|Couple quick thoughts on the iPad:
It's pretty much what I expected. I thought there might be a camera (front facing, for video chat,) but I'm not so surprised there isn't. I didn't expect there to be a keyboard, so I'm a little surprised at that external option.
The price is a little lower than I expected. Apple is clearly gunning to dominate this space in a way they never tried with personal computers.
People are complaining that it's just a big iPhone, but I think John Gruber has it right that the truth is the other way around: the iPhone is a small iPad. This is the product Apple has been trying to make for many years, but it wasn't possible until now. A few years ago they took a lot of the ideas from the then still incubating tablet and made what they could at the time: the iPhone. But this is what they wanted to make, and they just had to wait for the reality of what is possible to catch up with their ideas.
As I predicted, lots of people seem to be enraged by this device (and all the hype around it). They point out, with some accuracy, that there isn't anything new here. Or worse, that there is even less here than lots of other devices that have been on the market for far longer. But this misses the point. The iPod was arguably worse (certainly had less features) than the Creative Jukebox when the iPod debuted. And there was this same sort of "Apple is doomed" talk at the time. How'd that work out for Creative?
The Apple branded CPU ("Apple A4") is interesting for how little Apple will say about it. This isn't to keep things secret, but just because this isn't a product for people who care about the hardware details. But for people who do care, like me, it looks like it's basically an ARM Cortex A9 that Apple tweaked and mated with a GPU (again, pretty much what we thought.) I'm hoping more details come up, but I doubt they'll be from Apple.
And finally, yes, I still think it will be a winner. But remember, it's not the tablet (I think iPad is a bad name, but I absolutely don't think it will matter) that is key. It's the whole hardware / software ecosystem they are building. Nobody else is even thinking on this scale. And that's why it's interesting.
I guess there is some interest in the question of openness. Clearly this isn't an open device. That's the point, Apple wants to completely control the experience because they think they know better than "most people". And while I don't personally like that aspect of it (for instance, I don't think I'll be buying one) that doesn't really have much to do with anything. When it comes to computers most people are not like me. But still, in the abstract, there's nothing ethically wrong with closed computing systems. Nobody complains that they can't load their own software onto a remote control; or that they can't reprogram their refrigerators. If Apple was somehow pushing to make general purpose computers illegal - then there would be a problem. But they're not, and they won't be. Personal computers will continue to exist, with a variety of open and closed architectures and operating systems. The iPad is something different. If someone doesn't like the closed nature, then they shouldn't buy it. But I think a ton of people are going to.
What will the entry level price for the iPad be by Christmas? $399 seems like a not too extravagant guess. $349 maybe? $299 possibly? Probably that's too low too soon, but you get the idea. Can anyone catch up? Does anyone else control the whole stack (from CPU through the OS and all applications? Does anyone else have the volume (iPad + iPod + iPhone - all sharing a ton of components) to drive prices down? I don't think so.
Microsoft didn't have to seek a ban on Netscape--it just took over the show, and we are all the worse for it. You're saying Apple's thinking ahead, but won't contemplate that they're thinking ahead to total integrated market domination (iEverything). Look how the brand-specific term podcast became synonymous with "spoken word mp3 on blog." Juan Cole on his blog today said "log into iTunes to hear my podcast," not "here's a link to the mp3." I can't unconvince you but will keep trying--Apple has turned to the Dark Side, and you can still turn away from it, Luke.
i think of podcasts as deep internet history. (just sound!) JC clings to it since it was his medium and he advances slowly. i dont think internet media consumers cling to the term at all w/ you tube as the replacement medium. they moved along as media advanced with broadening bandwidth. do people still listen to podcasts? is that still an ipod habit? is that how people use ipods still or a too labor intensive app/fad? do people use ipods anymore or prefer iphones and other advanced cell phones?
I think platform wars are so 20th century. The ideal client is thin and open
Thanks Tom. I share your concerns. But things have so far worked out. Maybe I'm being naive. But even though IE beat Netscape, it only resulted in IE taking over temporarily. It's still dominant, but on the decline against multiple strong competitors. 'Podcasts' became a word, but I'm not really seeing the danger. Apple doesn't make money from them; in fact, quite the opposite as they host things like Juan Cole's talks that would otherwise cost him money to distribute. I think Apple is executing their strategy. If they are evil then they've always been evil, rather than having suddenly turned. Total market domination is one way to look at their desires. They would say something more like "changing the world". I just look on with some awe at their ability to execute. Computers are still a little geeky. A little out of reach for a lot of people. They are changing that. In the process they are building some walled garden situations, but not in any way that will harm people who choose to stay outside. No one is taking over the show. They are just expanding the tent to include a lot of people who had previously been left out.
"The ideal client is thin and open...." Yup. In other words the browser. In other words WebKit.
And ree, yes, I agree. "Podcast" isn't even a big term anymore. And iPods do still sell, but they have merged with the iPhone (the iPod Touch) and the boundary gets more and more fuzzy. Advanced cell phones are the personal computer evolved to the point where we begin to stop thinking of them as computers and they become something more intimate. Apple's been thinking about this for a while. Of course, they're not the only ones, but they do seem to be executing better than anyone else.
"Or other advanced cell phones" sounds OK but since Apple has all these integrated features to hook in casual users they will likely be the winner. As for YouTube being an alternative/successor to iPods (if it is), Apple and Google are all over that one with shifting the 'Tube to a new, Apple-friendly (and others not so friendly) video standard, as we've discussed elsewhere. When my workplace goes all-Apple I'll say I told you so but until then I'm hoping simple entropy (and world poverty) will keep Jobs' control schemes at bay.
I just don't see the danger. Why is somewhat proprietary h.264 worse than the current somewhat proprietary flash format? Video has to be served in some format. Why will Apple's success be worse than the current Microsoft hegemony? (Or in video delivery the current Adobe stranglehold?) You'd just rather have no standards and lots of incompatible offerings just to see that nobody "wins"? Why is expanding the marketplace (i.e., "hooking in casual users") a bad thing? You'd rather that more people just didn't care about computers? Or that computers stay too complicated for some part of the market to use?
I can see that there is a danger with single corporate entities controlling too much. But the opposite just means that nothing gets done and we'd all be using Visicalc and watching grainy Real Video clips on our Windows 3.1 computers. I'm not concerned with a corporation using a dominant market position to push technology forward so much as with technology just stagnating - whether due to a dominant corporation that doesn't know how to innovate or a chaotic market place that doesn't standardize enough to allow creativity to flow frictionlessly to a mass audience.
But I think I do agree with you in the sense that Apple is likely to be the "winner" in some sense due to the "integrated features" they are building into their products, and that does make them less interesting to watch in some sense. It will be fun to spot the next upstart on the horizon who will eventually disrupt Apple. But in the meantime I think Apple is ruling (or will rule) fairly benevolently.
The licensing for h.264/AVC is complicated, however monetary transfer for the use of technology ideas is the basis for how I make a living. I'm pretty comfortable with the general model and AVC in particular. There's competition, so if the licensors have an incentive to not do anything stupid (as in the original MPEG-4 licensing).
The thing about video encoding/decoding is that it's hard. Having a standard allows people to make codec chips, GPUs, etc. that can support the requisite video processing efficiently and at low cost.
Long term, there is NFW for h.264, or h.265, or whatever to have hegemony. Like all else in technology, these things are transient. General purpose HW will take over the functions, and the need for a single ubiquitous standard goes away. Off the cuff prediction: 5 years. That's a long time in internet years, but in real life that's not so much.
To use an analogy from a business I'm immersed in, Apple is trying (with the iPhone, iPod, iPad, Apple TV, iTunes, iTunes Store, App Store) to be Comcast and Vizio simultaneously. By being a content intermediary, aggregator, store front, they're competing with Comcast et al. By making a client, they're competing with Vizio et al.
The television set is the classic thin client. I don't need a DuMont to watch the DuMont Network, a Sony TV to see Sony Pictures content, etc. By being so thin, it's also become somewhat of a commodity.
The analogy breaks down in that there are actually two layers of client: the set top box, and the TV. The STB is proprietary and closed, while the TV is open and a commodity. I argue that the STB as we know it is a living fossil. It's taken 50 years for the cable/satellite TV market to get where it is today, so it's not going to drop dead next week. But it is in the process of dying. The client of the future *has* to be different. The closed proprietary HW clients will be left in the dust. Another layer of analogy ... imagine buying a video conferencing system with piles and piles of proprietary hardware and proprietary protocols. AND, you could only talk to people who had the same gear. This is how people video conferenced in 1990. By '91 at the latest, it was crystal clear that open standards implemented largely (but not completely) on generic technology platforms was going to win.
Perhaps I'm overemphasizing an obvious point, because I'm around people all day who don't understand the living fossil aspect. Drives me nuts.
Ultimately the storefront, the access network, the client HW and the client SW will be deconstructed. In cable, all four of those are bundled. In iPad, three of the four are bundled (three and a half if you count the Apple/AT&T partnership for the 3g). Apple can make a tidy little business from bundling now. But they ought to understand that it's a temporary situation. If they appear to be overreaching and forcing too much conformance to their environment, there will be a backlash. If they make the stuff all work together in an awesome fashion (but don't make it mandatory to use it all together) then people will naturally be drawn to staying within the Apple environment. That's got to be their long-term play: don't make it hard to stray, but do make it awesome to stay.
Ultimately, I'll be able to buy whatever media reader gizmo I want, select what client SW I want to load, pick with media I want, and pick which media store I want to buy it from. Apple's iPad has to evolve to that paradigm or it will eventually and inevitably die.
An aside: A friend mentioned OSF types protesting at the announcement of the iPad. His reaction "That's not the right response. Go make something better. That's how you respond."
Yeah, the video compression issue is really complex. And it's probably the thing I worry about the most. (Or, the thing I wish we could say for sure was not going to be a problem down the road.) One thing I didn't understand before starting to read is Mark's point:
The thing about video encoding/decoding is that it's hard. Having a standard allows people to make codec chips, GPUs, etc. that can support the requisite video processing efficiently and at low cost.This is something that non-hardware engineers like myself can maybe miss. Given mobile devices with thermal and size constraints on the CPU, you're not going to be able to decode high quality video in software (on the CPU,) you're going to need a dedicated decoder chip (which may or may not be part of the GPU.) And, to get to the point, nobody is going to design and fab computer chips using technology (a specific video codec) that has an unclear patent situation. It's really expensive to set up chip fabrication. So if someone made a hardware Ogg Theora decoder, and then a third party came forward with patent claims it would be a mess. And there are companies that hold patents and just wait until someone is invested enough to strike (i.e., no one is going to come forward now with patents on Ogg because there is no one with deep pockets to sue; but if intel started cranking out tens of millions of mobile GPUs with built in Ogg decoding support, then they would suddenly surface with their claims - hence "submarine patent.")
So it becomes very difficult to build hardware around open formats since there is no one with any cash to defend the format from submarine patents. It's better for a lot of big businesses just to pay MPEG-LA for the h.264/AVC and then rely on them to defend the intellectual property.
But, with all that said, Google might still rescue us all. They bought On2 which controlled the rights to the vp6 codec which is what the Flash player uses. So they could open that up, or at the very least they could use that as a starting point for a next gen codec if h.264/AVC terms turn ugly. So it's not like the internet community is going to get totally stuck on this. But sure, I wish it looked better at this point.
Jim, since I learned so much from you about open standards and the importance of general purpose computers, it's hard to read you defending Apple this way.
I was reading on Open Left that the Congressional Black Caucus opposes net neutrality because they believe the telecoms' arguments "Let us become more profitable and we will build out broadband to your communities." (So we can then control your cable experience.)
Big companies (especially "benign" ones like Apple--though I don't believe they are) can educate people and even make money being somewhat open. Windows was smart in one way, which was plug and play and 3rd party applications. The argument "the iPad isn't intended for sophisticated users" is an abdication of social responsibility. What Jobs wants is a "convergence TV" with book reader, music player, etc--same as Comcast wants, so he can offer you 1000 channels of company-vetted crap. Right now you have your computer so you're OK but Jobs would love to take it away from you and give a set top box. This is not good and I wish you'd stop saying it is, just because you're pulling back into your market-watcher, "I'm not involved" posture. You are involved--you are enabling a closed system.
Apple is the company best positioned to establish a monopoly in a converged portable media player and associated server-side products. Whether or not they want to, I don't think they can do it. I don't think anyone can.
I think we're moving towards a post-platform-war environment. E.g., for most stuff it really doesn't matter if I have a Mac or a PC, especially media I want to consume. (Yeah, there's extra pain in getting WMV content on a Mac, but it's certainly doable.)
In a post-platform-war environment, it's exceedingly difficult to build a vertically integrated monopoly, esp. with the number of players in the game. People will be able to pick and choose between competitors for the different layers of the stack. Both content owners and content consumers will undermine anyone's attempt to establish a monopoly in the intermediate layers. E.g., if the iTunes store sucks, I can just download directly from the content owners' web sites. There is no inherent need for a retailer.
If Apple gets too rigid about their particular stack of store front, media player software, media player hardware, they will lose. Best of breed solutions cobbled together from disparate suppliers will work, and will be evolving in performance and functionality very quickly. At least that's the outcome I see. If I'm right, then if Apple plays nice, everyone benefits -- but if they don't, they lose.
As a point of contrast, here are some places I see where we're not yet in a post-platorm-war environment:
- Optical disks. BD won HD-DVD lost. Due to the molecules involved (manufacturing plants, inventory, etc.), one had to lose. We may never have another similar format war either because physical media distribution becomes irrelevant or everyone recognizes it's a stupid fight.
- Game platforms. These are so compute intensive, that like video compression, special purpose hardware is required. Thus we have PS3, Wii, Xbox 360, and PCs. With video compression, over the years, the bit rate decreases exponentially -- meaning the decreases in bit rate are smaller and smaller (and harder and harder). Eventually (I'm predicting) radical new compression standards will be hard to justify, and implementations will be able to get by with generic hardware. The rendering in video games is a different matter. The complexity is unbounded, so HW acceleration will be with us for a while.
Digital media playback doesn't fit within the constraints I cited (molecules or unbounded complexity). There's no war to be won.
"...but Jobs would love to take it away from you and give a set top box."
I just don't think that is true. I think killing the Macintosh (his baby) is something he would never do under any circumstances. Do you have any evidence, or just a belief that Apple must be evil? And just from a purely capitalistic greedy point of view, do you know what the margins are on a set top box? Compared to a Mac? It would kill Apple to go this way.
If I did think he wanted to do that (get rid of general purpose computers) then I would agree with you. So I don't see how I am enabling a closed system. I'm merely writing about where I think the market is going. I think Apple is going to control the market for digital media consumption in the near term. I've already stated I'm not going to buy one. But I am impressed by what they are doing. I'm impressed by the technology as well as the scale of their vision. Sort of like I'm impressed by the technology behind an Abrams M1A1 tank, but that doesn't make me a war monger, or make me a supporter of wars. Or the way I'm impressed with LeBron James, but that doesn't make me a Cavaliers fan.
It's true that in the past I did write a lot, somewhat more passionately than I do now, about the importance of general purpose computers. That was back when I thought there might be some way for the government to actually outlaw them. I just don't think so anymore, so my ideas have evolved and my ability to get politically worked up ("To the barricades!") has gone down.
Now I'm closer to how Mark's friend expressed it when I see something in the tech world I don't like: "Go make something better." If you're talking about pipes (like Comcast or the cellular networks or any of the other telecoms) then obviously you can't just go and make something better. But Apple doesn't have any choke points like that. They have the App store and the iTunes store. There is a ton of competition for those things, and there is no way Apple can get into a position to kill off other competitors. So it will always be possible to route around Apple, provided someone can make something better. So far no one has made anything better from my point of view. I think my "market-watcher" position is ethically valid.
I think Mark's last point (didn't see it before I posted) is correct, and that's exactly why I say I think Apple will dominate in the "near term". Eventually, and sooner if they get too greedy, someone will do something better, and then everyone will switch to that.
In the meantime, I think the iPad might prove to be an excellent old-lady computer. Might get one for my mom.
Mark, I hope you're right.
Jim, I'm basing it on the iPod originally being more closed than it is and only opening somewhat after people howled, and now the iPad, for all the reasons outlined by Create Digital Music. These are Apple's bleeding edge products and this is where they show their true colors.
I think your early concerns about the computer were valid. It's not a matter of outlawing it so much as concentrating power--as we learned under Bush, if the government says "mind if we browse through your customer info?" your large US company will say, "Go right ahead, we'll build a room for you if it will help."
Earlier you asked if I was advocating chaos. You're turning my argument around. I'm arguing for agreed open standards--it's Apple that won't play, as I recall.
As for the threat of submarine patents--I think it's bogus. For a large company nuisance lawsuits are part of doing business. It really bugs me that Mozilla is out in the cold and the for-profit gets to write the rules.
I think you are under estimating the billions of dollars it takes to fab microchips. This is more than a nuisance. There simply won't ever be Ogg Theora decoding hardware for this reason. I'd like to see more open video standards too, but going firefox's way means no high quality video on small devices. Is that really what you want?
As for chaos, I say it because companis making hardware are not going to sign on for your open video standards for the reasons above. So without h.264 we'd be back to each vendor having their own video codec. It's a mess.
Also, fwiw, mobile firefox just dropped flash support for the same reason apple did - too CPU intensive.
But I'm listening too your arguments and I appreciate your point of view.
I'm saying if Apple is the leader they pay the cost of the buildout, reap the vast profits of their total-integration-media system, fight the lawsuits against spurious, late-arriving patent claims and then third parties using the standard don't have to worry. Apple basically said f.u. to the people whose job it is to come up with open standards and is using "the threat to the little guy" as an excuse.
I think your missing my point. Apple doesn't build hardware video decoders. Only the big guys can fab those. Apple buys from one of them (say Intel) and integrates them into their device. But apple can't buy an ogg video decoder chip because no one makes one. I don't see how it's apples fault that no one makes one. If an open standards group would fab the chip, at competitive cost, I'm sure apple might well use it.
Well see what the new mobile firefox does. If I'm right they won't be able to support ogg either for these same reasons. It's not a conspiracy.
But who will Intel sell them to? If Apple is going to be their biggest customer you make the contract contingent on joint defense of goofy patent claims from low-lying schmucks. Or a consortium of the largest customers does this. You don't spit on the open standards guys--this is the point I don't seem to be making any headway with. That's what Microsoft did/does with the Internet. (And lots of people are saying Apple is getting worse than Microsoft--it's not just me tilting at windmills on the Tree.)
Here is an interesting post about Amazon disabling the "buy" button on all McMillan books. That's the kind of thing Apple's total domination of all media will allow them to do on a much bigger scale. They are already showing their bad faith by censoring apps from people whose politics they don't like.
"If Apple is going to be their biggest customer you make the contract contingent on joint defense of goofy patent claims from low-lying schmucks..." That is exactly what they have done. It's called MPEG-LA. They license the rights to h.264. Anyone can join, as Apple has, and then use the IP while MPEG-LA will use the money to fight patent claims.
I think you are confusing open and free. H.264 is open. Anyone can produce an implementation. It's just not free. You have to license the rights from MPEG-LA. But anyone can do so. Neither Apple, nor anyone else, can say to a third party, "you can't have a license." That's open. But it's not free. (And personally I think the fees are too high and there aren't enough protections against the fees going up in the future - that's why I agree and say I'm worried about the situation.)
But you seem to be suggesting that Apple should go on the hook to defend against possible billions of dollars in legal damage and then give away that protection for free? To open source projects and competitors alike? Why would they do that?
I think your claim that Apple is spitting on open source advocates shows an unfamiliarity with Apple's involvement with open source. I'm sure Richard Stallman is opposed to what Apple has done, and I'm sure you can find any number of other super hard liners who are similarly opposed. But they have built a huge chunk of their technology on open source, and they have given back a significant amount of work to these open source projects, BSD and KHTML being two of the largest.
I'm trying to hear what you are saying. And if I come across as saying that there is no chance that Apple could do something evil then I have misspoken. I think there is a chance for anyone to suddenly do something evil. But I just don't agree that Apple could choose to go with Ogg Theora for video playback and make it work if they really wanted to. I think this is untrue, and I think this is part of what you are claiming. So I just disagree on that point.
But I like talking it over with you, and there are certainly other points on which I could be convinced that Apple poses a threat to the open internet. But their insistence on using H.264 for video encoding isn't evil, and isn't done out of any attempt to control the internet. (They would have made their own encoding technology in order to do that, and assumed the risks themselves like you advocate, but then *not* made it open.)
I'm reading the articles and I'm not confused. I'm sure we could think of examples where a company has "given away the source code" as a way of attracting business to its products.
I gave some examples of where I think Apple has acted in bad faith. I take that as evidence of what kind of corporate citizen they will be as they consolidate media and power and market share. I'm sorry you can't "hear" this.
As for Ogg, yes, we disagree. We're not talking about imaginary patent suits breaking Apple here but rather that they went with the cheaper route of paying the consortium while at the same time throwing the "democrats" under the bus. Without studying their balance sheets I think they could have supported an open standard for the media, and made money off their hardware and media-files-for-grandma database schemes. But they want more, more, more.
Paddy Johnson made a comment somewhere about the iPad not having Flash and therefore being bad for porn consumption. Maybe that will be Jobs' Achilles heel.
I looked back over the whole thread and I can't find any concrete examples you've given of Apple acting in bad faith. I thought there was something you wrote about the App store blocking apps because of the authors political beliefs but now I can't find that. Do you really have any examples of Apple acting in bad faith?
In any case, I don't feel like this is going anywhere. Perhaps I am either evil or an idiot. Probably we should leave it at that.
Tom, your use of the term "open standard" is open to confusion. The various standards produced by ITU-T are described by them as "open standards"
What is meant by this is that the process for creating the standard is transparent, participation is open, freely available documentation of the proposals and research results supporting the proposals (including proposals that don't make the cut), and "reasonable and non-discriminatory licensing" of patents. This is a stark contrast to closed, proprietary systems. For example, in the early aughties Microsoft was trying to get their codec (Windows Media Video) adopted as THE advanced compression protocol, without telling anyone what was in it. Eventually, due to the competition of something known as "ISO MPEG-4 part 10 Advanced Video Coding / ITU-T h.264" they decided to open up WMV to the standards process with an organization called SMPTE. They gave up control (to a bunch of argumentative, anal-retentive standards wonks), and had to reveal to the whole world what was in it. Once the mystery was revealed, it was greeted with a ho-hum. (VC-1 is the name given by SMPTE. It's largely irrelevant.)
I don't know if you appreciate the huge bullet that was dodged. This was a major turning point. (As someone who was involved in the process of SW Bell (now AT&T) turning towards 264 and away from WMV, I think it was a huge victory against closed media protocols.)
While everything in the h.264 (and MPEG-2 and AAC and ...) protocols are completely out in the public, and there are freely available implementations (i.e. "test models"), commercial use of these is a different matter.
There are certain totally awesome techniques that are covered by patents. A great example is "B frames". These are video frames that are predicted from previous and subsequent frames. It's a pretty obvious technique to a video compression geek *in retrospect*, but a researcher at Bell Labs came up with this idea and got a patent on it. Someone else can't just pick up that idea and make money off of it without paying.
Copyright also comes into play. One can start with the copyright-free test model (which is spaghetti code written by academics) and try to make an implementation -- a lot of work. Tapping into open source code bases (e.g Source Forge) is another approach. There are strings attached that some companies can deal with, and others don't want to deal with. What some people do is license code from public institutions like Fraunhofer and HHI, who use software royalties to help fund their research activities.
While the software royalties can be avoided by making a novel implementation or using open source, the patent royalties can't be avoided.
Companies are worried about both copyright violations and patent violations. When it comes to open source issues, they employ people whose only job is to pour through code and look for instances of SW engineers doing cut and paste operations that can contaminate an entire code base. They also "give back" to the open source community in a hope of providing some inoculation against the wrath of said community if a violation slips through.
In the case of patents, they try to build up a portfolio to be used in defense against infringement claims against other companies. But that does no good against those companies whose only revenue is patent royalties. This is something that legal departments worry about.
Companies are comfortable with MPEG and ITU-T. Digital video and video conferencing have been based on these approaches for about 20 years now. There are certain aspects of the MPEG-LA approach that have people uncomfortable, but this isn't enough to get people to abandon an approach that has worked for a couple of decades.
Open protocol standards (as defined above): imperfect.
Closed protocol standards: bad.
Open source protocol standards: scary for commercial entities.
The reason the open source approach is scary is two fold: the intellectual property underlying the algorithms may not be vetted and certainly hasn't been packaged in a reasonable and non-discriminatory patent license portfolio, and secondly, open source isn't free as in "free beer" it's free as in "publish all of your source code that you've spent many years and millions of dollars developing".
Microsoft assured everyone that they had the patent issues covered in WMV -- it was all their own stuff. Once it was public, it was clear they didn't.
China has developed an advanced video codec that does not* infringe on anyone's patents. This is called AVS.
* The word "not" is open to interpretation. No one is going to attempt to deploy AVS in countries that enforce patent rights.
In a quick sampling of comments on the interwebs, it's interesting to note the hostility towards MPEG-LA but absolutely no suspicion towards On2's motives. This is pure fabrication, but how about this ... On2 gives away inferior late 20th century technology (VP3, which is the basis for Theoria) to undermine the adoption of superior industry standard protocols which would crystallize the market. In the meantime they continue developing their own proprietary technology (currently at VP8) with the hope that the market instability they previously unleashed means there's a bigger opportunity.
I mean, if we're going to assume commercial entities are out to fuck us all, why let any of them off the hook?
Jim, sorry, I thought you had seen the stories about Apple's rejections of iPhone Apps with political content (specifically, satire), that's why I only mentioned it and didn't provide links. Here are a few stories:
1 / 2 / 3
I hear your arguments and was an Apple user from 1987-2002. They lost me with the iPod and what seems like an escalating effort to control the user's experience. I know you said I could strip down an Apple computer but I don't even like the design of their products all that much. (Give me a grey box any day.)
Mark, thanks for the info, it's very detailed and insightful. I know patent is a swamp but that cuts two ways: if the big fear is companies that make their money shaking down others for patents, a big legal department can grind them down, right? Who is going to run out of money first?
I also find the all enveloping aspect of the itunes/apps store creepy.
I frankly haven't put a lot thought into theoria over the past few years, because it's not relevant to the broadcast business. For starters, it doesn't do interlace. While I'd love for interlace to die a well deserved death, there's too much interlaced video in the archives and in current production. For any hope of being ubiquitous, a standard needs interlace.
Also, VP3/Theoria is a snapshot of video compression protocol from about 10 years ago. Yes, they've done some tweaks, but there's some major advances that are present in h.264, wmv, etc., etc. For many applications, compression efficiency is the crux of the issue. Satellite, cable and telco TV all have various bottlenecks to deal with. If Theoria was out in front of the other competing codecs, I think they would find a way to overcome their discomfort. (Look at how much headway wmv made in arenas dominated by MPEG-2. What got them there was claims of superior protocol which were based on the fuzzy science of comparing implementations. Msft was perceived to have an advantage until good implementations of h.264 started to show up.)
Legal departments rail against GPL software (perhaps the most "open" of open source licenses.)
"Thou shalt not ever under any circumstances ever, ever, ever use GPL! It exposes the company to too much risk."
"Um, dude, Linux? You're soaking in it."
If an open source product offers enough advantage, it will get adopted. I don't think Theoria crosses that threshold. Linux clearly has.
great thread. please stick with it. there are still some good points made and good questions asked that have not been addressed and are still sitting on the table.
I just feel like I've spent a bunch of time in an argument where Tom and I won't ever see eye to eye since we disagree on some of the starting points. So, after having tried for some time, I'd rather write about things that I see as important rather than defending charges I'm not particularly worried about and that I see as beside the main point. But I agree there is a bunch of good stuff in this thread. Marks comments in particular are extremely interesting and enlightening.
So I'll try to just say again, hopefully more clearly, why I'm excited about this stuff.
The Apple App Store is a closed ecosystem. Apple reserves the right to pretty much reject anything they don't like (although they try to spell out the rules as clearly as they can.) If the App store was the only way to get applications onto the iPad (or iPhone) then this would be a huge problem (although not really, since there is no chance the iPad would be interesting or important or successful if this was true.) But the App Store is not the only way to get apps onto the device, so I don't feel strongly about whether a political bobblehead application was rejected by Apple (one week later they reversed course and allowed the app after it was reviewed by someone higher up the food chain at Apple.) Other apps have been rejected for, mostly I think, nudity or pornographic content. But also for what might be more troubling reasons - I guess like the app where you could throw a shoe at George Bush which was rejected and never reaccepted. Or, more troubling to my mind, applications that compete with features offered by Apple. But these are minor disappointments to me and don't factor very largely into my assessment of the device, or of Apple's trajectory in terms of moving the computing experience forward.
The reason I'm not concerned about Apple's control over the App Store is because going through the App Store is not the only route available to developers. And this gets to one of the main reasons why this device, and how Apple is thinking about the future of computers, is so exciting. And so open to everyone. Namely: web applications are basically* first class citizens on the iPad. And there is no censorship of the web on this device, and definitely no indication that Apple has any plans to censor the web. It would be suicide if they did. Plus it's basically impossible (see China as an example.) It's just really not going to happen. If you don't believe me on that then we're definitely going to disagree.
Now maybe I'm biased since I develop web applications, so this direction plays into my strengths. And I'm not the only one. This Joe Hewitt post says a bunch of stuff I believe much better (and with much more authority) than I could. Joe is the FaceBook iPhone developer (i.e., huge job) who somewhat famously quit developing iPhone apps because of Apple's control over the App Store approval process. But he's excited (to put it mildly) by the iPad, and the reason is the same as what I'm trying to say here. Web applications are the future of our computing experience, and the iPad is the device on which most people will see that future for the first time. I've been talking about web applications evolving to replace desktop apps for many years, and the moment is finally here.
[n]ew web browsers have been built around WebKit such as the S60 browser on Symbian mobile phones, Midori, Shiira and Google's Chrome browser, and it has also been adopted as the rendering engine in OmniWeb, iCab and Epiphany replacing their original rendering engines. Epiphany supported both Gecko and WebKit for some time, but the team decided that Gecko's release cycle and future development plans would make it cumbersome to continue supporting it. Palm's WebOS, the Palm Pre being the first phone device to carry the OS, is based on WebKit.Basically, all modern mobile browsers (S60: Nokia, iPhone, Android, Palm Pre) are powered by WebKit thanks, historically, to the KHTML project, and currently and much more importantly Apple. The WebKit team, led by Dave Hyatt, are simply kicking ass and almost single handedly (Google is of course doing a lot too) pushing the capabilities of the open web forward.
iPad is an incredible opportunity for developers to re-imagine every single category of desktop and web software there is. Seriously, if you're a developer and you're not thinking about how your app could work better on the iPad and its descendants, you deserve to get left behind.Sure, on the other side of the iPad equation, the App Store is not completely open. There are some great and beneficial reasons for this (Hewitt goes into them, but it's about stability and security of the device,) as well as some that make Apple seem big brother-ish. They want that side of the device (the applications you buy through them) to work correctly and be safe (thus they have to all be extensively vetted by Apple) but it's clear they also want to create a family safe place (no porno) where that definition of family safe is going to rub some people the wrong way ("why can't I demean public figures in my application?") I very much want Apple to do the first part of this (safe & secure - like web apps already are by design since they run in a sandbox) while I'm obviously (or maybe not so obviously, but I am) against the stronger steps they've taken for in terms of the second half. But since the whole point is that the web is the future, and the web is totally uncensored and uncontrolled by Apple, I think dwelling on the minor points of App Store TOS is seriously missing what's going on. The App store is going to be great for L.M.'s Mom but definitely isn't what's great about the iPad or the future.
Now, for me, back to Augmented Reality on the blog since I still have to tie that and a few other things together with the iPad (and the web as the future of computing) to move towards my larger point.
Thanks for the insight on web apps. (Web 3.0?)
I've been using and updating a google doc for a couple of years. It's just one stupid spreadsheet, but the power of that idea is one of the reasons I think we're in a post-platform-war environment.
(I'm still waiting for google to have add an rbase doc.)
Another take on the iPad. He's right that the iPad won't offer an opportunity to tinker, but it seems like there are plenty of routes to tinkering that weren't open at the dawn of personal computers.
If you can't demean public figures in an application imagine trying to write one as a tool to organize an anti-government or anti-corporate protest. Or change sexual mores through art some dweeb views as pornographic. I'll stop giving examples, since I know you're tired of me being a fly in your Apple ointment.
Right, you do that on the server side and distribute it as a web app where there is no censorship. Problem solved. You can't use Apple's payment mechanism, but that seems fair to me. From the users point of view there is no difference (i.e., they can still put the icon on their home screen, and it feels just like a native app.) So you can demean public figures or display pornography or whatever you like on the iPad, you just can't force Apple to facilitate a financial transaction for this sort of thing (but you are free to make a financial transaction in your web app.) Apple is just putting a moral sandbox, so to speak, around one particular way of getting apps (apps coming from the Apple approved App Store.) It's not to my liking, but doesn't bother me all that much either since there is another way to do whatever you want.
The store is where 99% of people will get apps for Mac products. Mac is the leading edge of the new web. Hence, "control of the web"--you can parse this argument a million ways, I know.
Back to something Mark wrote: "If an open source product offers enough advantage, it will get adopted. I don't think Theoria crosses that threshold. Linux clearly has."
We changed the subject here from "Theoria is bad because of the threat of submarine patents" to "Theoria is bad because its core functions aren't as good." Maybe those are related but we had been discussing two more or less equal standards up until this point.
You're right that the benefit and risk aren't separable topics. My quick assessment of Theoria (from an "algorithm toolbox" perspective) is that it's okay, not great. It's more likely that it's comparable to MPEG-4 part 2, which also never set the world on fire. But I just don't see how it could be equal to MPEG-4 part 10 (aka H.264).
I think there's a fair amount of buzz about Theoria being as good as h.264, and there's even a test that "proved" it was better. Despite all of that, I'll go out on a limb and say it's not as good. It lacks interlace, which is huge for traditional video. It also lacks bi-predicted frames, which is a very powerful tool for compression. I'm sure given the time line of VP3, it lacks other critical tools, but I without digging in much deeper, I can't say.
It's easy for mythology to grow regarding the comparative qualities of video compression codecs. Performance varies widely with material. Performance varies widely with the quality of the implementation. And the assessed performance varies widely with assessment technique.
Example: In the early days of h.264, experts in the field were fooled into thinking h.264 had 4x the compression performance of MPEG-2. There's a commonly used test sequence called "mobile and calendar" that is way easier in h.264 (1/4 pixel vs. 1/2 pixel motion estimation is a major factor). But that scene misrepresents the typical improvement available in h.264. (2x is a better estimate.)
Example: The first implementations of MPEG-2 broadcast encoders needed about 5.5 to 6 Mbit/sec to do SD. And it didn't look great. Today variable bit rate MPEG-2 is often done in the 1.5 to 2 Mbps range (average bit rate) and constant bit rate MPEG-2 is often done in the 2.5 to 3.5 Mbps range. The MPEG-2 protocol hasn't changed, but the implementation of encoders has. (Those rates are for broadcast quality rather than internet quality.)
Further, comparison of video quality can't be done with a video quality meter. No such thing. Actually there are measures ... PSNR, JND, etc., etc. But none of these is adequate for anything more than gross comparisons. People use them, but experts don't consider them authoritative assessments of quality. For example, one can introduce distortion that is very difficult to see. A PSNR measure will weight that equally with distortion that's easy to see. Another example, noise reduction prior to compression is a common technique to improve picture quality and compression efficiency. Since noise reduction changes the image, it will show up as distortion in a simple PSNR measure -- when it actually improves the picture from a viewers perspective.
Shorter: comparing the performance of codecs is a sub-specialization of a specialized field, and is easy for non-experts to fuxxor up.
I haven't done a critical viewing comparison of VP3/theoria and h.264, but I'm very, very skeptical that it's equivalent. It might be closer to baseline profile h.264 (which is what a lot of people use today on the internet), but main profile h.264? Not likely.
Very good + capped financial exposure >> Good + unknown financial exposure.
couple quick thoughts indeed.
Thanks, Mark. I wonder if Apple went into at the level you went into it. I bet not.
I read the Hewitt article Jim linked to and it sounds like Joe drank the Apple-aid. Interesting insight into how developers talk to other developers: "now that I have a largely-accepted system that those ordinary users are locked out of and can't screw up, I feel an enormous sense of freedom to create Apps--it's like Viagra." Well, bully.
I for one don't mind reinstalling my OS periodically--clears out the cobwebs and makes me feel empowered.
I'm going to link to this discussion from my blahgh--I appreciate the work everyone's doing to make these issues comprehensible to others.
From time to time we get reports in the local paper of high tech execs talking smack to each other. Here's a relevant one that follows up on an earlier report of comments Jobs made at an Apple employee meeting. Jobs was jabbing both Adobe and Google.
It is very amusing to read an Adobe executive saying "open access has proven to be more effective in the long term than a walled approach."
for the life between buildings, ipad notes.
More on theoria v. 264. Conclusions on compression support my comments. I haven't looked at how he got there. It's a non-trivial eval process.