I'm guessing that a larger share of such work is being done at research universities, by math or CS people looking to publish academic papers.
I can neither prove nor disprove that assertion. The fact than H.264 is encumbered by a ton of patents tells me that many non-academic parties were involved. Having worked in the video industry, I know tons of companies just implement the standards, they don't try to extend them.
I don't agree. I think an algorithm is more like a device and fits the patent model better. The flow chart and quantization matrix were the important parts, not the characters in a file. Besides, do you want h.264 copywritten for the next 75 years?
Not really. Math research was alive and well before software patents came about. And implementing a mathematical algorithm in software... well, isn't that the obvious fricking point of a computer?
My point is, with the prospect of licensing revenue, math research is aliver and weller.
So the benefit to society is we get a 2160i video standard this decade, not next. Is that worth it?
Maybe we get, maybe we don't. In the meantime, software patents are screwing a lot of people over who are just marginally tied to some software. I have to figure out whether we have to rearchitect our entire video delivery platform because I don't know how much the patents and royalties on h.264 are going to screw us over. That's a real cost.
I'm not up on the details, but I thought that was the whole point of MPEG-LA, one stop shopping for all your video licensing fees. I'm much more concerned about everyone else. I have no idea if any software I write infringes on some submarine patent. I agree, that's a real cost. But I think the solution is not to toss all software patents but instead to tighten the requirements before issuing one.
Maybe. Sure, lots of companies want better compression (Google, Netflix, Apple to name a few). Is that enough? I don't know. My point is, there are a ton of other small, no-name players who can contribute, but only if they have the prospect of licensing revenue down the line. With out patents, at least some of them will not bother doing the research, so the pace of innovation will slow down. The judgment call is whether that speedup is worth collateral damage.
Also, don't get me wrong. A lot of software patents are asinine. My company encourages us to file patents to build a portfolio. We don't generally intend to license any of them, it's only a defense when we get sued for infringement. It's silly, but that's the way the system works today. I'd be perfectly happy setting a much higher bar for what is obvious, plus requiring a working implementation.
It's defensible because someone had to do the research to figure out the H.264 algorithms. In retrospect, it's easy to say "Duh, of course quarter-pixel motion estimation is a good idea", but someone had to do a lot of grunt work to prove that's really the case.
I'm quite certain math geeks are beavering away at new compression algorithms in corporate labs. Much of that research will screech to a halt if there's no prospect of making money licensing the resulting patents. Not all of it, just a lot. So the benefit to society is we get a 2160i video standard this decade, not next. Is that worth it? I don't know, maybe, but it's not cut and dried.
All the DVR did was replace the magnetic tape storage with magnetic disk storage. Nothing revolutionary... it was an evolutionary change.
What was revolutionary about a DVR vs. VCR is (a) you can record while watching and (b) it's much more reliable (since you don't have to pre-position the blank tape). All DVRs also include a program grid, something that I've never seen in a VCR. That's not really required, per se, it's just one of those things that was possible in the timeframe DVRs were introduced but not when VCRs came out.
Anyway, having used both, a DVR really is a different experience, not an evolution, much like a bird evolved from a dinosaur but really isn't one any more.
As long as you don't mix spaces and tabs, you're fine.
That's where things don't work for me. I don't like to always line things up on rigid tab stops. But you're right, if you always use only one or the other, things work out fine.
It's things like this that convinced me using spaces is the only way to fly. You also need an editor that knows how to indent to the right place by inserting the right number of spaces (i.e. not vi out of the box or notepad). Anything else will always get screwed up by someone or some tool.
Let's be clear on this situation: HDD makers, instead of making larger HDDs would rather spin the numbers to make them appear larger instead of actually being larger.
I don't think disk drive makers are avoiding making larger disks, they just want to promote the disks in the best light possible. And when it comes down to it, changing units doesn't change the number of sectors on the platter.
So long as we're all clear on which units are being used, either one is fine. Since most humans don't know the binary units, and there's no natural reason why the number of sectors on a disk should match power of two boundaries, I'm find with using the more common decimal prefixes.
Plus, I've got five extension phones scattered around the house so I don't have to run around like a madman to find the mobile when it rings (not that I'd have a chance of hearing it ringing three rooms away).
This is a perfectly reasonable view, so long as they made it clear up front that you're buying a limited time, limited rights license, not a perpetual license. I don't think that's what most consumers thought they were buying. I think most consumers expected they were buying something which would work forever.
I was chatting about this this morning. Assume you trust a DRMed music retailer, e.g. Apple. Do you really think their DRM servers will be up and running 20 years from now? And that you won't have bought more than 7 (or however many) computers or iPodsby then? I have CDs I bought in the late 80s that still play fine, so expect my mp3s will too. Us techies all realized this was a problem from the beginning, but it's totally unreasonable to think non-geeks have internalized this.
Not to dis Google or anything, but many people in the Bay Area hire goats for munching grass. There was a herd (probably the same one) "mowing" a hill near my office in Sunnyvale last week.
This works well around here. Wild grass grows only from December to April, at which point it dries and turns golden brown (to goats: golden, brown and delicious.) If they eat it now, it doesn't regrow until next winter.
Hear, hear! That's exactly the real consumer issue.
If there is no real competition for broadband service, no kidding prices will go up and service will go down. When there is real competition (think dialup in the '90s or long distance phone service since the '80s), prices tend to go down and service quality improves.
The only government action required it to ensure there is robust competition to provide broadband access to most homes.
I think the point might be that perhaps the school should deal with a disruptive child in a more constructive fashion than kicking them out without needing a full-blown autism diagnosis. How 'bout a little middle ground?
OTOH, not knowing the details, perhaps they already did. Certainly, at some point they need to consider the other children in the class and take action.
I don't know how notifying customers would help. "In 90 days, some number of web sites won't be reachable from your mobile phone. We don't know which ones or even how many. To avoid this, switch to another cell service provider. Have a nice day." I'm a geek and I don't know how to use this information.
When it's all said and done, I think the free market worked. Sprint cuts Cogent, customers bitch, Sprint sees the light and decides losing subscribers isn't worth the $100k a month, reconnects Cogent. How was this not a success?
The best AT&T offers to my house is a half-assed 1.5 Mbps download. Given that bitrate, I can either wait forever to download a HD movie (best case would be something like 6-8 hours) or watch it now with much lower video quality. Even DVD quality video won't stream, I'd have to download the entire thing before watching. More buffering won't help because it's downloading at something like 1/4 to 1/15 of the real-time bitrate.
I watch a lot of stuff from NetFlix using my laptop. I totally expected the video quality would be too poor to watch, but was quite surprised that it's good enough (but just barely). It's about the same as a beat up old VHS tape, if you're old enough to remember that.
Actually this could be a very sweet deal for developers. Perhaps, but what's annoying is that the developer doesn't get to decide. Maybe I decide Apple is offering a great deal. Maybe I've already got all the infrastructure lying around and the incremental cost to me is close to zero, in which case Apple's deal stinks. With no free market, there's no pressure on Apple to offer a better deal.
Granted, there's a large, sorta open market for mobile apps on smart phones which aren't controlled by Apple, so if you define "free market" broadly enough, this isn't a problem. But as a consumer, I'd prefer there to be a free (read: not controlled by Apple) market for iPhone software.
but because the hydrogen -> energy process is essentially the reverse of the water -> hydrogen process that will never happen due to the laws of thermodynamics. Sure, but that's not what I'm suggesting. Take gasoline or ethanol, crack hydrogen off that, use the hydrogen to run a fuel cell, which makes electricity to drive the wheels. Don't bother cracking water because it isn't useful or necessary. Sure, it sounds inefficient, but internal combustion engines are insanely inefficient too.
Let me throw out some imaginary numbers. Say an internal combustion engine is 40% efficient. Suppose cracking gas->hydrogen is 80% efficient (which I doubt), and the fuel cell is another 20% energy loss. That means the gas->hydrogen->fuel cell process has a 64% net efficiency, which beats the internal combustion engine. Those numbers are certainly incorrect, but if we're lucky, this is how the numbers might work out.
I don't get it. I can burn gas to run a generator to produce electricity to crack water so the hydrogen can run a fuel cell to make electricity to turn the wheels. Or I can skip the whole hydrogen step and just drive the wheels directly (either from the engine or using electric motors). It seems doubtful to me that producing hydrogen from water in a car is has a point.
Now, producing hydrogen from either gasoline or ethanol and using that to drive the wheels, that's a different story. You'd have to run the numbers to see if that's more or less efficient than internal combustion.
They should watch it as the people who created intended it, first and foremost. Well, that's certainly your choice, but why would you force it upon others? By all means, watch it as the director intended if that's what you like, but I may have valid reasons for not wanting to do that. What those reasons are are is nobody's business but my own. Lemme give you a far-out hypothetical. Suppose my SO had persistent genital arousal disorder so she's always on the verge of orgasm. Perhaps seeing a sex scene sets her off and she doesn't like that. Perhaps she still likes explosions and violence. Seems like trimming a gratuitous sex scene of questionable plot value is a reasonable choice.
No go ahead and watch it as you will with the tools you have at your own disposal... But these companies essentially were a tool at my disposal. They're just saving me the trouble of ripping/editing/re-authoring the DVD by myself. I never bought one of these DVDs, but my understanding was you had to buy the original DVD first (so you can't claim the original creators weren't getting their fair compensation), then charging an additional fee to edit out the naughty bits.
In the end, I believe this comes down to freedom. The creator has the right to produce the work they want and sell it or not. Once they sell it to me, I have the right to enjoy it as I want.
I got my first phone late last year. I'm a cheap bastard and wanted the same thing -- a cheap phone to make calls with. No camera, no texting, no web browsing, just a phone.
Virgin Mobile had the best deal I could find. I got a phone for something like $50, no service contract, the service is $7 a month and I pay $.10 a minute for calls. I think my average monthly bill is around $10-$15. The phone actually has text messaging and some "Virgin Connect" thingy that I can't figure out how to disable, but I just ignore that part.
Slashdot Mirror
I'm guessing that a larger share of such work is being done at research universities, by math or CS people looking to publish academic papers.
I can neither prove nor disprove that assertion. The fact than H.264 is encumbered by a ton of patents tells me that many non-academic parties were involved. Having worked in the video industry, I know tons of companies just implement the standards, they don't try to extend them.
Then they should copyright it.
I don't agree. I think an algorithm is more like a device and fits the patent model better. The flow chart and quantization matrix were the important parts, not the characters in a file. Besides, do you want h.264 copywritten for the next 75 years?
Not really. Math research was alive and well before software patents came about. And implementing a mathematical algorithm in software... well, isn't that the obvious fricking point of a computer?
My point is, with the prospect of licensing revenue, math research is aliver and weller.
So the benefit to society is we get a 2160i video standard this decade, not next. Is that worth it?
Maybe we get, maybe we don't. In the meantime, software patents are screwing a lot of people over who are just marginally tied to some software. I have to figure out whether we have to rearchitect our entire video delivery platform because I don't know how much the patents and royalties on h.264 are going to screw us over. That's a real cost.
I'm not up on the details, but I thought that was the whole point of MPEG-LA, one stop shopping for all your video licensing fees. I'm much more concerned about everyone else. I have no idea if any software I write infringes on some submarine patent. I agree, that's a real cost. But I think the solution is not to toss all software patents but instead to tighten the requirements before issuing one.
Maybe. Sure, lots of companies want better compression (Google, Netflix, Apple to name a few). Is that enough? I don't know. My point is, there are a ton of other small, no-name players who can contribute, but only if they have the prospect of licensing revenue down the line. With out patents, at least some of them will not bother doing the research, so the pace of innovation will slow down. The judgment call is whether that speedup is worth collateral damage.
Also, don't get me wrong. A lot of software patents are asinine. My company encourages us to file patents to build a portfolio. We don't generally intend to license any of them, it's only a defense when we get sued for infringement. It's silly, but that's the way the system works today. I'd be perfectly happy setting a much higher bar for what is obvious, plus requiring a working implementation.
It's defensible because someone had to do the research to figure out the H.264 algorithms. In retrospect, it's easy to say "Duh, of course quarter-pixel motion estimation is a good idea", but someone had to do a lot of grunt work to prove that's really the case.
I'm quite certain math geeks are beavering away at new compression algorithms in corporate labs. Much of that research will screech to a halt if there's no prospect of making money licensing the resulting patents. Not all of it, just a lot. So the benefit to society is we get a 2160i video standard this decade, not next. Is that worth it? I don't know, maybe, but it's not cut and dried.
I got a crazy idea: how about charging the same rate, regardless of which artificial political boundary they cross?
All the DVR did was replace the magnetic tape storage with magnetic disk storage. Nothing revolutionary... it was an evolutionary change.
What was revolutionary about a DVR vs. VCR is (a) you can record while watching and (b) it's much more reliable (since you don't have to pre-position the blank tape). All DVRs also include a program grid, something that I've never seen in a VCR. That's not really required, per se, it's just one of those things that was possible in the timeframe DVRs were introduced but not when VCRs came out.
Anyway, having used both, a DVR really is a different experience, not an evolution, much like a bird evolved from a dinosaur but really isn't one any more.
As long as you don't mix spaces and tabs, you're fine.
That's where things don't work for me. I don't like to always line things up on rigid tab stops. But you're right, if you always use only one or the other, things work out fine.
It's things like this that convinced me using spaces is the only way to fly. You also need an editor that knows how to indent to the right place by inserting the right number of spaces (i.e. not vi out of the box or notepad). Anything else will always get screwed up by someone or some tool.
How about they divert snow from Moscow first?
Let's be clear on this situation: HDD makers, instead of making larger HDDs would rather spin the numbers to make them appear larger instead of actually being larger.
I don't think disk drive makers are avoiding making larger disks, they just want to promote the disks in the best light possible. And when it comes down to it, changing units doesn't change the number of sectors on the platter.
So long as we're all clear on which units are being used, either one is fine. Since most humans don't know the binary units, and there's no natural reason why the number of sectors on a disk should match power of two boundaries, I'm find with using the more common decimal prefixes.
Plus, I've got five extension phones scattered around the house so I don't have to run around like a madman to find the mobile when it rings (not that I'd have a chance of hearing it ringing three rooms away).
This is a perfectly reasonable view, so long as they made it clear up front that you're buying a limited time, limited rights license, not a perpetual license. I don't think that's what most consumers thought they were buying. I think most consumers expected they were buying something which would work forever.
I was chatting about this this morning. Assume you trust a DRMed music retailer, e.g. Apple. Do you really think their DRM servers will be up and running 20 years from now? And that you won't have bought more than 7 (or however many) computers or iPodsby then? I have CDs I bought in the late 80s that still play fine, so expect my mp3s will too. Us techies all realized this was a problem from the beginning, but it's totally unreasonable to think non-geeks have internalized this.
Charcoal is very stable and won't re-enter the atmosphere for millions of years.
Not to dis Google or anything, but many people in the Bay Area hire goats for munching grass. There was a herd (probably the same one) "mowing" a hill near my office in Sunnyvale last week.
This works well around here. Wild grass grows only from December to April, at which point it dries and turns golden brown (to goats: golden, brown and delicious.) If they eat it now, it doesn't regrow until next winter.
Hear, hear! That's exactly the real consumer issue.
If there is no real competition for broadband service, no kidding prices will go up and service will go down. When there is real competition (think dialup in the '90s or long distance phone service since the '80s), prices tend to go down and service quality improves.
The only government action required it to ensure there is robust competition to provide broadband access to most homes.
I think the point might be that perhaps the school should deal with a disruptive child in a more constructive fashion than kicking them out without needing a full-blown autism diagnosis. How 'bout a little middle ground?
OTOH, not knowing the details, perhaps they already did. Certainly, at some point they need to consider the other children in the class and take action.
We did this in a previous lab. When the eighth and ninth servers showed up, we had to add the little know dwarves Sleezy and Queezy.
When it's all said and done, I think the free market worked. Sprint cuts Cogent, customers bitch, Sprint sees the light and decides losing subscribers isn't worth the $100k a month, reconnects Cogent. How was this not a success?
I watch a lot of stuff from NetFlix using my laptop. I totally expected the video quality would be too poor to watch, but was quite surprised that it's good enough (but just barely). It's about the same as a beat up old VHS tape, if you're old enough to remember that.
Granted, there's a large, sorta open market for mobile apps on smart phones which aren't controlled by Apple, so if you define "free market" broadly enough, this isn't a problem. But as a consumer, I'd prefer there to be a free (read: not controlled by Apple) market for iPhone software.
Let me throw out some imaginary numbers. Say an internal combustion engine is 40% efficient. Suppose cracking gas->hydrogen is 80% efficient (which I doubt), and the fuel cell is another 20% energy loss. That means the gas->hydrogen->fuel cell process has a 64% net efficiency, which beats the internal combustion engine. Those numbers are certainly incorrect, but if we're lucky, this is how the numbers might work out.
Now, producing hydrogen from either gasoline or ethanol and using that to drive the wheels, that's a different story. You'd have to run the numbers to see if that's more or less efficient than internal combustion.
In the end, I believe this comes down to freedom. The creator has the right to produce the work they want and sell it or not. Once they sell it to me, I have the right to enjoy it as I want.
The three most important words in science are not "I've got it!" but "Gee, that's odd."
Virgin Mobile had the best deal I could find. I got a phone for something like $50, no service contract, the service is $7 a month and I pay $.10 a minute for calls. I think my average monthly bill is around $10-$15. The phone actually has text messaging and some "Virgin Connect" thingy that I can't figure out how to disable, but I just ignore that part.