Voice data just CAN'T be securely encrypted. That's because the spacetime information HAS to be there because we inherently interpret voice data according to these characteristics. Either you reveal this information in the stream, or you must increase the latency to the point that communication is impossible. If you want security, don't speak, WRITE, and use a cryptosystem that isn't a piece of shit.
I disagree. The problem pointed at in this article can be easily solved on many SIP endpoints. I spend all day working on VoIP phones from vendors such as Linksys, Polycom, Aastra, Cisco, and if I really have to snom. Most of these have an option where it'll just send blank full bitrate audio rather than the usual "put silence here" instructions on G.711 calls. In fact that is the default behavior on some, since it makes the latency a bit more predictable to have a constant-rate data stream. If you want to use a VBR codec, of course this is a problem, but don't act like it's impossible or even hard to solve. If you are concerned enough to encrypt your conversations, use a CBR codec. 64 kbit/sec is not hard to free up.
Hell with easily available, they're trivial to build. Less than $100 at Home Depot will get you everything needed to build something I consider nearly equal to a Volcano. If you already have a decent heat gun, you can cut more than half that cost out.
Taking a look at the PCL-R, depending on what the standards are I just may qualify as a psychopath, so I'll give you that, and I completely agree on the procedures likely going similarly to how you stated.
Good points. (bet you don't hear that often on the internet...)
You would have to be a cold-hearted asshole to not be affected, but when it comes down to it overall you are not making the choice to end those lives. They're already going to die one way or another at that point, it's merely a matter of whether they die over the ocean or potentially take out a small town in the process of crashing down.
That's how I look at it at least, if the button has to be pushed the lives of the astronauts are already written off and it's potential lives on the ground that the RSO is affecting. With that in mind, you're not making the decision to end 7 lives, just the decision to save possibly tens or hundreds.
A point that's been made quite a few times so far is that if they get far enough off course that the destruct sequence has to be kicked off, they're doomed to fireball status one way or another. It's just a matter of what else explodes with them.
Even as a fan of PHP, I have to agree that it's a language problem in some way.
Many PHP security issues come from how terrible the default settings were on old versions, and unfortunately for compatibility with old code those settings are still often set to the shitty mode, even occasionally by default.
If you configure PHP 5 as recommended it's pretty decent, but unfortunately most are still not set up that way. I hope that PHP 6 finally drops the crap like register_globals.
It's just a US thing. What we saw as SMB2 here was Doki Doki Panic in Japan. They just stuck characters from the SMB world in to another game and called it a day. If you've played "The Lost Levels" on Super Mario All-Stars, that's the real SMB2.
It's still not rocket science to "tunnel" the modem traffic with VoIP over to the location it's supposed to be. Hell, depending on how much of the system you control you can even spoof the CID and not have to tunnel it.
It is better to have companies competing, then to have them holding hands. While I agree with this in most markets, Internet connectivity doesn't really follow the same rules. With CDs, if one label decides to jack up their prices I can still get plenty of music from their competitors. On the other hand, I typically get the choice between Telco A and Cableco B for connectivity at most locations (don't get me started on the pointlessness of resi/SOHO satellite). If I go far enough out of the city, this might drop to one (or in the case of the city I live in, the cable company is absolutely inept and lacks any modern technology like HDTV or broadband speeds that don't belong in 1999).
Because of this lack of last-mile competition, I want the ISPs to cooperate with each other as much as possible. If Verizon (my current ISP) got in a fight with either Global Crossings or Level(3) and did the same thing, I would no longer be able to access parts of my work from home.
If there was real competition, where I could connect to a number of respectable providers at decent speeds (had to add that since there still technically is competition in dialup), then I would love to see the ISPs beating each other up. Until then...
The different Blu-Ray "profiles" require different hardware, which is obviously not something that can be fixed by a firmware upgrade.
Profile 1.0, otherwise known as the grace period profile, only required 64KB of local storage for key revocation lists.
Profile 1.1, which is the "final standard" profile (though it was only required for players released after 11/1/2007, leaving over a year of BD player production supporting an incomplete featureset) requires 256MB of local storage as well as secondary audio and video decoders to allow for PIP and overlay audio commentary.
Profile 2.0 adds networking and Internet connectivity to the mix and ups the local storage requirement to 1GB. This profile is equivalent to the features that have been mandatory in HD-DVD from day one.
The only upgradable hardware BD player is the PS3, since it already had the hardware for other purposes. Profile 1.1 support was pushed out in a software update soon after it became mandatory in standalone players and profile 2.0 support was announced yesterday and is expected some time next month.
If you're buying your cables from Best Buy or Radio Shack, no matter what it is it's crazy expensive because you're getting ripped off.
Take a look at monoprice or Blue Jeans Cables. Both are highly regarded on AVS Forum, SA's A/V Arena, and other large home theater forums while charging prices that have a lot more to do with reality. Last time I checked, a 25 foot HDMI cable at Best Buy was in the $200 range. The same length can be had for between $25 and $75 from Blue Jeans or $15 to $50 from monoprice. Both of those places offer multiple grades of cable that correspond with real specs and ratings rather than typical audiophile garbage, and the prices fit for the quality. The Monster Cable model sold by Best Buy will fall roughly equal to the $25-30 range parts from either of these companies.
As for price and quality, sure my Macbook Pro and my desktop both beat the 720p resolution of my projector, but I've spend a total of $2100 on the home theater including every game system but the Wii while also having HD Satellite, HD-DVD, Blu-Ray, and an Apple TV. Both of my computers combined total up to around $3000 and can not play HD movies or watch HDTV (I do have a USB analog tuner which works on both).
On top of that, a 15 or 20" screen, no matter how high resolution it is, coupled to desktop speakers can in no way compete with a 110" projected image being backed up by 1000 watts of 8 speaker surround sound.
Still, I know not enough about the mac to change to it, and it's not like you're buying a mac to try it. *cough*
This is what sold me on the mac.
When PearPC popped up I messed around with OS X a bit which eventually led to me resurrecting a shitty first-gen iMac i got for $20 to play with 10.3.
Then back in (late '05? early '06?) when the first hacked copies of OS X that ran on normal PCs started showing up, I loaded it on my Athlon 64 and played around a bit more...
Then I spend a bit of time in the Mac labs on campus and got used to it. Then a suitemate of mine went to sell his G4 Powerbook and I bought it.
September of '06 I decided I was done playing around, went up to the Apple Store near Detroit and bought the base Macbook along with an iPod nano on the student combo deal.
A few weeks back when the new Macbook Pro came out I decided to upgrade.
Emulation, then the ability to natively run it on my homebrew PC is what sold me on the Mac platform as a whole. OS 8.0-9.x turned me off to the Mac, 10.0 interested me, 10.2 got me checking it out, 10.3 got me using it, 10.4 saw me buying one, and the Intel switch got me in to a new one and as a full on Mac user.
I now have a 2.4GHz Core 2 Duo, 4GB of RAM, and a GeForce 8600 GT. This thing is more powerful than my desktop and can run Windows if I need it while keeping me in the wonderful Mac OS X when I want to get things done. A decent UI, commercial apps, bash, and X11. What more could I ask for?
I disagree. I have a computer that was built in 2005 for under $1000. Since then I've doubled the cores (2GHz Athlon 64 to 2GHz Athlon X2), doubled the RAM (1GB to 2GB), and bumped up the graphics card, going from the GeForce 6600GT which was mid-range when I built it to a GeForce 7900GT which was mid-range when I did the upgrade. The total cost for those upgrades was under $500, bringing the total spent on this machine to roughly $1500 over three years.
It still plays everything I throw at it without having to drop the resolution from the native 1680x1050 of my 20" LCD. Unreal Tournament 3, sure. Crysis, bring it on. It gets mostly used for Portal and Audiosurf, but it can still take more stressful games without a hitch.
I just specced out a similar (slightly better, some of my parts are no longer in production) system on newegg for $555. AM2 socked Athlon X2 versus my 939, nForce5 successor to my nForce4 mobo, DDR2-800 vs. DDR1-400, GeForce 8600GT vs. 7900GT, TruePower3 550w vs my TP2, and Barracuda 7200.10 HD vs my 7200.8
All except the video card should be within a few percentage points ahead of what my current machine has, I think the video card is a bit slower clocked but it has a newer generation GPU so if I remember correctly it trades off depending on the game. Older games are typically faster on mine, newer games are faster on the new one.
Anyways, I can't see any complaints about needing a PC costing in the mid-three digit range to play modern games. Bumping up to around $1000 will let you future-proof it by going up another level in the CPU and graphics card, as well as doubling the RAM to 4GB (note that fully utilizing this requires a 64 bit OS, so if you can't stand Vista prepare to deal with the half-baked abortion that is XP64. Since we're talking gaming I'll leave the three major unixes out of the discussion...)
When the Xbox 360 package that's actually worth buying runs $350 and the PS3 $400, I see nothing wrong with another $155 plus a bit of build time to have the option of better controls, higher resolution displays, and multitasking.
Note that I am not a console hater, with the exception of the aforementioned Portal and Audiosurf my desktop PC rarely gets touched anymore where either my 360 or my PS3 are on pretty much every day.
What I personally disagree with is Sweeney's whining about the fact that not every PC has a fast GPU. There are many uses for which you don't need to spend an extra $200 on a fast graphics card, compared to a chipset with integrated graphics. Actually, 90% of office PCs match that description. So there is a large market for PCs where a fast GPU would just be a waste of time. Get used to it, Tim... I think Tim's complaint is more that those PCs with integrated graphics are being advertised as having a "256MB* Intel Integrated Extreme 3D Graphics Accelerator" so the average person is tricked in to believing that their $250 Dell can actually play games which take more power than World of Warcraft.
I look at it sort of like the Vista Certified thing where it's true that those computers _can_ run the game, it's just at 640x480 with everything turned off while still only managing 4 FPS. If the integrated pieces of shit weren't advertised as if they are actually useful for modern games, we probably wouldn't have this problem.
If I had to compare internet telephony (and other internet services) to the standard set of traditional land-line TDM telephone service, I'd be tossing my desktop SIP phone (not to mention my cell phone) in the rubbish bin. If you notice a quality difference, your VoIP provider or internet connection sucks.
I work in the VoIP industry where part of my job is fixing quality problems. Our standard codec is G.711u, which is the exact same codec used on the digital trunk lines in the traditional telecom system. Given the same input, the bitstream should be identical and thus the output on the other end will be just as identical. We can even go beyond that. In the traditional telecom world a T1 means 24 calls maximum, end of story. If quality can be sacrificed, I can drop the codec down to G.729 or GSM and fit 3-5x as many calls in the same bandwidth. Both codecs are roughly equivalent to a good cell phone (or exactly equivalent in the case of GSM since again it's the same codec).
On the other side of things, we have G.722AB, which is marketed as "HD Voice" by Polycom. It takes up the same 64kbit chunk of bandwidth as G.711 but is a much more complicated design allowing for more than double the sampling rate and bit depth, thus significantly improved call quality. This can be used between compatible systems on private trunk lines, but you can't get the benefits when calling outside of your company without using VoIP. Calls in HD sound absolutely incredible, I was talking to someone in Chicago from Cleveland and it sounded like he was in the same room.
VoIP does benefit from much cheaper long distance and international, but it also has a lot more flexibility than traditional telecom systems and if it's done right the reliability is equal or better. I have customers that have gone up to two years between problems, including internet downtime. Amusingly enough, I've found a decent cable provider to be more reliable than a T1 over that time. DSL brings up the rear, tied with some of the shittier cable providers.
If Linux were even half as much better as people like you thought it was, business would be falling over themselves trying to save money using it.
I think you underestimate the PHB effect. My boss is practically falling over himself to get us "upgraded" to Windows Server 2008 from our current Debian setup simply because it has a familiar GUI so he can think he'll actually be able to administer the damn thing.
I've managed to hold him off for 6 months, we'll see how long I can keep it up.
Idiot bosses who fancy themselves skilled at IT plus wannabe admins with MCSEs probably account for the majority of Windows Server installs in my mind.
So...you're saying to use JPEG in a lossless mode? While I believe such a thing does exist, no one uses it and it would be pointless to argue.
Obviously going from anything to a lossless format, assuming the new format is capable of handling the number of channels, frequency, and bit depth of the original, will be lossless as long as there is not a major failure somewhere in the chain as that is pretty much the point.
What I'm saying is that with the exception of very simple lossy compression schemes if you go lossy -> lossy you will lose more data. In most cases the encoder has no idea what format the data was in previously, it simply sees the raw decoded output and whatever parameters were given to control the encoding process.
Now, I will concede that there may be some tools designed specifically for re-encoding at a lower quality which work within the already compressed data to reduce the size with less or no impact from existing compression artifacts, but those would have to be designed specifically for a format and the efficiency of such a process would vary dramatically by the format chosen. The reduction in data will still cause more loss than if the original compression run had targeted that file size (as it could have optimized things better), but it could be better than just using the same compressor over again.
I fail to see how you expect it to be possible to convert from lossy to lossless and back to lossy without losing more information when discussing modern compression algorithms.
Maybe if we're talking something like simple audio compression (I think G.711u as used in the telephone system counts here) or dithering to 256 colors for GIF compression where it just drops off data that doesn't meet certain requirements...
If I save an image as a JPEG or an audio file as MP3, artifacts are introduced. If I then save that file back as a PNG or WAV, it's now in lossless form but it still has the artifacts and missing data. When I then go back to save it in its respective lossy format, now the compressor has to deal with the artifacts and in the process introduces new artifacts as well as discarding more data to make room for the data describing the previous artifacts.
Transcoding from one lossy format to another or even re-encoding with the same encoder is as bad as if not worse than copying analog tapes in terms of quality loss. The first one will usually be reasonable, the second bearable, and beyond that it just goes downhill dramatically.
I am certainly not a lawyer, but here's how I understand the situation:
GeForce 8 cards have had CUDA support from day one. nVidia bought Ageia, and with it all they need involving the PhysX API. This upcoming download to enable physics acceleration will be a PhysX-to-CUDA wrapper that is in no way locked down to the Geforce 8 architecture (which is the point of CUDA).
By my understanding of SarbOx (which admittedly is not great) this falls under the same category as programs being written for an Intel processor. It certainly adds value, but since the ability to program it was in the design from the beginning it shouldn't pose any accounting problems.
A dynamic IP hasn't been a decent protection against any form of attack for ages. Sure, back when most attacks were directed at a specific target it might have helped, but for at least the last few years the kind of threats targeting the average Windows XP (or any other desktop OS for that matter) user are automated and basically use the shotgun approach against entire subnets at a time.
For example, drop an unpatched Windows XP RTM box on the internet, no firewalls or anything, and watch it get infected within seconds. No one knew you were going to put that box online, but the quantity of machines scanning the internet for new vulnerable targets is just so high that any publicly routable IP probably gets scanned by every single major worm at least once a minute.
I know looking through my logs that before I implemented DenyHosts and a blacklist preventing logins from outside the country all of my Linux boxes were getting over 100 root login attempts per minute from a few different hosts trying to brute force my password. They didn't get in since I had password login disabled, but the logs were annoying.
The page you link to appears to say otherwise. The Adobe case listed shows that the EULA doesn't apply until you actually agree to it (presumably by installing it) but the next case after that seems to have the clear result that once you have entered in to the license agreement the publisher can limit your rights as outlined in the license.
Given that Steam (and pretty much every other online digital content store I've ever seen) requires you to agree to the EULA before you can even get an account, you can't claim any of the excuses you could against physical EULAs.
I must admit I did not even think about a POTS line. I work in Voice over IP, so my home phone is delivered via that system and the endpoints I use allow me to reject a call easily. Before that, I only had a cell phone where as we all know the same is typically true. This has all but insulated me from ordinary land lines for quite a few years.
I guess you are correct and the vast majority of the population does have a system where you just have to ignore the ring, which does tend to interrupt things and lead to an uncomfortable silence (doubly so if you have an answering machine with call screening enabled so whatever the caller decides to say is now being broadcast to the entire group).
Actually, making a phone call is more "immediate" and intrusive than a text message. Try sitting in a group and just let the phone ring. Everyone will get antsy until the phone stops ringing. That's because you're an asshole if you just let your phone ring. Every phone I've ever owned can be easily muted blindly. In the case of the iPhone and the LG CU500 (both of which I have in front of me) I can say for certain it is as easy as pressing any of the side buttons while the phone is ringing. Hit the button and magically the phone stops ringing without me picking up the call.
and I don't know how someone can be so useless behind the wheel to do that. Yes, I've seen it too, but it's all laziness on their part. I drive a stick shift and often forget my Bluetooth headset, yet I still manage to use my signals when changing lanes in traffic. The signal stalk is designed to be convenient and unless you have really stubby fingers it's easy to activate it with your left hand while still using the same hand to steer.
Slashdot Mirror
Voice data just CAN'T be securely encrypted. That's because the spacetime information HAS to be there because we inherently interpret voice data according to these characteristics. Either you reveal this information in the stream, or you must increase the latency to the point that communication is impossible. If you want security, don't speak, WRITE, and use a cryptosystem that isn't a piece of shit.
I disagree. The problem pointed at in this article can be easily solved on many SIP endpoints. I spend all day working on VoIP phones from vendors such as Linksys, Polycom, Aastra, Cisco, and if I really have to snom. Most of these have an option where it'll just send blank full bitrate audio rather than the usual "put silence here" instructions on G.711 calls. In fact that is the default behavior on some, since it makes the latency a bit more predictable to have a constant-rate data stream. If you want to use a VBR codec, of course this is a problem, but don't act like it's impossible or even hard to solve. If you are concerned enough to encrypt your conversations, use a CBR codec. 64 kbit/sec is not hard to free up.Hell with easily available, they're trivial to build. Less than $100 at Home Depot will get you everything needed to build something I consider nearly equal to a Volcano. If you already have a decent heat gun, you can cut more than half that cost out.
Taking a look at the PCL-R, depending on what the standards are I just may qualify as a psychopath, so I'll give you that, and I completely agree on the procedures likely going similarly to how you stated.
Good points. (bet you don't hear that often on the internet...)
You would have to be a cold-hearted asshole to not be affected, but when it comes down to it overall you are not making the choice to end those lives. They're already going to die one way or another at that point, it's merely a matter of whether they die over the ocean or potentially take out a small town in the process of crashing down.
That's how I look at it at least, if the button has to be pushed the lives of the astronauts are already written off and it's potential lives on the ground that the RSO is affecting. With that in mind, you're not making the decision to end 7 lives, just the decision to save possibly tens or hundreds.
A point that's been made quite a few times so far is that if they get far enough off course that the destruct sequence has to be kicked off, they're doomed to fireball status one way or another. It's just a matter of what else explodes with them.
Even as a fan of PHP, I have to agree that it's a language problem in some way.
Many PHP security issues come from how terrible the default settings were on old versions, and unfortunately for compatibility with old code those settings are still often set to the shitty mode, even occasionally by default.
If you configure PHP 5 as recommended it's pretty decent, but unfortunately most are still not set up that way. I hope that PHP 6 finally drops the crap like register_globals.
It's just a US thing. What we saw as SMB2 here was Doki Doki Panic in Japan. They just stuck characters from the SMB world in to another game and called it a day. If you've played "The Lost Levels" on Super Mario All-Stars, that's the real SMB2.
It's still not rocket science to "tunnel" the modem traffic with VoIP over to the location it's supposed to be. Hell, depending on how much of the system you control you can even spoof the CID and not have to tunnel it.
I assumed PC players were considered a given.
Because of this lack of last-mile competition, I want the ISPs to cooperate with each other as much as possible. If Verizon (my current ISP) got in a fight with either Global Crossings or Level(3) and did the same thing, I would no longer be able to access parts of my work from home.
If there was real competition, where I could connect to a number of respectable providers at decent speeds (had to add that since there still technically is competition in dialup), then I would love to see the ISPs beating each other up. Until then...
The different Blu-Ray "profiles" require different hardware, which is obviously not something that can be fixed by a firmware upgrade.
Profile 1.0, otherwise known as the grace period profile, only required 64KB of local storage for key revocation lists.
Profile 1.1, which is the "final standard" profile (though it was only required for players released after 11/1/2007, leaving over a year of BD player production supporting an incomplete featureset) requires 256MB of local storage as well as secondary audio and video decoders to allow for PIP and overlay audio commentary.
Profile 2.0 adds networking and Internet connectivity to the mix and ups the local storage requirement to 1GB. This profile is equivalent to the features that have been mandatory in HD-DVD from day one.
The only upgradable hardware BD player is the PS3, since it already had the hardware for other purposes. Profile 1.1 support was pushed out in a software update soon after it became mandatory in standalone players and profile 2.0 support was announced yesterday and is expected some time next month.
If you're buying your cables from Best Buy or Radio Shack, no matter what it is it's crazy expensive because you're getting ripped off.
Take a look at monoprice or Blue Jeans Cables. Both are highly regarded on AVS Forum, SA's A/V Arena, and other large home theater forums while charging prices that have a lot more to do with reality. Last time I checked, a 25 foot HDMI cable at Best Buy was in the $200 range. The same length can be had for between $25 and $75 from Blue Jeans or $15 to $50 from monoprice. Both of those places offer multiple grades of cable that correspond with real specs and ratings rather than typical audiophile garbage, and the prices fit for the quality. The Monster Cable model sold by Best Buy will fall roughly equal to the $25-30 range parts from either of these companies.
As for price and quality, sure my Macbook Pro and my desktop both beat the 720p resolution of my projector, but I've spend a total of $2100 on the home theater including every game system but the Wii while also having HD Satellite, HD-DVD, Blu-Ray, and an Apple TV. Both of my computers combined total up to around $3000 and can not play HD movies or watch HDTV (I do have a USB analog tuner which works on both).
On top of that, a 15 or 20" screen, no matter how high resolution it is, coupled to desktop speakers can in no way compete with a 110" projected image being backed up by 1000 watts of 8 speaker surround sound.
This is what sold me on the mac.
When PearPC popped up I messed around with OS X a bit which eventually led to me resurrecting a shitty first-gen iMac i got for $20 to play with 10.3.
Then back in (late '05? early '06?) when the first hacked copies of OS X that ran on normal PCs started showing up, I loaded it on my Athlon 64 and played around a bit more...
Then I spend a bit of time in the Mac labs on campus and got used to it. Then a suitemate of mine went to sell his G4 Powerbook and I bought it.
September of '06 I decided I was done playing around, went up to the Apple Store near Detroit and bought the base Macbook along with an iPod nano on the student combo deal.
A few weeks back when the new Macbook Pro came out I decided to upgrade.
Emulation, then the ability to natively run it on my homebrew PC is what sold me on the Mac platform as a whole. OS 8.0-9.x turned me off to the Mac, 10.0 interested me, 10.2 got me checking it out, 10.3 got me using it, 10.4 saw me buying one, and the Intel switch got me in to a new one and as a full on Mac user.
I now have a 2.4GHz Core 2 Duo, 4GB of RAM, and a GeForce 8600 GT. This thing is more powerful than my desktop and can run Windows if I need it while keeping me in the wonderful Mac OS X when I want to get things done. A decent UI, commercial apps, bash, and X11. What more could I ask for?
I disagree. I have a computer that was built in 2005 for under $1000. Since then I've doubled the cores (2GHz Athlon 64 to 2GHz Athlon X2), doubled the RAM (1GB to 2GB), and bumped up the graphics card, going from the GeForce 6600GT which was mid-range when I built it to a GeForce 7900GT which was mid-range when I did the upgrade. The total cost for those upgrades was under $500, bringing the total spent on this machine to roughly $1500 over three years.
It still plays everything I throw at it without having to drop the resolution from the native 1680x1050 of my 20" LCD. Unreal Tournament 3, sure. Crysis, bring it on. It gets mostly used for Portal and Audiosurf, but it can still take more stressful games without a hitch.
I just specced out a similar (slightly better, some of my parts are no longer in production) system on newegg for $555. AM2 socked Athlon X2 versus my 939, nForce5 successor to my nForce4 mobo, DDR2-800 vs. DDR1-400, GeForce 8600GT vs. 7900GT, TruePower3 550w vs my TP2, and Barracuda 7200.10 HD vs my 7200.8
All except the video card should be within a few percentage points ahead of what my current machine has, I think the video card is a bit slower clocked but it has a newer generation GPU so if I remember correctly it trades off depending on the game. Older games are typically faster on mine, newer games are faster on the new one.
Anyways, I can't see any complaints about needing a PC costing in the mid-three digit range to play modern games. Bumping up to around $1000 will let you future-proof it by going up another level in the CPU and graphics card, as well as doubling the RAM to 4GB (note that fully utilizing this requires a 64 bit OS, so if you can't stand Vista prepare to deal with the half-baked abortion that is XP64. Since we're talking gaming I'll leave the three major unixes out of the discussion...)
When the Xbox 360 package that's actually worth buying runs $350 and the PS3 $400, I see nothing wrong with another $155 plus a bit of build time to have the option of better controls, higher resolution displays, and multitasking.
Note that I am not a console hater, with the exception of the aforementioned Portal and Audiosurf my desktop PC rarely gets touched anymore where either my 360 or my PS3 are on pretty much every day.
Actually, 90% of office PCs match that description. So there is a large market for PCs where a fast GPU would just be a waste of time. Get used to it, Tim... I think Tim's complaint is more that those PCs with integrated graphics are being advertised as having a "256MB* Intel Integrated Extreme 3D Graphics Accelerator" so the average person is tricked in to believing that their $250 Dell can actually play games which take more power than World of Warcraft.
I look at it sort of like the Vista Certified thing where it's true that those computers _can_ run the game, it's just at 640x480 with everything turned off while still only managing 4 FPS. If the integrated pieces of shit weren't advertised as if they are actually useful for modern games, we probably wouldn't have this problem.
I work in the VoIP industry where part of my job is fixing quality problems. Our standard codec is G.711u, which is the exact same codec used on the digital trunk lines in the traditional telecom system. Given the same input, the bitstream should be identical and thus the output on the other end will be just as identical. We can even go beyond that. In the traditional telecom world a T1 means 24 calls maximum, end of story. If quality can be sacrificed, I can drop the codec down to G.729 or GSM and fit 3-5x as many calls in the same bandwidth. Both codecs are roughly equivalent to a good cell phone (or exactly equivalent in the case of GSM since again it's the same codec).
On the other side of things, we have G.722AB, which is marketed as "HD Voice" by Polycom. It takes up the same 64kbit chunk of bandwidth as G.711 but is a much more complicated design allowing for more than double the sampling rate and bit depth, thus significantly improved call quality. This can be used between compatible systems on private trunk lines, but you can't get the benefits when calling outside of your company without using VoIP. Calls in HD sound absolutely incredible, I was talking to someone in Chicago from Cleveland and it sounded like he was in the same room.
VoIP does benefit from much cheaper long distance and international, but it also has a lot more flexibility than traditional telecom systems and if it's done right the reliability is equal or better. I have customers that have gone up to two years between problems, including internet downtime. Amusingly enough, I've found a decent cable provider to be more reliable than a T1 over that time. DSL brings up the rear, tied with some of the shittier cable providers.
If Linux were even half as much better as people like you thought it was, business would be falling over themselves trying to save money using it.
I think you underestimate the PHB effect. My boss is practically falling over himself to get us "upgraded" to Windows Server 2008 from our current Debian setup simply because it has a familiar GUI so he can think he'll actually be able to administer the damn thing.I've managed to hold him off for 6 months, we'll see how long I can keep it up.
Idiot bosses who fancy themselves skilled at IT plus wannabe admins with MCSEs probably account for the majority of Windows Server installs in my mind.
So...you're saying to use JPEG in a lossless mode? While I believe such a thing does exist, no one uses it and it would be pointless to argue.
Obviously going from anything to a lossless format, assuming the new format is capable of handling the number of channels, frequency, and bit depth of the original, will be lossless as long as there is not a major failure somewhere in the chain as that is pretty much the point.
What I'm saying is that with the exception of very simple lossy compression schemes if you go lossy -> lossy you will lose more data. In most cases the encoder has no idea what format the data was in previously, it simply sees the raw decoded output and whatever parameters were given to control the encoding process.
Now, I will concede that there may be some tools designed specifically for re-encoding at a lower quality which work within the already compressed data to reduce the size with less or no impact from existing compression artifacts, but those would have to be designed specifically for a format and the efficiency of such a process would vary dramatically by the format chosen. The reduction in data will still cause more loss than if the original compression run had targeted that file size (as it could have optimized things better), but it could be better than just using the same compressor over again.
I fail to see how you expect it to be possible to convert from lossy to lossless and back to lossy without losing more information when discussing modern compression algorithms.
Maybe if we're talking something like simple audio compression (I think G.711u as used in the telephone system counts here) or dithering to 256 colors for GIF compression where it just drops off data that doesn't meet certain requirements...
If I save an image as a JPEG or an audio file as MP3, artifacts are introduced. If I then save that file back as a PNG or WAV, it's now in lossless form but it still has the artifacts and missing data. When I then go back to save it in its respective lossy format, now the compressor has to deal with the artifacts and in the process introduces new artifacts as well as discarding more data to make room for the data describing the previous artifacts.
Transcoding from one lossy format to another or even re-encoding with the same encoder is as bad as if not worse than copying analog tapes in terms of quality loss. The first one will usually be reasonable, the second bearable, and beyond that it just goes downhill dramatically.
I am certainly not a lawyer, but here's how I understand the situation:
GeForce 8 cards have had CUDA support from day one.
nVidia bought Ageia, and with it all they need involving the PhysX API.
This upcoming download to enable physics acceleration will be a PhysX-to-CUDA wrapper that is in no way locked down to the Geforce 8 architecture (which is the point of CUDA).
By my understanding of SarbOx (which admittedly is not great) this falls under the same category as programs being written for an Intel processor. It certainly adds value, but since the ability to program it was in the design from the beginning it shouldn't pose any accounting problems.
A dynamic IP hasn't been a decent protection against any form of attack for ages. Sure, back when most attacks were directed at a specific target it might have helped, but for at least the last few years the kind of threats targeting the average Windows XP (or any other desktop OS for that matter) user are automated and basically use the shotgun approach against entire subnets at a time.
For example, drop an unpatched Windows XP RTM box on the internet, no firewalls or anything, and watch it get infected within seconds. No one knew you were going to put that box online, but the quantity of machines scanning the internet for new vulnerable targets is just so high that any publicly routable IP probably gets scanned by every single major worm at least once a minute.
I know looking through my logs that before I implemented DenyHosts and a blacklist preventing logins from outside the country all of my Linux boxes were getting over 100 root login attempts per minute from a few different hosts trying to brute force my password. They didn't get in since I had password login disabled, but the logs were annoying.
The page you link to appears to say otherwise. The Adobe case listed shows that the EULA doesn't apply until you actually agree to it (presumably by installing it) but the next case after that seems to have the clear result that once you have entered in to the license agreement the publisher can limit your rights as outlined in the license.
Given that Steam (and pretty much every other online digital content store I've ever seen) requires you to agree to the EULA before you can even get an account, you can't claim any of the excuses you could against physical EULAs.
IANAL and such
I must admit I did not even think about a POTS line. I work in Voice over IP, so my home phone is delivered via that system and the endpoints I use allow me to reject a call easily. Before that, I only had a cell phone where as we all know the same is typically true. This has all but insulated me from ordinary land lines for quite a few years.
I guess you are correct and the vast majority of the population does have a system where you just have to ignore the ring, which does tend to interrupt things and lead to an uncomfortable silence (doubly so if you have an answering machine with call screening enabled so whatever the caller decides to say is now being broadcast to the entire group).
and I don't know how someone can be so useless behind the wheel to do that. Yes, I've seen it too, but it's all laziness on their part. I drive a stick shift and often forget my Bluetooth headset, yet I still manage to use my signals when changing lanes in traffic. The signal stalk is designed to be convenient and unless you have really stubby fingers it's easy to activate it with your left hand while still using the same hand to steer.