and completely wrecks the author's credibility.
To those who seem to be enjoying this article so much: If you just like the adrenaline rush, consider playing a game of racquetball instead. For the sake of the rest of us, please leave Slashdot for actual, honest-to-goodness news and analysis.
You're correct in that even this is better than no security at all. However, what I (and the summary) meant by "increase security" was security with respect to the traditional method of locking one's phone, which is with a PIN or a password of some kind. In that sense, this system will not increase security, as fingerprint authentication systems are demonstrably less secure than a well-chosen password.
Frighteningly, I don't think there's that great a difference, in terms of technical feasibility, cost of implementation, or legal dubiousness, between the NSA clandestinely spying on the private conversations of U.S. citizens by the aid of AT&T and others; and that same agency, hypothetically, collecting fingerprint data from consumers by the aid of whichever cellular carriers will offer this phone.
It may seem improbable, but we've already seen equally grievous government intrusions into personal privacy. Such a scenario would honestly just be more of the same at this point.
More realistically, you'd also have to worry about somebody lifting your fingerprint from, say, the phone itself, then using that to log in. The MythBusters did a segment showing how easy it is to lift somebody's fingerprint, then use that print to defeat a scanner.
This thing isn't going to increase security, it's only going to increase convenience.
Where possible, Vista's BitLocker uses a Trusted Platform Module to encrypt and decrypt data to and from the disk. The article was light on details (needless to say; it's The Register), but if this hypothetical cryptographic back door were implemented in the TPM rather than in the OS, there wouldn't be much the end user could do to get around it.
"I personally hate Scientology but they are a religion and must be respected as one."
I disagree. Actually, this reminds me of something that Douglas Adams once said:
The BSD guys are (clearly) amazing, but you're terribly misguided if you think they had anything to do with the fact that 32-bit IOKit drivers can be loaded into the 64-bit kernel, or that 64-bit Cocoa is 32-bit safe.
I agree completely. Looking forward, the best way to add value to The Children's Machine is to increase the amount and diversity of software available for it. And the best way to do that is to sell the machine to those who want to tinker with, and possibly program, it.
I agree. My kneejerk reaction to the news that Apple TV only plays h.264 video was that this was a weird attempt at pseudo- vendor lock-in. While that still may have been a small factor in this design choice, it seems far more likely that the decision was due to h.264's high compression ratio, which is a major advantage not only when you're storing movies on that 40GB hard drive, but also when trying to stream 720p video over 802.11.
No, it's not DRM. It's a watermark. It does not and cannot prevent you from doing anything at all with the video; all it does is provide the content owners with a means of identifying copyright violators, should the video show up in a torrent somewhere.
I agree with shirizaki - this is the proper way for media to be distributed online. This watermark will never be an issue for you until after you've already, publicly violated someone else's copyright.
If you forget the password you'll have to contact Apple, but the Apple ID of your account is stored within your purchased music files. Attempting to play a FairPlay DRMed track for which your computer is not authorized will cause iTunes to prompt you for the password to the appropriate Apple account.
No, he's right on the point. The reason our old Apple ][s are easier to program than a modern computer is that the Apple ][ didn't have an operating system with nearly as much functionality as a modern version of Windows, OS X, Linux, or BSD. Today's more sophisticated (yes, I would argue sophisticated) systems necessitate more complex APIs.
The large number of functions and libraries that a modern programmer has to dig through while writing software for today's systems is an unfortunate – but necessary – byproduct of these more advanced OSes. You generalize that these APIs are a waste of time: "Not worth the effort: just do it yourself and get it over with." That may be fine for some very small tasks, but when it comes to the more complex aspects of writing software, that stance is completely unfeasible. Would you like to implement that TCP stack you're using all by yourself? How about that windowing system which your customers seem to love so much? Maybe the filesystem, too? Where do we draw the line?
A similar case can be made for less fundamental system APIs, such as Apple's Core Data API in Tiger. You might think of it as a waste of time, and if so you're free to roll your own data persistence framework in each of your applications; by merely providing the API, Apple isn't forcing you to waste your time learning Core Data if you don't want to use it. But when I need to roll out an OS X application on a deadline, having something like Core Data to save me development time can be a real lifesaver.
If you know what you're looking for, having a variety of pre-built solutions to build on top will save you way more time than you will "waste" looking through the docs. Sure, that shifts some of the difficulty of software development from the skill of laying down code to the skill of figuring out which pieces to glue together under certain circumstances, but that's just the art of programming. As developers, it is our job to solve problems in the most efficient way feasible.
I think one of the single most important factors holding back the Macintosh in the corporate arena is the lack of a clearly defined product lifecycle for OS X. Correct me if I'm wrong, but nobody outside Apple seems to know, on authority, how much longer we'll receive security updates for 10.3 or 10.4.
It's difficult to justify widely deploying any given platform, even one as nice as OS X, if you don't know when the product will be forcefully obsoleted.