Agreed. And I do have some evidence to back up why you're right. On my 11 hour flight yesterday the plane was crowded with people using their iPads, and they all seemed extremely satisfied with them. The couple in the isle next to me had one each and have been watching movies on it almost the entire freaking flight, all on a single charge, and surprise, without hearing a single complaint about how the 4:3 aspect ratio was bothering them or whatver. People were playing games on them, having fun, showing off stuff and just enjoying it in general. Somehow I don't have the impression they would have been happier with a Galaxy Tab or some other nerdy Android device with no tablet-optimized stuff whatsoever available for it.
How did they 'irrevocably give up their rights to enforce their patents' any more than MPEG-LA 'giving up their rights to levy royalties for non-profit use of H264'. In terms of hypothetical scenario's for getting screwed as a non-profit customer, Google and MPEG-LA are in the same boat.
As for the indemnification against patent claims, it's pretty obvious that you'd have to be epically stupid to invest heavily in WebM as a company, because Google is specifically saying (in the licensing terms of WebM), that as soon as you get involved in a patent lawsuit and try to settle so you can continue selling your products, the 'free' license you had on WebM will be revoked. Yes, that's true, go look it up: the licensing terms for WebM prohibit settling any patent infringement lawsuits, and Google will not help you out if you get sued because you make money off of WebM based products and it turns out WebM isn't so 'patent-free' after all. At least the MPEG-LA has a large patent pool that was specifically founded to protect the H264 format and the MPEG-LA members from patent infringement claims.
Then why waste so many words on it? I can appreciate that you're trying to "win", but you're losing. Spectacularly.
Oh really, because the way I'm reading it, node 3 and the other few people who don't seem to blindly ride the Google 'free', Google 'open'-train are the only ones that actually seem to be able to formulate any substantial arguments in this discussion.
Meanwhile, people like yourself, are stuck at hypothesizing how MPEG-LA will rape you, how WebM is somehow less proprietary or closed than H264, how throwing away billions of dollars of vested interest in the most advanced video coding technology is a good thing, how WebM is supposedly 'on par' with H264 in terms of encoding quality (it isn't, read up on the analysis of actual experts like the people behind x264', how WebM is somehow going to 'take over' because it 'is open' and 'will be improved by the open-source community' while it is in fact a spec set in stone and completely controlled by Google, and last but not least, how advanced technology like h264 is apparantly not worth paying for, because ripping it off, crippling it and then giving it away 'for free' to support a strategy of turning everything into an ad-supported illusion of 'freedom' is the way to go.
I tend to have an open mind about software patents, the advantages and disadvantages of both FOSS and closed-source software, and the value of technological innovation, but the way the FOSS crowd is cheering for Google here really makes me sick, it seems like every last bit of rational thinking disappears as soon as someone breaks out the 'free' and 'open' aspect. Filtering out all the arguments in favor of WebM that are bollocks, really only a single argument remains: it's free as in beer (for now). It's free as in beer, and it's crap, so we should prefer it over 'good and paid for' because 'free as in beer' sounds so much like 'free as in speech'.
I never thought I'd ever say something like this, but the way you and your likewise-minded co-posters here are arguing genuinely makes me hope MPEG-LA will soon file and win some patent suits against WebM, just to show how narrowminded and short-sighted the arguments in favor of WebM are, everyone can continue enjoying the most advanced video coding technology in existence, and the MPEG-LA can continue working on making it even better with H265, without having to worry someone like Google will again rip them off and repackage their work as 'open' and 'free' software.
Don't bother trying to come up with reasonable arguments in this discussion. I tried it with the last topic on H264 vs. VP8, and it didn't work. Just reading the majority of the comments in this topic makes me cringe, to the point that I'm starting to wonder why I ever started liking FOSS in the first place, because the discussion on this issue seems to imply that FOSS has become mostly about 'OMG everything has to be free in beer' and how every piece of patented technology is supposedly so trivial it should be unpatentable.
It's really sad how apparently the most rabid FOSS supporters, the ones commenting here, prefer ripping off other people's work, reformatting it and taking features out so they can sell it as 'open' and 'free' just so said companies can keep pretending their mission to turn the world in one big ad-supported 'free as in beer' world of crippled crappy technology. And that this is somehow better for anyone, compared to paying up to a reasonable licensing scheme to use an advanced piece of technology created by others, which cost millions to develop, instead of re-inventing the wheel badly. The amount of FUD about the supposed intentions of the MPEG-LA and the hypothetical things that could happen that would make 'the web' somehow 'closed' or 'proprietary' or would somehow trick us all into getting raped by MPEG-LA would have been laughable if it wasn't so terribly sad.
Which goes to show how Slashdot is slowly becoming irrelevant to people who are genuinely interested in technology and sensible discussion about it. Another tragedy of the commons, the flaming fanboy horde has now taken over this site by democratic force:-(
I agree 100%, what kind of an editor let this kind of inflammatory summary hit the front page? I feel like I'm reading the brain-dead comments you'd expect on Gizmodo or Engadget with this article.
And the worst part about it, is that the analysis this article is linking to isn't even remotely classifiable as FUD, it's not long-winded, and it's actually insightful if you ask me. It's not even a subjective analysis, it's just an analysis of the current state of video codecs, how they relate to each other in terms of openness, how 'open' is not the same as 'royalty free', and how the 'closed' nature of H264 can impossibly be the motive for Google to scratch H264 from Chrome, seeing that it has Flash player (which is proprietary and fully controlled by a single company, only 'open' in the sense that partial specs are availble (DRM system and codec specs are not open), and -that's the best part- includes H264), and in addition to that includes built-in support for other closed/patent-encumbered format such as MP3, AAC or JPEG).
Is Slashdot trying to become the next podium for uninformed hating, flaming, trolling and FUD sponsored by the editors or what?
One more thing to note is the data partitioning scheme used by VP8. This scheme is much like VP3/Theora’s and involves putting each syntax element in its own component of the bitstream. The unfortunate problem with this is that it’s a nightmare for hardware implementations, greatly increasing memory bandwidth requirements. I have already received a complaint from a hardware developer about this specific feature with regard to VP8.
non-adaptive arithmetic coding has to have some serious penalties. It may also be a hardware implementation problem.
I would expect, with equally optimized implementations, VP8 and H.264 to be relatively comparable in terms of decoding speed. This, of course, is not really a plus for VP8: H.264 has a great deal of hardware support, while VP8 largely has to rely on software decoders, so being “just as fast” is in many ways not good enough
The VP8 interpolation filter is likely slightly better, but will definitely be slower to implement, both encoder and decoder-side. A staged filter allows the encoder to precalculate all possible halfpel positions and then quickly calculate qpel positions when necessary: an unstaged filter does not, making subpel motion estimation much slower
This is all only about the parts of VP8 that might be difficult or expensive to implement in hardware in the first place. The rest of the article is riddled with cases where VP8 does things completely different from H264, which means many hardware implementations that implement H264 will not be able to handle them as well as they handle H264. Companies designing SoC's that decode video spend millions on optimizing the hardware down to the last CPU cycle and milli-Watt, often taking shortcuts or accepting limitations on what you can do with them as long as it serves their primary purpose, which is decoding a specific type of video. I've worked on such DSP's and just getting them to run the stuff they are designed for is already very difficult, let alone trying to repurpose them to handle codecs that were never considered in the design phase.
You forgot about the part where you'll need to encode at higher bitrates because WebM is technically inferior to H264, how you will need more power and time to stream or transcode since the only available WebM encoder is slow as molasses and not nearly as good as the better h264 encoders, and how you will have to replace hardware that doesn't support WebM and never will because the chips in it weren't designed for them.
WebM brings nothing to the table that actual users would benefit from. Literally nothing. The only thing it brings is a format Google can use for more tracking and ad serving, while posturing how 'open' and 'free' we are. If WebM prevails, users will get lower quality, less privacy and more ads, but somehow Google tricked them into thinking it benefits them. It's fascinating.
The sheep are people like you who mindlessly repeat the same mantra's about who licensed software stifles innovation and locks out the small players, like you did.
Fact: the most popular, most widely used, and as regarded by many the best H264 encoder/decoder on the market is *free* as in speech, *free* as in beer, *open* as in open-source, and *100% legal* that they negotiated a licensing scheme with MPEG-LA, to allow commercial use of x264, taking royalties from x264 licensees.
Your arguments are defective and irrelevant in the context of H264 codecs.
And yet, MPEG-LA has no problem with anyone using x264, and are in fact taking x264 licensees who want to use it in commercial products. Maybe in your head the world will come to an end and everyone will be tricked into paying shitloads of money to MPEG-LA, but in reality, it's just a codec like so many others with patents on them, and despite that are still used and available everywhere, both as open-source and commercial software.
You know there is an upper bound of $5 million on the H264 royalties, right? How is paying over $100 million for a company with an inefficient, badly documented codec saving Google money?
Google is in this for only 2 things: controlling another standard they can use for tracking and serving ads, and posturing how 'open' and 'free' they are compared to the likes of Microsoft and Apple, so they can add tick off another irrelevant box in the marketing department for their browser and phone OS.
You are both wrong. The most resource-intensive parts of a video codecs are handled by DSP's that are very specific to the codecs they support. While some parts of WebM will translate to current hardware just fine, some parts of the standard have been found not to translate to it at all. Just read this to educate yourself on the subject before assuming hardware WebM support will be a matter of a simple firmware update:
You don't get it. Video decoding hardware is very specific in what parts of what codecs it supports, and it can't be upgraded through software. The x264 devs already determined that WebM contains algorithms that don't translate well to efficient hardware, and that it will be a huge resource hog compared to current h264 solutions, until dedicated WebM hardware is released to the market.
As for the whole licensing discussion: I think everyone should pull their head out of their asses and stop spreading the H264 licensing and royalty FUD. The H264 patent pool serves only a single purpose, which is licensing H264 for use in commercial products and services. The terms are very clear, only if you make more than x amount of money (somewhere in the neighbourhood of a few hundred thousand dollars) you have to pay a very reasonable royalty fee as a compensation for using the work done by the MPEG group and ITU. I don't see what's wrong with that.
The only arguments against H264 that people can come up with are irrational, and hypothetical, and none of them make any sense at all. What if MPEG-LA reverses their decision and asks everyone to pay up for watchin youtube? What if MPEG-LA challenges open-source codecs in court to crush them? What if the lock the specifications and extort everyone hosting an H264 to pay up? None of these make sense unless you think MPEG-LA are codec fascists who are only out to screw everyone, instead of just trying to make money off a very advanced piece of technology that is widely regarded as the best you can get for video coding.
Does the fact that x264 negotiated a licensin scheme with MPEG-LA for 100% legal distribution of x264 for commerical purposes make any sens if they want to extort non-profit use? MPEG-LA is effectively taking x264 licensees now, or in other words: they make money off the commercial use of an open-source codec that's freely available for non-profit use.
The latter, in theory the idea is that 'scharrel-ei' (which would translate best to 'walk-about-egg' I guess) can walk around freely, peck around for food etc. In practice the regulations for calling your eggs 'scharrel-ei' are so lax that the chickens are packed together in a barn so tight that you compare it with taking the subway during rush hours, all day long, have to sleep on the ground, no green or outside spots etc.
The regulations for free-range eggs are a lot tighter here, these chickens can go outside, they can cleep on a dowel, etc.
So because he thinks democracy doesn't always work so well, he's a fascist? Brilliant...
Even the most rabid democratic (and I'm talking about someone who supports democracy, not a member of the democratic party in the US) will agree that democracy is not perfect, and that it doesn't always lead to the 'best' decisions, or the decisions supported by the most people, no matter how you qualify 'good' and 'bad' (be it based on utalitarian, deontological or other ethics). The reason democracy is seen as the 'best' way to run a state is not because it actually 'the best', but because it's the 'the least bad' political system. In history there have been benevolent dictators who did a better job than democracy, but unfortunately it always turns out that power corrupts, and dictatorships turn bad.
I can attest to this. I didn't start buying free-range eggs because I thought they would taste better or would be more healthy, just because I hated the way other eggs are produced, but it turned out that's exactly what I found out. Compared to the eggs I used to buy, which weren't even battery-eggs but what they call 'scharrel-ei' here (no idea what the translation for this would be, but it means the chickens have about 1m^2 of space per chicken, but still live in crap conditions), the 'real' free-range eggs have stronger yellow yoke and more taste.
Anyway, I think it's pretty obvious that meat and dairy produced in an animal-friendly way would be much healthier, if not because of the conditions the animals live in, it would be because of the stuff they get fed. Right now there's a really big scandal in Germany, where it was found out a big producer of animal food has been mixing polluted oils containing a high level of dioxins through their products to make it cheaper, probably for years already. Eggs have been found and tested to have elevated dioxin levels, which is not something you want to eat. Seeing that biologically produced meat and dairy has much higher profit margins, it only seems logical that the farms producing it do not have to shave every last penny off the price of animal food just to get by. So in practice, I think it's very likely cheap meat & dairy is less healthy.
One thing I always wonder is why the *** people come up with this 'prediction' every time, except because they are regurgitating some paranoid internet meme about Apple restricting everything, everywhere, using only closed and Apple-approved stuff for everything, trying to put brain control on you, and trying to lock out applications just for fun, out of pure evilness. It's fascinating, especially if you consider that of all the commercial operating systems from a single vendor, OS X is probably the most 'open' one you can get, except for the UI layer and the Objective-C frameworks in the SDK, about everything in OS X is based on open-source software, and the list of source packages on developer.apple.com/opensource has about 2500 entries.
My question is always this: "What does Apple have to win by locking down OS X the way they locked down iOS?". Even a single good argument would surprise me. The only thing people can come up with is 'make more money by selling applications through the Apps store'. Meanwhile Apple barely breaks even on the iOS app store, while they make billions selling hardware and selling music (DRM free, by the way). Somehow it doesn't really make sense introducing reasons for people not to buy Apple hardware, such as restricting what they can install on it. For a phone, there are good reasons to do so, something Microsoft figured out after WM tanked, and Google will soon find out after Android becomes such a big mess no developer will know what they're developing for. For a desktop OS, everybody would lose by deliberately gimping it.
Keep in mind that Apple is a company that dictated what programming languages developers could use to develop software for one of its platforms. Do you realize how absurd that is? Do you realize how absolutely wrong it is?
No, in fact I don't realize that, maybe you can elaborate? How is this different from about *every freaking other* integrated consumer electronics product on the market? Can I program Java on WP7? Can I program C# on Android? Can I program Java on the Nintendo DS? Can I program Visual Basic on Blackberry?
Here's the thing: Apple created an operating system, a buttload of frameworks, devices that run them, and a set of development tools, the latter of which you can even get for free. All of this was designed and implemented with a number of technologies in mind that fit the hardware and the platform. In terms of programming languages that's Objective-C, C, C++, Fortran. In terms of application and UI frameworks that's Cocoa, UIKit, etc. In terms of development tools (including packaging, provisioning, code signing, and submission to the application store) that's XCode. It's actually all pretty complex, and probably took a lot of time and millions of investment to get everything together. Because Apple provides both the hardware and the retail channel for applications running on the hardware, it is very important for them that applications use the features the platform offers as much as possible, because a crap user experience will hurt the perception of their own products. Which is why they spent a lot of time on the SDK and the development toolchain. Ask any iOS developer and they will tell you that they did in fact do a pretty good job.
Now how absurd and wrong is it that they don't allow every idiot who knows some random programming language to distribute their stuff via the iOS app store? If you want to program Haskell on your iPhone, go ahead, nobody is stopping you, but don't expect Apple to put your work in the app store, just like Microsoft will not allow you to publish a GW-BASIC program on the Xbox 360, or Sony will allow you to distribute a Java application through PSN. Other companies also provide SDKs that you have to use to publish on their platforms, there's nothing absurd or wrong about that. Stop seeing a phone platform as some kind of hobbyist playground that should allow you to do everything with it you desire.
When was the last time you complained you can't reprogram the scaler in your HDTV, or write a Java program for your car ECU?
The real question is how do people manage to charge $0.99 for an iPhone game when they are much closer to the free flash games available on the web or even the free games available on the iPhone.
1. Because not every iOS game is also available as a Flash game, not even remotely so 2. Because Flash games are usually made for kb + mouse control, which you don't have on an iPhone 3. Because the iPhone doesn't support Flash anyway 4. Because even if the iPhone did support Flash, the gaming experience and battery life playing Flash games on it would be terrible. Android has Flash but I don't think many people use it for gaming. 5. Because $0.99 is actually dirt cheap???
You could say the same about buying a hotdog, a carton of milk, or going to a movie theatre. Why pay for that if you can also find someones leftovers in the trash can, you can also drink water from the tap, or you can also watch regular TV at home?
As a spare-time iOS developer I always get a little sad reading stuff like this. We've come to the point where you can pick up nice games for the ridiculously low price of $0.99, games that took hundreds of hours of development, games that are often a lot better in every way and contain a lot more content than games you used to pay $20 or more (remember NeoGeo? $250 for a some games!) for 10 years ago, and yet, people all still complaining. Now it should all be free... People really have become cheap-ass bastards...:-S
When you optimize to GPUs, you have to optimize to all GPUs. I realize there are common instruction sets but the main selling point of Android is its versatility. If I start coding for only Snapdragon processors with PowerVR GPUs because it has a better UX [..]
OpenGL ES 2 == OpenGL ES 2. Just write one compositor that uses it, and one that doesn't. Problem solved. The iPhone 3G and earlier didn't have OpenGL ES 2 either, but that didn't stop Apple from utilizing it for the UX on the 3GS and upwards.
Yet all the low end Android phones combined probably outsell the Droid by a factor of 20 or so. And these phones will not run the same Android apps as well as the Droid, which in itself won't run many applications as well as a nexus s or a galaxy.
My impression is that the reasons why people buy Android phones are overrated. Some people know about Android, and research what phone they will buy. Those people end up with phones like the Droids, the Desires, The Nexuses and the galaxies. The rest just gets their crappy, badly supported Wildfire or Tattoo with their phone plan without having a clue what OS it runs.
Apple doesn't offer a whole lot of choice, that's true, but it's not a problem for them. Somehow a lot of people have the impression marketshare is what Apple is after, but it isn't. Profit margins and customer satisfaction is what Apple tries to maximize, and a single iPhone model a year is what enables them to do exactly that. Many people will pick another phone because of this, but as long as every iPhone generation sells 15 to 20 million units, Apple is on target.
What I meant when I said that the Android model is not sustainable is that another OS will take over eventually. Maybe WP7, maybe meego, maybe even something completely new, but I sincerely think Android will fall apart in the long run by manufacturers abusing it to serve their own agenda, which will hurt platform consistency and image.
Agreed. The good thing about Apple's strategy (good for them, that is, you don't necessarily have to agree that it's good for buyers, even though I personally think it is), is that they target their products specifically at the group of buyers they know will appreciate exactly those attributes of their products they spend the most time on: ease of use, polish (both in terms of software and hardware), longevity (in terms of planned obsolesence). Affordability is not one of these attributes, and people getting iPhones instead of Androids get what they expected from the product, which explains the high customer satisfaction rates.
Meanwhile, Android handset manufacturers mainly target the demographic that wants to save money on their phone, ie: they want it to be cheap, or at least: cheaper than comparable alternatives. Sure enough Android is also great if you are a geek, and sure enough there are also high-end Android phones that are as expensive as the iPhone, but they constitute a pretty minor subset of all Android buyers. The problem with this tactic is that to make money using this strategy, means you have to sell lots of phones, and to do that, you have to introduce lots of new models, to get people to replace their phones faster. You also have to cut down production costs which means making design compromises. Eventually this will hurt Android as a platform and it will hurt customers, because there will be many crappy Android phones on the market, and many phones will end up unsupported within a year. Someone who gets burned by a crappy Android phone will choose something different next time.
I don't think the Google model is sustainable in the long run, and will seriously limit the usefulness of the Android platform. Not because it is a bad platform, but because too many buyers will have a negative experience with their purchase, but also because the insane variety of brands, specifications and OS versions will mean developers will never be able to achieve the same baseline quality level in their apps without having to shut out a very large part of Androids installed base. This will be very confusing and frustrating for end-users who expect to get their phone, go on the Android market, install stuff, only to find out their phone doesn't handle the application, or because the quality is abysmal. Apple got it right with their single-model-1-year-update-cycle, sure, it means you have less choice if you want an Apple phone, but at least you can be pretty sure you won't run into any surprises if you try to use it they way you expect it to work.
This last paragraph is exactly why I find the statement in the article by this guy named Tim Bray pretty stupid. Even if one or two vendors introduced phones that are better than the current iPhone in terms of hardware (such phones are already on the market) *and* software (Android is almost there), you'd still have only a few handset models, which combined will sell only a fraction of what the iPhone sells, and will never get individual marketshare big enough for developers to spend enough time extracting all their capabilities from the hardware and software. Most developers will go for a set baseline much lower than the current iPhone model, just to make sure they target a sufficiently large installed base. That way, the ecosystem of Android apps will always be one or two years behind iOS.
Slashdot Mirror
Agreed. And I do have some evidence to back up why you're right. On my 11 hour flight yesterday the plane was crowded with people using their iPads, and they all seemed extremely satisfied with them. The couple in the isle next to me had one each and have been watching movies on it almost the entire freaking flight, all on a single charge, and surprise, without hearing a single complaint about how the 4:3 aspect ratio was bothering them or whatver. People were playing games on them, having fun, showing off stuff and just enjoying it in general. Somehow I don't have the impression they would have been happier with a Galaxy Tab or some other nerdy Android device with no tablet-optimized stuff whatsoever available for it.
So does or did this whitelist ever contain the default Android browser?
How did they 'irrevocably give up their rights to enforce their patents' any more than MPEG-LA 'giving up their rights to levy royalties for non-profit use of H264'. In terms of hypothetical scenario's for getting screwed as a non-profit customer, Google and MPEG-LA are in the same boat.
As for the indemnification against patent claims, it's pretty obvious that you'd have to be epically stupid to invest heavily in WebM as a company, because Google is specifically saying (in the licensing terms of WebM), that as soon as you get involved in a patent lawsuit and try to settle so you can continue selling your products, the 'free' license you had on WebM will be revoked. Yes, that's true, go look it up: the licensing terms for WebM prohibit settling any patent infringement lawsuits, and Google will not help you out if you get sued because you make money off of WebM based products and it turns out WebM isn't so 'patent-free' after all. At least the MPEG-LA has a large patent pool that was specifically founded to protect the H264 format and the MPEG-LA members from patent infringement claims.
Then why waste so many words on it? I can appreciate that you're trying to "win", but you're losing. Spectacularly.
Oh really, because the way I'm reading it, node 3 and the other few people who don't seem to blindly ride the Google 'free', Google 'open'-train are the only ones that actually seem to be able to formulate any substantial arguments in this discussion.
Meanwhile, people like yourself, are stuck at hypothesizing how MPEG-LA will rape you, how WebM is somehow less proprietary or closed than H264, how throwing away billions of dollars of vested interest in the most advanced video coding technology is a good thing, how WebM is supposedly 'on par' with H264 in terms of encoding quality (it isn't, read up on the analysis of actual experts like the people behind x264', how WebM is somehow going to 'take over' because it 'is open' and 'will be improved by the open-source community' while it is in fact a spec set in stone and completely controlled by Google, and last but not least, how advanced technology like h264 is apparantly not worth paying for, because ripping it off, crippling it and then giving it away 'for free' to support a strategy of turning everything into an ad-supported illusion of 'freedom' is the way to go.
I tend to have an open mind about software patents, the advantages and disadvantages of both FOSS and closed-source software, and the value of technological innovation, but the way the FOSS crowd is cheering for Google here really makes me sick, it seems like every last bit of rational thinking disappears as soon as someone breaks out the 'free' and 'open' aspect. Filtering out all the arguments in favor of WebM that are bollocks, really only a single argument remains: it's free as in beer (for now). It's free as in beer, and it's crap, so we should prefer it over 'good and paid for' because 'free as in beer' sounds so much like 'free as in speech'.
I never thought I'd ever say something like this, but the way you and your likewise-minded co-posters here are arguing genuinely makes me hope MPEG-LA will soon file and win some patent suits against WebM, just to show how narrowminded and short-sighted the arguments in favor of WebM are, everyone can continue enjoying the most advanced video coding technology in existence, and the MPEG-LA can continue working on making it even better with H265, without having to worry someone like Google will again rip them off and repackage their work as 'open' and 'free' software.
Don't bother trying to come up with reasonable arguments in this discussion. I tried it with the last topic on H264 vs. VP8, and it didn't work. Just reading the majority of the comments in this topic makes me cringe, to the point that I'm starting to wonder why I ever started liking FOSS in the first place, because the discussion on this issue seems to imply that FOSS has become mostly about 'OMG everything has to be free in beer' and how every piece of patented technology is supposedly so trivial it should be unpatentable.
It's really sad how apparently the most rabid FOSS supporters, the ones commenting here, prefer ripping off other people's work, reformatting it and taking features out so they can sell it as 'open' and 'free' just so said companies can keep pretending their mission to turn the world in one big ad-supported 'free as in beer' world of crippled crappy technology. And that this is somehow better for anyone, compared to paying up to a reasonable licensing scheme to use an advanced piece of technology created by others, which cost millions to develop, instead of re-inventing the wheel badly. The amount of FUD about the supposed intentions of the MPEG-LA and the hypothetical things that could happen that would make 'the web' somehow 'closed' or 'proprietary' or would somehow trick us all into getting raped by MPEG-LA would have been laughable if it wasn't so terribly sad.
Which goes to show how Slashdot is slowly becoming irrelevant to people who are genuinely interested in technology and sensible discussion about it. Another tragedy of the commons, the flaming fanboy horde has now taken over this site by democratic force :-(
I agree 100%, what kind of an editor let this kind of inflammatory summary hit the front page? I feel like I'm reading the brain-dead comments you'd expect on Gizmodo or Engadget with this article.
And the worst part about it, is that the analysis this article is linking to isn't even remotely classifiable as FUD, it's not long-winded, and it's actually insightful if you ask me. It's not even a subjective analysis, it's just an analysis of the current state of video codecs, how they relate to each other in terms of openness, how 'open' is not the same as 'royalty free', and how the 'closed' nature of H264 can impossibly be the motive for Google to scratch H264 from Chrome, seeing that it has Flash player (which is proprietary and fully controlled by a single company, only 'open' in the sense that partial specs are availble (DRM system and codec specs are not open), and -that's the best part- includes H264), and in addition to that includes built-in support for other closed/patent-encumbered format such as MP3, AAC or JPEG).
Is Slashdot trying to become the next podium for uninformed hating, flaming, trolling and FUD sponsored by the editors or what?
Why not..
One more thing to note is the data partitioning scheme used by VP8. This scheme is much like VP3/Theora’s and involves putting each syntax element in its own component of the bitstream. The unfortunate problem with this is that it’s a nightmare for hardware implementations, greatly increasing memory bandwidth requirements. I have already received a complaint from a hardware developer about this specific feature with regard to VP8.
non-adaptive arithmetic coding has to have some serious penalties. It may also be a hardware implementation problem.
I would expect, with equally optimized implementations, VP8 and H.264 to be relatively comparable in terms of decoding speed. This, of course, is not really a plus for VP8: H.264 has a great deal of hardware support, while VP8 largely has to rely on software decoders, so being “just as fast” is in many ways not good enough
The VP8 interpolation filter is likely slightly better, but will definitely be slower to implement, both encoder and decoder-side. A staged filter allows the encoder to precalculate all possible halfpel positions and then quickly calculate qpel positions when necessary: an unstaged filter does not, making subpel motion estimation much slower
This is all only about the parts of VP8 that might be difficult or expensive to implement in hardware in the first place. The rest of the article is riddled with cases where VP8 does things completely different from H264, which means many hardware implementations that implement H264 will not be able to handle them as well as they handle H264. Companies designing SoC's that decode video spend millions on optimizing the hardware down to the last CPU cycle and milli-Watt, often taking shortcuts or accepting limitations on what you can do with them as long as it serves their primary purpose, which is decoding a specific type of video. I've worked on such DSP's and just getting them to run the stuff they are designed for is already very difficult, let alone trying to repurpose them to handle codecs that were never considered in the design phase.
You forgot about the part where you'll need to encode at higher bitrates because WebM is technically inferior to H264, how you will need more power and time to stream or transcode since the only available WebM encoder is slow as molasses and not nearly as good as the better h264 encoders, and how you will have to replace hardware that doesn't support WebM and never will because the chips in it weren't designed for them.
WebM brings nothing to the table that actual users would benefit from. Literally nothing. The only thing it brings is a format Google can use for more tracking and ad serving, while posturing how 'open' and 'free' we are. If WebM prevails, users will get lower quality, less privacy and more ads, but somehow Google tricked them into thinking it benefits them. It's fascinating.
The sheep are people like you who mindlessly repeat the same mantra's about who licensed software stifles innovation and locks out the small players, like you did.
Fact: the most popular, most widely used, and as regarded by many the best H264 encoder/decoder on the market is *free* as in speech, *free* as in beer, *open* as in open-source, and *100% legal* that they negotiated a licensing scheme with MPEG-LA, to allow commercial use of x264, taking royalties from x264 licensees.
Your arguments are defective and irrelevant in the context of H264 codecs.
And yet, MPEG-LA has no problem with anyone using x264, and are in fact taking x264 licensees who want to use it in commercial products. Maybe in your head the world will come to an end and everyone will be tricked into paying shitloads of money to MPEG-LA, but in reality, it's just a codec like so many others with patents on them, and despite that are still used and available everywhere, both as open-source and commercial software.
You know there is an upper bound of $5 million on the H264 royalties, right? How is paying over $100 million for a company with an inefficient, badly documented codec saving Google money?
Google is in this for only 2 things: controlling another standard they can use for tracking and serving ads, and posturing how 'open' and 'free' they are compared to the likes of Microsoft and Apple, so they can add tick off another irrelevant box in the marketing department for their browser and phone OS.
You are both wrong. The most resource-intensive parts of a video codecs are handled by DSP's that are very specific to the codecs they support. While some parts of WebM will translate to current hardware just fine, some parts of the standard have been found not to translate to it at all. Just read this to educate yourself on the subject before assuming hardware WebM support will be a matter of a simple firmware update:
http://x264dev.multimedia.cx/archives/377
You don't get it. Video decoding hardware is very specific in what parts of what codecs it supports, and it can't be upgraded through software. The x264 devs already determined that WebM contains algorithms that don't translate well to efficient hardware, and that it will be a huge resource hog compared to current h264 solutions, until dedicated WebM hardware is released to the market.
As for the whole licensing discussion: I think everyone should pull their head out of their asses and stop spreading the H264 licensing and royalty FUD. The H264 patent pool serves only a single purpose, which is licensing H264 for use in commercial products and services. The terms are very clear, only if you make more than x amount of money (somewhere in the neighbourhood of a few hundred thousand dollars) you have to pay a very reasonable royalty fee as a compensation for using the work done by the MPEG group and ITU. I don't see what's wrong with that.
The only arguments against H264 that people can come up with are irrational, and hypothetical, and none of them make any sense at all. What if MPEG-LA reverses their decision and asks everyone to pay up for watchin youtube? What if MPEG-LA challenges open-source codecs in court to crush them? What if the lock the specifications and extort everyone hosting an H264 to pay up? None of these make sense unless you think MPEG-LA are codec fascists who are only out to screw everyone, instead of just trying to make money off a very advanced piece of technology that is widely regarded as the best you can get for video coding.
Does the fact that x264 negotiated a licensin scheme with MPEG-LA for 100% legal distribution of x264 for commerical purposes make any sens if they want to extort non-profit use? MPEG-LA is effectively taking x264 licensees now, or in other words: they make money off the commercial use of an open-source codec that's freely available for non-profit use.
The latter, in theory the idea is that 'scharrel-ei' (which would translate best to 'walk-about-egg' I guess) can walk around freely, peck around for food etc. In practice the regulations for calling your eggs 'scharrel-ei' are so lax that the chickens are packed together in a barn so tight that you compare it with taking the subway during rush hours, all day long, have to sleep on the ground, no green or outside spots etc.
The regulations for free-range eggs are a lot tighter here, these chickens can go outside, they can cleep on a dowel, etc.
So because he thinks democracy doesn't always work so well, he's a fascist? Brilliant...
Even the most rabid democratic (and I'm talking about someone who supports democracy, not a member of the democratic party in the US) will agree that democracy is not perfect, and that it doesn't always lead to the 'best' decisions, or the decisions supported by the most people, no matter how you qualify 'good' and 'bad' (be it based on utalitarian, deontological or other ethics). The reason democracy is seen as the 'best' way to run a state is not because it actually 'the best', but because it's the 'the least bad' political system. In history there have been benevolent dictators who did a better job than democracy, but unfortunately it always turns out that power corrupts, and dictatorships turn bad.
I can attest to this. I didn't start buying free-range eggs because I thought they would taste better or would be more healthy, just because I hated the way other eggs are produced, but it turned out that's exactly what I found out. Compared to the eggs I used to buy, which weren't even battery-eggs but what they call 'scharrel-ei' here (no idea what the translation for this would be, but it means the chickens have about 1m^2 of space per chicken, but still live in crap conditions), the 'real' free-range eggs have stronger yellow yoke and more taste.
Anyway, I think it's pretty obvious that meat and dairy produced in an animal-friendly way would be much healthier, if not because of the conditions the animals live in, it would be because of the stuff they get fed. Right now there's a really big scandal in Germany, where it was found out a big producer of animal food has been mixing polluted oils containing a high level of dioxins through their products to make it cheaper, probably for years already. Eggs have been found and tested to have elevated dioxin levels, which is not something you want to eat. Seeing that biologically produced meat and dairy has much higher profit margins, it only seems logical that the farms producing it do not have to shave every last penny off the price of animal food just to get by. So in practice, I think it's very likely cheap meat & dairy is less healthy.
Exactly...
One thing I always wonder is why the *** people come up with this 'prediction' every time, except because they are regurgitating some paranoid internet meme about Apple restricting everything, everywhere, using only closed and Apple-approved stuff for everything, trying to put brain control on you, and trying to lock out applications just for fun, out of pure evilness. It's fascinating, especially if you consider that of all the commercial operating systems from a single vendor, OS X is probably the most 'open' one you can get, except for the UI layer and the Objective-C frameworks in the SDK, about everything in OS X is based on open-source software, and the list of source packages on developer.apple.com/opensource has about 2500 entries.
My question is always this: "What does Apple have to win by locking down OS X the way they locked down iOS?". Even a single good argument would surprise me. The only thing people can come up with is 'make more money by selling applications through the Apps store'. Meanwhile Apple barely breaks even on the iOS app store, while they make billions selling hardware and selling music (DRM free, by the way). Somehow it doesn't really make sense introducing reasons for people not to buy Apple hardware, such as restricting what they can install on it. For a phone, there are good reasons to do so, something Microsoft figured out after WM tanked, and Google will soon find out after Android becomes such a big mess no developer will know what they're developing for. For a desktop OS, everybody would lose by deliberately gimping it.
Keep in mind that Apple is a company that dictated what programming languages developers could use to develop software for one of its platforms. Do you realize how absurd that is? Do you realize how absolutely wrong it is?
No, in fact I don't realize that, maybe you can elaborate? How is this different from about *every freaking other* integrated consumer electronics product on the market? Can I program Java on WP7? Can I program C# on Android? Can I program Java on the Nintendo DS? Can I program Visual Basic on Blackberry?
Here's the thing: Apple created an operating system, a buttload of frameworks, devices that run them, and a set of development tools, the latter of which you can even get for free. All of this was designed and implemented with a number of technologies in mind that fit the hardware and the platform. In terms of programming languages that's Objective-C, C, C++, Fortran. In terms of application and UI frameworks that's Cocoa, UIKit, etc. In terms of development tools (including packaging, provisioning, code signing, and submission to the application store) that's XCode. It's actually all pretty complex, and probably took a lot of time and millions of investment to get everything together. Because Apple provides both the hardware and the retail channel for applications running on the hardware, it is very important for them that applications use the features the platform offers as much as possible, because a crap user experience will hurt the perception of their own products. Which is why they spent a lot of time on the SDK and the development toolchain. Ask any iOS developer and they will tell you that they did in fact do a pretty good job.
Now how absurd and wrong is it that they don't allow every idiot who knows some random programming language to distribute their stuff via the iOS app store? If you want to program Haskell on your iPhone, go ahead, nobody is stopping you, but don't expect Apple to put your work in the app store, just like Microsoft will not allow you to publish a GW-BASIC program on the Xbox 360, or Sony will allow you to distribute a Java application through PSN. Other companies also provide SDKs that you have to use to publish on their platforms, there's nothing absurd or wrong about that. Stop seeing a phone platform as some kind of hobbyist playground that should allow you to do everything with it you desire.
When was the last time you complained you can't reprogram the scaler in your HDTV, or write a Java program for your car ECU?
The real question is how do people manage to charge $0.99 for an iPhone game when they are much closer to the free flash games available on the web or even the free games available on the iPhone.
1. Because not every iOS game is also available as a Flash game, not even remotely so
2. Because Flash games are usually made for kb + mouse control, which you don't have on an iPhone
3. Because the iPhone doesn't support Flash anyway
4. Because even if the iPhone did support Flash, the gaming experience and battery life playing Flash games on it would be terrible. Android has Flash but I don't think many people use it for gaming.
5. Because $0.99 is actually dirt cheap???
You could say the same about buying a hotdog, a carton of milk, or going to a movie theatre. Why pay for that if you can also find someones leftovers in the trash can, you can also drink water from the tap, or you can also watch regular TV at home?
As a spare-time iOS developer I always get a little sad reading stuff like this. We've come to the point where you can pick up nice games for the ridiculously low price of $0.99, games that took hundreds of hours of development, games that are often a lot better in every way and contain a lot more content than games you used to pay $20 or more (remember NeoGeo? $250 for a some games!) for 10 years ago, and yet, people all still complaining. Now it should all be free... People really have become cheap-ass bastards... :-S
When you optimize to GPUs, you have to optimize to all GPUs. I realize there are common instruction sets but the main selling point of Android is its versatility. If I start coding for only Snapdragon processors with PowerVR GPUs because it has a better UX [..]
OpenGL ES 2 == OpenGL ES 2. Just write one compositor that uses it, and one that doesn't. Problem solved. The iPhone 3G and earlier didn't have OpenGL ES 2 either, but that didn't stop Apple from utilizing it for the UX on the 3GS and upwards.
Sounds like a bad gay porn advertisement to me.
What is gay about squirting?
Yet all the low end Android phones combined probably outsell the Droid by a factor of 20 or so. And these phones will not run the same Android apps as well as the Droid, which in itself won't run many applications as well as a nexus s or a galaxy.
My impression is that the reasons why people buy Android phones are overrated. Some people know about Android, and research what phone they will buy. Those people end up with phones like the Droids, the Desires, The Nexuses and the galaxies. The rest just gets their crappy, badly supported Wildfire or Tattoo with their phone plan without having a clue what OS it runs.
Apple doesn't offer a whole lot of choice, that's true, but it's not a problem for them. Somehow a lot of people have the impression marketshare is what Apple is after, but it isn't. Profit margins and customer satisfaction is what Apple tries to maximize, and a single iPhone model a year is what enables them to do exactly that. Many people will pick another phone because of this, but as long as every iPhone generation sells 15 to 20 million units, Apple is on target.
What I meant when I said that the Android model is not sustainable is that another OS will take over eventually. Maybe WP7, maybe meego, maybe even something completely new, but I sincerely think Android will fall apart in the long run by manufacturers abusing it to serve their own agenda, which will hurt platform consistency and image.
Agreed. The good thing about Apple's strategy (good for them, that is, you don't necessarily have to agree that it's good for buyers, even though I personally think it is), is that they target their products specifically at the group of buyers they know will appreciate exactly those attributes of their products they spend the most time on: ease of use, polish (both in terms of software and hardware), longevity (in terms of planned obsolesence). Affordability is not one of these attributes, and people getting iPhones instead of Androids get what they expected from the product, which explains the high customer satisfaction rates.
Meanwhile, Android handset manufacturers mainly target the demographic that wants to save money on their phone, ie: they want it to be cheap, or at least: cheaper than comparable alternatives. Sure enough Android is also great if you are a geek, and sure enough there are also high-end Android phones that are as expensive as the iPhone, but they constitute a pretty minor subset of all Android buyers. The problem with this tactic is that to make money using this strategy, means you have to sell lots of phones, and to do that, you have to introduce lots of new models, to get people to replace their phones faster. You also have to cut down production costs which means making design compromises. Eventually this will hurt Android as a platform and it will hurt customers, because there will be many crappy Android phones on the market, and many phones will end up unsupported within a year. Someone who gets burned by a crappy Android phone will choose something different next time.
I don't think the Google model is sustainable in the long run, and will seriously limit the usefulness of the Android platform. Not because it is a bad platform, but because too many buyers will have a negative experience with their purchase, but also because the insane variety of brands, specifications and OS versions will mean developers will never be able to achieve the same baseline quality level in their apps without having to shut out a very large part of Androids installed base. This will be very confusing and frustrating for end-users who expect to get their phone, go on the Android market, install stuff, only to find out their phone doesn't handle the application, or because the quality is abysmal. Apple got it right with their single-model-1-year-update-cycle, sure, it means you have less choice if you want an Apple phone, but at least you can be pretty sure you won't run into any surprises if you try to use it they way you expect it to work.
This last paragraph is exactly why I find the statement in the article by this guy named Tim Bray pretty stupid. Even if one or two vendors introduced phones that are better than the current iPhone in terms of hardware (such phones are already on the market) *and* software (Android is almost there), you'd still have only a few handset models, which combined will sell only a fraction of what the iPhone sells, and will never get individual marketshare big enough for developers to spend enough time extracting all their capabilities from the hardware and software. Most developers will go for a set baseline much lower than the current iPhone model, just to make sure they target a sufficiently large installed base. That way, the ecosystem of Android apps will always be one or two years behind iOS.