Of course, it might not be my webmail, if someone has DNS poisened the system for example. But that is beside the point
Well. It sounds to me like your problem is the webmail provider. If you really don't care about security, then you should use a webmail provider that provides plain http access, and if you do care about security, then you should use a provider with a valid certificate. If you can't change webmail providers, then you should petition them to fix their problem. It's not IE's fault that your webmail sucks, just like it's not Firefox's fault that all those IE-only websites don't work. In fact, IE7 will help fix the problem for you, because when your webmail finds out that everyone with IE7 can't access their mail, they will *have* to fix it.
Risking a "man in the middle" attack once the first time you visit a site is worse than sending private info unencrypted over and over again?
That's a false dichotomy. First of all, there's no mechanism in SSL like there is in SSH to remember keys, so your first option is impossible for now. Secondly, there is a third option: force the site owners to get their freaking certificates signed properly and actually be secure! When browsers act like IE7 and Firefox 3, site owners will have no choice but to do things the secure way, and that is a good thing for the Internet as a whole. Users will see fewer errors overall once the transition is complete, and they will be more secure.
Actually, to go back to your first option, I do agree that there should be some mechanism in SSL implementations like SSH has to remember keys (other than adding people to your trusted root). But you would have to design the user interface for such a mechanism *extremely* carefully. Remember, when you put something in a web browser, it is going to get used by hundreds of millions of people who will trust it with their most important data. It would need to be truly idiot-proof, and that would be extremely hard as you'd have to get idiots to understand the subtle implications of their security-related actions.
how is having an unsigned/self-signed certificate WORSE than having NO certificate?
It's a false sense of security, leading users to do stupid things. People tend to believe that encryption is a brick wall that will protect them against anything, and it's almost true when you do both encryption and authentication, like SSL. Trouble is, on a LAN such as a public WiFi network, doing man-in-the-middle is almost as easy as sniffing, so defending against one but not the other is hardly an improvement, and outweighed by the damage done by false user expectations. Encryption without authentication is just wrongheaded.
It refuses to let me go directly to a secure website that has been signed by itself
This is a good thing, and if I understand correctly Firefox 3 will be doing the same. If you think otherwise, you do not understand how SSL works and why encryption without authentication is worse than useless. Anyone can self-sign a certificate claiming to be your webmail provider, or Amazon.com, or whoever they want. If you click past the warning, you are owned, because encryption without authentication is *always* vulnerable to a man-in-the-middle attack. And if you don't believe man-in-the-middle attacks are a threat, you are a fool. Every time you are on the same LAN as other users who you don't completely trust, you are vulnerable. Additionally, you are vulnerable to every ISP in the chain between you and your destination.
Now that browsers are making it harder for users to bypass the security warnings, there will be a lot more pressure on sites to fix the problems instead of instructing users to click past the warnings, so you will see less of these errors in the future, and the web will be a safer place for everyone.
Well, there was also the fact that his stories tended to be hype-filled press-release copy-paste jobs with sensationalized headlines but little real information, while the subject matter was typically mundane vaporware products or crackpot science. And yet somehow, his stories were accepted at an alarming rate. To his credit, lately they have been better. (Note that while his main story links are now direct, he still links to his blog from his name).
The wiggle room there is that statutory damages are used because calculating the actual damages is hard. The song costs $1, but the defendant probably uploaded it multiple times (and with the expectation that those copies would be further distributed). Therefore the damages are a *multiple* of $1. The RIAA would probably argue that the multiple could easily be 75 or more, making the damages fall under the 10:1 threshold you quote.
But I think a little critical thinking could topple that argument. It's easy to see that across a P2P system, the average number of times each copy of a song has been uploaded is 1. The number of uploads is by necessity exactly equal to the number of downloads. So in order for it to be plausible that the defendant uploaded a particular song 75 times, the defendant would have to be uploading that song *much* more than average. I don't think there's any evidence to prove that, though I might be wrong.
The RIAA might then argue that any uploading by the defendant was done with the expectation that the recipient would further upload the song to other people, and therefore uploads done by those people should factor in. Also, if this argument succeeded in reducing the damages, the RIAA would start collecting evidence on things like connection speed and uptime, and start suing people on fast pipes with 24/7 availability, using that as evidence that they uploaded more than average. There are plenty of those people to sue, and taking them out preferentially would kill P2P networks faster anyway.
Re:Yet ANOTHER sound server?
on
Fedora 8 Released
·
· Score: 5, Interesting
Oh, and I forgot to mention that PulseAudio has aspirations to become "Compiz for audio", providing earcandy effects such as surround-sound positioning for on-screen events (so sounds from a window on the left of the screen come from the left speaker, etc) and muffled sound from background windows (so the Flash ad in Firefox's background tab doesn't blast your eardrums and the new-mail notification doesn't sound over the movie you're watching full-screen).
Re:Yet ANOTHER sound server?
on
Fedora 8 Released
·
· Score: 5, Interesting
Thankfully, it does appear that PulseAudio is the One True sound server that we can all finally agree on. It emulates esd, OSS, and ALSA, so legacy apps like Flash and your smartphone work. It supports hotplug of audio devices, including networked ones (using Zeroconf even). It supports synchronized output between multiple devices, even when those devices use different sampling rates or have out-of-sync clocks (it resamples automatically). It has a zero-copy low latency architecture, taking advantage of the latest high resolution timer and real-time scheduling capabilities in new Linux kernels (when available), and it supports latency measurement for sound/video sync even when high latency is unavoidable (such as over a network). It has a modern user interface that provides per-application volume sliders like Windows Vista, and allows on-the-fly routing of audio to devices, including "saving" audio streams to another device if the device they are using is unplugged.
The guys behind PulseAudio really "get it". They even decided to drop their typically-awful open-source project name "PolypAudio" in favor of the infinitely better "PulseAudio", for wider acceptance. You've got to give them points for that; the GIMP could learn a thing or two from them.
Time machine backs up to a different disk, so it can save you in the event of a hard disk crash or other catastrophic event. Time Machine has a cool interface that is deeper than just a starfield wallpaper and some 3D effects. For example, you can take a spotlight search back in time to see what it would have found in the past.
In some ways Time Machine is not as cool as shadow copies, because it doesn't keep every version of the file with disk-sector-level copy-on-write; it does backups at specified times on its own schedule, and if a file is modified at all it backs up a whole new copy. If someone integrated shadow copies with the time machine interface, and also kept the features that back up to a different disk, then it would be the perfect all-singing, all-dancing backup tool.
You miss his point; 802.11g, and even n, are just as bad as 802.11b for truly large-scale deployment. Wi-Fi just isn't designed for deployment over anything larger than a business or university campus, and it shows. 802.16 is WiMAX, a completely different standard, and it does require buying extra hardware because no laptops currently have it (though they will in the future). But remember, these people already bought $500-$2000 laptops; I think a $100 USB dongle for ubiquitous Internet access is a no-brainer purchase. If you build a municipal WiMAX network and offer cheap or free access, convincing people to buy the hardware is the *least* of your worries.
Those bots weren't following the road markings, they were following the invisible GPS waypoints DARPA provided, which were much more detailed and accurate than you'd find in any car navigation system today. At best, their road-finding algorithms were fixing the last meter of inaccuracy in their differential GPS. And did you actually watch the bots as they traversed that section? I don't think any of them went over 10 MPH with frequent braking. Even CMU looked quite sluggish. A person could easily do 20-30 over such a flat and well-defined road, and that's not even in a race.
Firstly, a car you can't ever drive would never sell in the US. People want control, they want the ability to drive off-road even if they never actually do (see SUVs), and they love their older cars too much to stop driving them. Secondly, even if every car was automated, that would only take care of a *few* of the problems faced by automated vehicles. They would still have to deal with all of the problems that are caused by things other than unpredictable drivers, such as: wind, rain, snow, ice, fog, loss of GPS, worn or obscured road markings, people walking in the road, things that fall from trucks on the freeway, tires that blow out, malfunctioning traffic signals, downed power lines, mechanical failures of all kinds, collapsed bridges, avalanches, sinkholes, people trying to trick the robot sensors, and all the other problems I didn't happen to think of just now. If you really want your robotic car to be 100% safe, you have to program it to handle so many varied situations that I believe programming it with traffic rules for safe driving around humans would be a relatively small part of your work.
Now it's true that you could drive more efficiently without humans, but that will have to be phased in gradually. For example, you could have special robot lanes, and perhaps eventually entire robot-only streets in big cities. But that would only be possible *after* the introduction of autonomous vehicles.
No, no open source code. But what the public does get out of this is advances in technology. Case in point: the *real* winners of this year's Urban Challenge are Velodyne. Their lidar sensor was invented by team DAD for the 2005 challenge. For the 2007 challenge, they decided that instead of losing the competition again, they would sell their lidar technology to the other teams. Over half of the 35 teams in the challenge bought one, and 5 of the 6 finishers (Virginia Tech being the exception).
This thing is a huge advance over previous technology for this application, and it directly owes its existence to this challenge. Thanks to DARPA, you can now buy a lidar that you can stick on top of a car and which gives you 360 degree range data in 3D at 10 Hz over Ethernet. Now that the company is jump-started, next year those specs will improve, costs will go down, and eventually something like this will be driving your car for you. That's the benefit everyone gets from this competition. Not to mention all the people whose imaginations have been captured by the competition; who have been working on the funding DARPA gave out, getting their PhDs, or even just working in their spare time, learning how to write the software to run these things. There's no doubt in my mind that DARPA has gotten far more mileage from their money in this contest than they would have dumping it in the accounts of some defense contractor.
So even though no open source was produced from the contest, the public will see a lot of benefit from the money DARPA has spent.
Not because monopolies naturally arise on their own (they don't). We have anti-monopoly laws to fix the monopolies that arise because of *other* laws. In Microsoft's case, the other laws in question are the copyright laws. Microsoft is a monopoly because the government *grants* them a monopoly on their product; that's the express intent of the copyright laws (and the patent laws too). We shouldn't be surprised when these laws achieve their goals. The music and media cartels are similar products of government-granted copyright monopolies. If that isn't what we wanted after all, well then maybe the law ought to be changed.
If the government really wanted to end Microsoft's monopoly, they should have attacked the problem at its source: copyright. Consider: if the government voided Microsoft's copyright on Windows and/or Office, their dominance of the market would quickly end. For a less extreme remedy, Microsoft's copyrights could be restricted in various ways or made subject to various compulsory licenses.
I think your justification for not using placeholders is rather, uh, wrong. I agree that the argument order thing can be an issue, but that's what *named* placeholders are for. The major benefit of placeholders is not speed, it's absolute resistance to SQL injection. You may be diligent in quoting, but standard software development wisdom is that it's always better to eliminate the possibility of a bug than rely on programmer checking all the time, due to Murphy's Law of course. Also, the admittedly small speed benefit of placeholders when used with precompiled statements is going to be on the DB side, not the web side, which can make a bigger difference (as the DB is harder to scale).
This is not a class action suit and there's no law firm. This is an investigation by a state Attorney General. Verizon is paying a hundred or so thousand dollars to the state of New York in penalties, and they estimate a million to actual customers, plus changing their advertising *and* they have stopped terminating people's contracts (though that isn't an explicit requirement of the settlement).
As someone whose service actually *was* terminated by Verizon and who stands to be reimbursed for the cost of his hardware, I am pretty happy with this turn of events. I do wish the penalty was larger, because I have a sneaking suspicion Verizon has saved at least that much by terminating their most expensive customers, but I didn't really expect to get a resolution at all so I am pleasantly surprised by the effectiveness of the justice system.
ATI committed to providing complete documentation, but did not provide the source code to their current driver. This is (partially) because their current closed driver contains proprietary IP that ATI does not own, so they can't open-source it. NVIDIA is in the same situation and will have to take the same slow route to open-source drivers when they eventually come to their senses. They cannot "open up any day now" and "have a working code base" instantly, any more than ATI could.
VMs add practically zero overhead if they are implemented with cooperation from the OS being virtualized. A new OS that has the explicit goal of virtualizing legacy Windows could do a near-perfect job, ensuring compatibility while allowing Microsoft to end work on the giant legacy codebase that even they admit has become a nightmare.
This approach would be similar to what Apple did with the Classic environment in OS X, but with advanced features similar to Parallels to make the virtualization seamless. If Microsoft's Windows unit has any sense, Windows 7 will work this way, with a new kernel based on some of the great work coming out of Microsoft Research (I'm not holding my breath).
Au contraire; now that the SDK is announced I am much more likely to buy one. I only want to know one more thing: will I be able to freely upload my own native unsandboxed code to my own iPhone/iPod, without paying Apple and then having to get a digital signature every time? If the answer is yes, I will buy an iPod Touch tomorrow.
I think it's worth noting that our current space efforts (ISS, moon and mars bases) are not remotely "economical" either. If we must spend billions of tax dollars on space missions, I think they would be better spent doing something that might conceivably free us from oil dependency and benefit the entire human race here on Earth, than on a manned Mars mission (for example). The main practical benefit to the human race of a Mars mission, if there is one, is as a step toward mining and colonization of Mars and eventually even interstellar travel, but those goals are even farther away than space-based solar power, if they are even feasible.
Have you even tried ABC's stuff over a decent connection? It's far beyond VHS. It's also far beyond, for example, the compressed crap spewed by DirecTV on its SD channels. You need to reconsider your "any TV station" statement; that bar is pretty low. It doesn't match Fios's quality yet, but it's only going to get better as connections get faster; in fact if ABC were targeting only customers with Fios level (30Mbps+) connections, they could probably beat Fios's own picture quality today. And if Fios's full bandwidth was available as neutral switched IP instead of being reserved for Verizon alone, then there would be no question: ABC could serve up crazy picture quality, with zero cooperation from Verizon.
Further, you did not even look at the requirements of providing two different services off of the same connection
Actually, you are just ignoring the obvious and perfectly reasonable solution to this problem that I presented, which doesn't require packet shaping *inside of ISPs*. I won't repeat myself here; just read it in my original post (2nd paragraph).
So I shouldn't be allowed to set up a gaming-oriented ISP where my customers pay me to prioritize gaming packets over all other packets?
You could do that, you would just have to implement it slightly differently. You would provide neutral service to the user's modem, then install a value-add hardware router or software program (on the customer's side of the modem, and not required for service) to do any shaping the customer desires. In fact, a better business would be providing the hardware and/or software which customers of *any* ISP could install to prioritize gaming packets over all other packets. Or, ISPs could contract you to provide your gaming performance hardware to customers who want it. Doing the shaping inside the ISP confers no benefits over doing it at the user's end, plus it is much less flexible *and* encourages anti-competitive behavior when ISPs (which are natural monopolies) become content providers (which should never be monopolies).
I am aware that the term "IPTV" is commonly used to refer specifically to delivering a legacy cable-style user interface to TV delivered over IP, but ABC.com is also delivering TV over IP and I think it is stupid to call it something else just because the term "IPTV" was taken. Anyway, pedantic debates over the definition of the term "IPTV" are not the point here, and neither are the implementation details of Fios. The point is that high-quality video can be delivered over the Internet without cooperation from ISPs, as ABC.com is proving right now. We don't need ISPs meddling with our packets.
IPTV can't possibly work without QoS, huh? I personally am using IPTV all the time right now, provided by ABC.com, NBC.com, Comedy Central, Joost, iTunes, and yes, even YouTube. None of these services are provided by telecoms, and all are high enough quality for me in the *complete absence* of QoS. Whoops! Turns out QoS isn't quite as vital as you thought. OTOH, without net neutrality, it is likely cable companies will begin throttling video packets (and phone companies too as they try to move into the triple-play market). Given a choice between ISP-enforced QoS and net neutrality, I choose net neutrality.
First of all, TV over IP is here, and it works right now without any "dedicated bandwidth", despite doom and gloom from ISPs. I've been watching shows from NBC.com and ABC.com and it's worked quite well for me. My connection isn't fast enough for streaming HD, but ABC does offer it and I hear from other people that it works well. Joost works less well for me, but I suspect that's because my upload is crappy (Joost is P2P).
But I'll give you the benefit of the doubt and concede that maybe some kind of shaping is helpful (maybe if you're running BitTorrent while watching streaming TV and talking on Skype). Even so, that kind of shaping can and should be done on the consumer's side of the last mile. If you want glitch-free TV, you can set the router sitting in front of your cable modem to give TV priority over your other traffic. The ISP can even provide you a router with this capability pre-set; no configuration necessary. ABC.com could provide instructions and/or software for configuring routers, and future home routers could attempt to automatically classify and shape traffic (this could be a differentiating factor in the home router market). Doing shaping inside the ISP, where the customer has no control over it, is stupid and wrong. It can only work to benefit the ISP's interests and lock out competition for their services. No ISP is going to configure their routers to favorably shape traffic to services competing with their own offerings.
Slashdot Mirror
That's a false dichotomy. First of all, there's no mechanism in SSL like there is in SSH to remember keys, so your first option is impossible for now. Secondly, there is a third option: force the site owners to get their freaking certificates signed properly and actually be secure! When browsers act like IE7 and Firefox 3, site owners will have no choice but to do things the secure way, and that is a good thing for the Internet as a whole. Users will see fewer errors overall once the transition is complete, and they will be more secure.
Actually, to go back to your first option, I do agree that there should be some mechanism in SSL implementations like SSH has to remember keys (other than adding people to your trusted root). But you would have to design the user interface for such a mechanism *extremely* carefully. Remember, when you put something in a web browser, it is going to get used by hundreds of millions of people who will trust it with their most important data. It would need to be truly idiot-proof, and that would be extremely hard as you'd have to get idiots to understand the subtle implications of their security-related actions.
It's a false sense of security, leading users to do stupid things. People tend to believe that encryption is a brick wall that will protect them against anything, and it's almost true when you do both encryption and authentication, like SSL. Trouble is, on a LAN such as a public WiFi network, doing man-in-the-middle is almost as easy as sniffing, so defending against one but not the other is hardly an improvement, and outweighed by the damage done by false user expectations. Encryption without authentication is just wrongheaded.
This is a good thing, and if I understand correctly Firefox 3 will be doing the same. If you think otherwise, you do not understand how SSL works and why encryption without authentication is worse than useless. Anyone can self-sign a certificate claiming to be your webmail provider, or Amazon.com, or whoever they want. If you click past the warning, you are owned, because encryption without authentication is *always* vulnerable to a man-in-the-middle attack. And if you don't believe man-in-the-middle attacks are a threat, you are a fool. Every time you are on the same LAN as other users who you don't completely trust, you are vulnerable. Additionally, you are vulnerable to every ISP in the chain between you and your destination.
Now that browsers are making it harder for users to bypass the security warnings, there will be a lot more pressure on sites to fix the problems instead of instructing users to click past the warnings, so you will see less of these errors in the future, and the web will be a safer place for everyone.
Well, there was also the fact that his stories tended to be hype-filled press-release copy-paste jobs with sensationalized headlines but little real information, while the subject matter was typically mundane vaporware products or crackpot science. And yet somehow, his stories were accepted at an alarming rate. To his credit, lately they have been better. (Note that while his main story links are now direct, he still links to his blog from his name).
The wiggle room there is that statutory damages are used because calculating the actual damages is hard. The song costs $1, but the defendant probably uploaded it multiple times (and with the expectation that those copies would be further distributed). Therefore the damages are a *multiple* of $1. The RIAA would probably argue that the multiple could easily be 75 or more, making the damages fall under the 10:1 threshold you quote.
But I think a little critical thinking could topple that argument. It's easy to see that across a P2P system, the average number of times each copy of a song has been uploaded is 1. The number of uploads is by necessity exactly equal to the number of downloads. So in order for it to be plausible that the defendant uploaded a particular song 75 times, the defendant would have to be uploading that song *much* more than average. I don't think there's any evidence to prove that, though I might be wrong.
The RIAA might then argue that any uploading by the defendant was done with the expectation that the recipient would further upload the song to other people, and therefore uploads done by those people should factor in. Also, if this argument succeeded in reducing the damages, the RIAA would start collecting evidence on things like connection speed and uptime, and start suing people on fast pipes with 24/7 availability, using that as evidence that they uploaded more than average. There are plenty of those people to sue, and taking them out preferentially would kill P2P networks faster anyway.
Oh, and I forgot to mention that PulseAudio has aspirations to become "Compiz for audio", providing earcandy effects such as surround-sound positioning for on-screen events (so sounds from a window on the left of the screen come from the left speaker, etc) and muffled sound from background windows (so the Flash ad in Firefox's background tab doesn't blast your eardrums and the new-mail notification doesn't sound over the movie you're watching full-screen).
Thankfully, it does appear that PulseAudio is the One True sound server that we can all finally agree on. It emulates esd, OSS, and ALSA, so legacy apps like Flash and your smartphone work. It supports hotplug of audio devices, including networked ones (using Zeroconf even). It supports synchronized output between multiple devices, even when those devices use different sampling rates or have out-of-sync clocks (it resamples automatically). It has a zero-copy low latency architecture, taking advantage of the latest high resolution timer and real-time scheduling capabilities in new Linux kernels (when available), and it supports latency measurement for sound/video sync even when high latency is unavoidable (such as over a network). It has a modern user interface that provides per-application volume sliders like Windows Vista, and allows on-the-fly routing of audio to devices, including "saving" audio streams to another device if the device they are using is unplugged.
The guys behind PulseAudio really "get it". They even decided to drop their typically-awful open-source project name "PolypAudio" in favor of the infinitely better "PulseAudio", for wider acceptance. You've got to give them points for that; the GIMP could learn a thing or two from them.
Time machine backs up to a different disk, so it can save you in the event of a hard disk crash or other catastrophic event. Time Machine has a cool interface that is deeper than just a starfield wallpaper and some 3D effects. For example, you can take a spotlight search back in time to see what it would have found in the past.
In some ways Time Machine is not as cool as shadow copies, because it doesn't keep every version of the file with disk-sector-level copy-on-write; it does backups at specified times on its own schedule, and if a file is modified at all it backs up a whole new copy. If someone integrated shadow copies with the time machine interface, and also kept the features that back up to a different disk, then it would be the perfect all-singing, all-dancing backup tool.
You miss his point; 802.11g, and even n, are just as bad as 802.11b for truly large-scale deployment. Wi-Fi just isn't designed for deployment over anything larger than a business or university campus, and it shows. 802.16 is WiMAX, a completely different standard, and it does require buying extra hardware because no laptops currently have it (though they will in the future). But remember, these people already bought $500-$2000 laptops; I think a $100 USB dongle for ubiquitous Internet access is a no-brainer purchase. If you build a municipal WiMAX network and offer cheap or free access, convincing people to buy the hardware is the *least* of your worries.
Those bots weren't following the road markings, they were following the invisible GPS waypoints DARPA provided, which were much more detailed and accurate than you'd find in any car navigation system today. At best, their road-finding algorithms were fixing the last meter of inaccuracy in their differential GPS. And did you actually watch the bots as they traversed that section? I don't think any of them went over 10 MPH with frequent braking. Even CMU looked quite sluggish. A person could easily do 20-30 over such a flat and well-defined road, and that's not even in a race.
Firstly, a car you can't ever drive would never sell in the US. People want control, they want the ability to drive off-road even if they never actually do (see SUVs), and they love their older cars too much to stop driving them. Secondly, even if every car was automated, that would only take care of a *few* of the problems faced by automated vehicles. They would still have to deal with all of the problems that are caused by things other than unpredictable drivers, such as: wind, rain, snow, ice, fog, loss of GPS, worn or obscured road markings, people walking in the road, things that fall from trucks on the freeway, tires that blow out, malfunctioning traffic signals, downed power lines, mechanical failures of all kinds, collapsed bridges, avalanches, sinkholes, people trying to trick the robot sensors, and all the other problems I didn't happen to think of just now. If you really want your robotic car to be 100% safe, you have to program it to handle so many varied situations that I believe programming it with traffic rules for safe driving around humans would be a relatively small part of your work.
Now it's true that you could drive more efficiently without humans, but that will have to be phased in gradually. For example, you could have special robot lanes, and perhaps eventually entire robot-only streets in big cities. But that would only be possible *after* the introduction of autonomous vehicles.
No, no open source code. But what the public does get out of this is advances in technology. Case in point: the *real* winners of this year's Urban Challenge are Velodyne. Their lidar sensor was invented by team DAD for the 2005 challenge. For the 2007 challenge, they decided that instead of losing the competition again, they would sell their lidar technology to the other teams. Over half of the 35 teams in the challenge bought one, and 5 of the 6 finishers (Virginia Tech being the exception).
This thing is a huge advance over previous technology for this application, and it directly owes its existence to this challenge. Thanks to DARPA, you can now buy a lidar that you can stick on top of a car and which gives you 360 degree range data in 3D at 10 Hz over Ethernet. Now that the company is jump-started, next year those specs will improve, costs will go down, and eventually something like this will be driving your car for you. That's the benefit everyone gets from this competition. Not to mention all the people whose imaginations have been captured by the competition; who have been working on the funding DARPA gave out, getting their PhDs, or even just working in their spare time, learning how to write the software to run these things. There's no doubt in my mind that DARPA has gotten far more mileage from their money in this contest than they would have dumping it in the accounts of some defense contractor.
So even though no open source was produced from the contest, the public will see a lot of benefit from the money DARPA has spent.
Not because monopolies naturally arise on their own (they don't). We have anti-monopoly laws to fix the monopolies that arise because of *other* laws. In Microsoft's case, the other laws in question are the copyright laws. Microsoft is a monopoly because the government *grants* them a monopoly on their product; that's the express intent of the copyright laws (and the patent laws too). We shouldn't be surprised when these laws achieve their goals. The music and media cartels are similar products of government-granted copyright monopolies. If that isn't what we wanted after all, well then maybe the law ought to be changed.
If the government really wanted to end Microsoft's monopoly, they should have attacked the problem at its source: copyright. Consider: if the government voided Microsoft's copyright on Windows and/or Office, their dominance of the market would quickly end. For a less extreme remedy, Microsoft's copyrights could be restricted in various ways or made subject to various compulsory licenses.
I think your justification for not using placeholders is rather, uh, wrong. I agree that the argument order thing can be an issue, but that's what *named* placeholders are for. The major benefit of placeholders is not speed, it's absolute resistance to SQL injection. You may be diligent in quoting, but standard software development wisdom is that it's always better to eliminate the possibility of a bug than rely on programmer checking all the time, due to Murphy's Law of course. Also, the admittedly small speed benefit of placeholders when used with precompiled statements is going to be on the DB side, not the web side, which can make a bigger difference (as the DB is harder to scale).
This is not a class action suit and there's no law firm. This is an investigation by a state Attorney General. Verizon is paying a hundred or so thousand dollars to the state of New York in penalties, and they estimate a million to actual customers, plus changing their advertising *and* they have stopped terminating people's contracts (though that isn't an explicit requirement of the settlement).
As someone whose service actually *was* terminated by Verizon and who stands to be reimbursed for the cost of his hardware, I am pretty happy with this turn of events. I do wish the penalty was larger, because I have a sneaking suspicion Verizon has saved at least that much by terminating their most expensive customers, but I didn't really expect to get a resolution at all so I am pleasantly surprised by the effectiveness of the justice system.
ATI committed to providing complete documentation, but did not provide the source code to their current driver. This is (partially) because their current closed driver contains proprietary IP that ATI does not own, so they can't open-source it. NVIDIA is in the same situation and will have to take the same slow route to open-source drivers when they eventually come to their senses. They cannot "open up any day now" and "have a working code base" instantly, any more than ATI could.
VMs add practically zero overhead if they are implemented with cooperation from the OS being virtualized. A new OS that has the explicit goal of virtualizing legacy Windows could do a near-perfect job, ensuring compatibility while allowing Microsoft to end work on the giant legacy codebase that even they admit has become a nightmare.
This approach would be similar to what Apple did with the Classic environment in OS X, but with advanced features similar to Parallels to make the virtualization seamless. If Microsoft's Windows unit has any sense, Windows 7 will work this way, with a new kernel based on some of the great work coming out of Microsoft Research (I'm not holding my breath).
Au contraire; now that the SDK is announced I am much more likely to buy one. I only want to know one more thing: will I be able to freely upload my own native unsandboxed code to my own iPhone/iPod, without paying Apple and then having to get a digital signature every time? If the answer is yes, I will buy an iPod Touch tomorrow.
I think it's worth noting that our current space efforts (ISS, moon and mars bases) are not remotely "economical" either. If we must spend billions of tax dollars on space missions, I think they would be better spent doing something that might conceivably free us from oil dependency and benefit the entire human race here on Earth, than on a manned Mars mission (for example). The main practical benefit to the human race of a Mars mission, if there is one, is as a step toward mining and colonization of Mars and eventually even interstellar travel, but those goals are even farther away than space-based solar power, if they are even feasible.
I am aware that the term "IPTV" is commonly used to refer specifically to delivering a legacy cable-style user interface to TV delivered over IP, but ABC.com is also delivering TV over IP and I think it is stupid to call it something else just because the term "IPTV" was taken. Anyway, pedantic debates over the definition of the term "IPTV" are not the point here, and neither are the implementation details of Fios. The point is that high-quality video can be delivered over the Internet without cooperation from ISPs, as ABC.com is proving right now. We don't need ISPs meddling with our packets.
IPTV can't possibly work without QoS, huh? I personally am using IPTV all the time right now, provided by ABC.com, NBC.com, Comedy Central, Joost, iTunes, and yes, even YouTube. None of these services are provided by telecoms, and all are high enough quality for me in the *complete absence* of QoS. Whoops! Turns out QoS isn't quite as vital as you thought. OTOH, without net neutrality, it is likely cable companies will begin throttling video packets (and phone companies too as they try to move into the triple-play market). Given a choice between ISP-enforced QoS and net neutrality, I choose net neutrality.
First of all, TV over IP is here, and it works right now without any "dedicated bandwidth", despite doom and gloom from ISPs. I've been watching shows from NBC.com and ABC.com and it's worked quite well for me. My connection isn't fast enough for streaming HD, but ABC does offer it and I hear from other people that it works well. Joost works less well for me, but I suspect that's because my upload is crappy (Joost is P2P).
But I'll give you the benefit of the doubt and concede that maybe some kind of shaping is helpful (maybe if you're running BitTorrent while watching streaming TV and talking on Skype). Even so, that kind of shaping can and should be done on the consumer's side of the last mile. If you want glitch-free TV, you can set the router sitting in front of your cable modem to give TV priority over your other traffic. The ISP can even provide you a router with this capability pre-set; no configuration necessary. ABC.com could provide instructions and/or software for configuring routers, and future home routers could attempt to automatically classify and shape traffic (this could be a differentiating factor in the home router market). Doing shaping inside the ISP, where the customer has no control over it, is stupid and wrong. It can only work to benefit the ISP's interests and lock out competition for their services. No ISP is going to configure their routers to favorably shape traffic to services competing with their own offerings.