more positive than having to take it back to a store or having the company assume i'm stupid so they limit development on the phone so i don't 'muck it up'...
But less positive than if it didn't break in the first place.
The Axims and Treos are wandering around with sales people. To my knowledge, the dance you speak of isn't 1) common, 2) causing mental anguish.
1) I didn't claim it was common. I simply claimed that as a troubleshooting procedure it's unacceptable.
2) The sad fact is that for most people it doesn't cause mental anguish. Our standards for phones are so low that we just accept things like removing the battery to find a hidden button and then resetting the phone to factory spec as par for the course.
Somehow the world has been able to have usable experience with smartphones/PDAs before Apple decided to boon us with the iPhone
I do not think the iPhone will do things like replace oxygen or solve global warming. I am excited that phones will start to be seriously evaluated in terms of usability as well as features. We take phone software's awfulness for granted, but we'd never accept the poor usability and low software quality on our home computers or from a webapp.
If nothing else, the iPhone really brings that aspect to the fore of public awareness. The products we've worked with up until now have been usable as you say, but I don't think anyone would say their usability is "good".
If you have a Treo running Windows Mobile or Palm OS you just do a hard reset and you're back to how it was when you opened the box. Why would users have to go back to store? Even then, the person at the store will just do a hard reset for the customer because they can't read the half page quick start/troubleshooting guide. It's a matter of pressing and holding two buttons, one of which you can't normally press since it is behind the battery.
Oh. So if my treo is acting up because of my third party apps, all I need to do is remove the battery, find the button behind it, and press and hold that button in unison with another button on the opposite side of the case? What could be simpler?
I wouldn't say this is a nightmare support experience at all considering all the software out there for these phones.
For Xenu's sake, listen to yourself! You're so desensitized to the awful experience and software quality of modern phones that you think the battery-removal dance is a positive experience!
Also, the notion of "no accountability" is suspect as well. With the massive array of lawsuits pointed at Sony/BMG, they're going to get creamed. Sure, they're litigating, but that's because of their fiduciary responsibility and the way our legal system is constructed. If the board said, "We're going to settle out of court for the terms named." then I'm reasonably certain there would be a new board in within 24 hours.
Fortunately both HD-DVD and Blu-ray have had their DRM been broken (they use the exact same DRM scheme). And it's staying broken. So we can start to factor the DRM out the way we did with DVDs once DeCSS started happening.
because the market/consumers have no real need of new media types
The recordable data disk market is in dire need, imho. A factor of 10 (or 5 if you're picky) reduction (4 -> 40gb) is a huge deal. That is the only real interest I have in the format war, I want the better data format to win. That means Blu-ray, because it has better per-disc capacity and a higher minimum data throughput.
I hope that's rational enough to cut the mustard.:)
I'm not saying which one you should prefer, but lots of people either hate HD-DVD or Blu-ray on irrational basis. For example, "HD-DVD players break too much! (even though I don't own them and the current generation is just fine)" or "I hate Sony/BMG, therefore I will boycott all of Sony (even though the connections between various divisions of such a large company are extremely slim)." Some people are even so foolish as to have decided-retroactively, of course-that the format they purchased is the superior one because, well... they spent a lot of money!
There is no reason to hope both lose. I'd really hate to be suck with DVDs for several years while the next-next-gen media gets its act together, and probably does the exact same thing all over again.
Me, I prefer Blu-ray because Sony takes their recordable-data business seriously and they're getting that stuff to market much faster. You might prefer something else, like HD-DVD because the hardware is a bit cheaper. Either way, there are plenty of rational non-fanboy reasons to prefer formats. The most irrational view I can think of is your position. How would the completely failure of the new media types benefit the market or consumers?
This solution seems chaotic to me. Now, instead of needing to pull all the changes from one central repository, I need to pull changes from the machines of all my co-workers individually?
Well sure it seems chaotic and different because you're not used to the idea. Assuming you're working with just 1-5 other people, it's a fairly simple cognative load. Heck, you'd probably even script it, so it'd just happen.
I submit to you, the reader, that the subversion method is pretty chaotic too. Because of the "Thunderdome" style of launching all patches into trunk without regard for if the build works, it can be really unclear if your checkout works, has passing tests, or any other thing. All you can do is hope the logs are accurate. And this assumes you have the patience to wait for SVN to tell you these things, given how slow it can be.
To me, that's one of the worst case scenarios. Because responsibilities are often delegated in this kind of situation, you seldom have any idea about the code that's being worked on "over there." So if it breaks in a reasonably-sized project, you're somewhat screwed.
Wouldn't this system make it difficult to guarantee that each developer was integrating the work of the others?
There is almost exactly the same amount of integration work. The difference is that you can defer it, or foist it off on other people. These other people may find a merge that baffles you to be utterly trivial.
Also, it doesn't seem very scalable. What if I have 20 co-workers?
Actually, it works fine. What you do naturally with such a large group is that you begin to delegate. You say, "It's Alice's job to get everyone's patches for this component, and she, Bob and Carlyle with be working on that part." Their patches feed up through her, and then you pull from Alice, confident that she's dealing with that part of the software.
Obviously git scales to the large-delegated-group solution, that's where it's being used to greatest effect (i.e., the linux kernel).
Exactly. This includes not thinking the PC games I play are "trite" or "boring" - maybe you just aren't playing the right ones?;) Regardless, it kind of goes without saying that I don't think they suck. If you do, that is fine, but it is not a fault of the platform - it's just not your cup of tea (just as the console games I have played are not mine).
No, no, you have it wrong. It's because of "To Each Their Own" that I can think your tired FPS rehashes are not just trite and boring, but an utter ripoff. But it also means that if you want to keep paying to play the same game over and over, you can feel free to do so. And you can even ridicule me for finding new games instead of playing the same thing over and over.
It also means that if you think that a 24" monitor, some nice speakers and a comfy chair approximates a home theater, I am free to ridicule your seclusion.:) For me, gaming is a social experience, and often people are sitting next to me enjoying it.
So, if you like to leave the house and be outdoors, WHY ON EARTH ARE YOU TAKING YOUR GAMING LAPTOP WITH YOU? For god's sake man, unglue yourself from the machine!
Maybe I'm not.com enough here, but I only have a "Laptop" and not a "GAMING LAPTOP". I have but one, and it is used for work, play, and everything inbetween.
But hey, if you want to hook me up with one or more, that'd be great.
You said "all snarkiness aside" but then continued to be snarky:P
You caught me.
In any event, I am sure the the poster was referring to PC games and, to go further, newer games that are coming out that the video card in the macbook wouldn't be able to handle well (as well as not being upgradeable and the fact laptop vid cards are always slightly behind their desktop counterparts).
Now, everyone can like whatever they like, that's only sensible... but... it just seems like there are two games on the professional PC game circuit, FPS flavor-of-the-month and RTS flavor-of-the-month games. I occasionally play some PC games, and some of them are good, but I certainly am not clamoring for anything that comes out on the PC. All the real public momentum outside of a few key institutions (e.g., Starcraft 2) are console based.
And quite frankly, my HD television setup is such a more powerful medium for gaming than my monitors could ever be, given the price-performance ratios. I know you could jerry-rig something, but PC gaming is poorly setup for this very natural idea, whereas consoles are designed ground-up for it.
I like nice graphics, but what's the point if I can't enjoy it in my home theater? And moreover, I'm not a slave to the eternal-graphics-upgrade saga that consitutes being a PC gamer. I just don't care, I have no desire to be constantly updating my hardware so that I can play good-looking but otherwise boring and trite games.
Oh, and there are currently 0 games I play or am planning to play that are available on consoles.
To each their own, but I think you're seriously missing out on Odin Sphere.
But if you're running games, why are you using a laptop.
I can't speak for you or the grandparent post's author, but I like to leave the house occasionally. A laptop is a good decision for people who occasionally stand, walk, or otherwise engage in self-locomotion.
I can understand using a laptop for a couple simple games,
I try to stick to the simple things, like World of Warcraft, Guild Wars, Quake 4, etc.
but if you're really hardcore into games, then why wouldn't you get a full sized computer.
All that snarkiness aside, I am really into games, so I did exactly what you said. Powerful system, peripherals, and huge monitor. It's called my Wii, PS3, and my HD television.:)
Git might suck badly if you have... or a centralized repository, a different development model, or simply want UI tools & IDE integration.
Git works perfectly well with a centralized repository. This use case is fully supported and actually has some unique support in the git-hooks. However, its discouraged socially.
I am not sure what kind of model git, darcs, mercuirial, bazaar, or monotone couldn't satisfy. They all scale from one person to many (albeit speed concerns with some). They can be used in a variety of ways. The only really annoying part I've ever found is the lack of support for empty directories, but that's what.anchor files are for, I guess.
Git actually comes stock with some of the best UI tools I've ever had included with my version control system. Seriously. Go take a look at gitk and check out git-gui.
As for IDE integration, a google search will show you that's already falling into place, and it took me all of 2 days to get work to play nice with TextMate (my editor of choice).
Don't get me wrong, a lot of people (myself included) would love to use Darcs. It's done a lot of thing right.
We can't.
Darcs is just too slow. You don't need a huge project to reveal this slowness, just a medium-sized one. It's merge algorithm freaks out when two patches appear at the same time that are identical (which happens a lot more than you might thing). It's also too freakin' slow! I'm using git until Darcs can un-slow-ify. Then I'll go back to it (unless, of course, git can simulate what darcs does, in particular interactive pulls).
No really. What toolkit does Linux have that's better than Cocoa? Certainly the only thing that's even _close_ is Qt, and their tech for GUI applications is still a few years behind. The way you build out an app in InterfaceBuilder as a serialized collection of objects that "wakes up" into an application state is absolutely brilliant, and mirror some of the (engineering-wise) best platforms ever devised.
Sure, there are some bugs and undocumented edges, but Qt isn't really different in that regard. So I would be hard pressed to believe anyone who knows both Cocoa and Qt could express some kind of longing for Qt.
Maybe you're upset about Objective-C? Sometimes people seem to think that Objective-C is bad, and this is an opinion that's not directly refutable... but in general Objective-C (and in particular Apple's implemention of it) is a pretty frikkin' awesome. I'm not sure what there is to complain about. So I'm going to rule that out.
Maybe you're pissed about Carbon. Here you have a legit and common complaint. Fortunately, every year (and more importantly, every major OS release) Carbon is driven back by fire into the dank hole of history from whence it came. So it does suck a little, but it's going away. I'm sure you can find at least one important API on linux that sucks to use, so this is a gimme.
What's left? CoreFoundation is a little weird, but not bad. Apple's IOKit is actually a really big step in the right direction. Nearly every library you could want from linux works on a mac already. Like, the only time I really pine for the Fjords-of-Linux is when I wish I had/proc.
So why is it so much more awsome on linux? And what apps have you developed, anyways?
What confuses me most about that article (besides the ugly picture of the inverse hydra) was that they sort of implied that it's impossible for any small company looking to oppose google to make any kind of capital, but nothing could be further from the truth.
What google has done to that space is remove all the BS, not all the "oxygen" as the article quotes. Your product has to be good, your plan has to be merciless, your people have to be dedicated not just to making a new product, but also to actually taking on the giant reverse-hydra on its home turf and its core competency. It's not impossible, it's not even unreasonable, it's just very hard.
As for the salary issue they mention, that confuses me. Who goes to a startup to make a higher salary? People who join startups join for the stock, not the awesome monthly compensation.
Powerset, the company I work for, is a real-life david-size phenomenon. It confuses me that the SF chronicle didn't even mention the SF-based startup that isn't dealing in billions of dollars to do what Microsoft is apparently struggling to do.
AAC is NOT a open standard, unless you consider MP4 to be an "open standard", and it is NOT royalty-free. In fact, I'm pretty sure the licensing for hardware players is slightly more than MP3. This is why most portable audio players don't support AAC, because then they would have to pay double licensing fees (one of MP3, one for AAC) and MP3 is vastly more popular than AAC especially overseas.
1. AAC is as "open" a standard as you are going to get in the audio world. You can get the spec, you can implement it. It's a controlled spec, but it's not like people are making audio codecs every day, nor is there much incentive to update a codec that is in deployment. The failure of widespread HE-AAC (aka AAC+) adoption should demonstrate this more eloquently than words can.
2. There is a licensing fee to sell an encoder, which is pretty small. There are no distribution fees. MP3, on the other hand, is a quagmire of submarine patents, royalties, and fees. Really, it's time for MP3's funeral barge to be let loose. The amount of legal restriction on that format is staggering, the history of it is pretty amazing.
AAC is the future, in a lot of ways. We might wish it were Ogg, but at least the format that will emerge is technically similar with what we wanted, if not legally similar.
You may not realize this, but the Zune plays music in.mp3 format, too, which would be an ideal, universally compatible format for DRM-less downloadable music.
Mp3 is a very old technology, and delivers poorer quality sound per a given bitrate. AAC at 256kbps is incredibly good, and AAC @ 128kb is very good (although Apple hasn't moved to support the extended support for low-bandwidth stuff, so it's not as good as it could be).
Most portable players bought within the past 3 years play AAC. AAC also is less encumbered by patents than MP3, and even better the laws are clear and there don't appear to be any submarine patents/copyrights waiting to trip people up. The legal world of MP3 was a quagmire, and there is no reason to plunge into that world if you don't have to. Apple has neatly sidestepped the MP3 patent issues by using the newer AAC, and gotten a technical advantage to boot.
Jobs didn't want that, though. He didn't want universally compatible music, he wanted ipod-and-only-ipod compatible music, which is why these new higher priced songs are only offered in AAC. It allows him to keep leveraging his near-monopoly between itunes and ipods in the same manner that Apple-DRM-Protected files did.
Since when is AAC ipod and only ipod? "M4P" files are ipod-and-only-ipod because of DRM, not because of AAC. The new iTunes store offerings will be playable on nearly every modern player, and if encoded properly will be about as close to CD quality as you get without hitting lossless. Moreover, there will be a surge of cheap players that play AAC, as if there aren't a ton already.
Please, remove the tinfoil hat. The only hidden agenda Apple has here is that they will sell even more music this way, and will charge people for upgrade info. But I'd definitely like the ability to move my music around, and I'd even pay more for it since it is not the norm in online sales world.
I hope that, now that we have the promise of downloadable iTMS 720p video via the AppleTV, Apple will get their asses in gear and upgrade Quicktime's H.264 support. Forget about DivX, that's a waste of time. What's meaningful is if the new, modern Codecs can be played easily.
The problem with Apple's H.264 support is that it only supports the Baseline and Main Profiles, and lacks the popular B-pyramids feature which x264 Main-profile encoders use. None of the High or Extended profile features are supported. These features exist for a reason, and they're there to improve your quality/datarate. iTunes Music Store movies are already a fairly long download. To increase them to 720p would make it unacceptably longer.
If Apple can get QT7's H.264 decode (and hopefully encode) support up to par with x264's, then the iTunes Music Store's video downloads can increase in resolution without a linearly proportional increase in file size.
NASA is sorely outnumbered on this issue. I don't claim to know why they decided on the G4s, but I can only assume it's because they have very wise mathmaticians who have decided that they don't need double-precision floating point. Even this presents problems, since C stdlib math functions all implicitly cast to double before doing their work. It's really annoying.
I can only assume that NASA had the resources to work around that problem. I sure didn't when I worked at Lockheed.
And the HPC page you linked is a software description, is that supposed to mean anything other than that yes, the software existed?
Have you heard something that the general public hasn't? As far as I can tell, the only limiting factor so far is how much Yellow Dog has managed to do with it.
Because it was, for a long time. In fact, it still seems to be somewhat faster than Intels SIMD instructions. The Altivec gets its amazing speed by being a very, very simple piece of hardware. Using it properly is not easy at all, and in some cases it simply can't provide the precision you want.
But Apple and Adobe seem to have discovered that it's even faster to have a dedicated GPU do this work. And so the important use cases for SIMD have evolved in Mac OS X, edging more towards a scientific and gaming bias. But anyone can tell you that the Altivec, with its lack of double precision floating point support, is not well suited to scientific applications. The difficulty of using it properly is also a major limiting factor in games. Intel chips have other advantages which end up giving a net win to things like CoreImage, but that assumes that you can farm out work to the GPU.
People like to say Apple pulled a 180 on this, and I suppose there is some merit to that. But it's also true that Apple Engineers decided they found a better way outright, and the marketing engine didn't fight them on that, unlike in some companies we've seen (Intel with the P4 architecture, for example).
Slashdot Mirror
But less positive than if it didn't break in the first place.
1) I didn't claim it was common. I simply claimed that as a troubleshooting procedure it's unacceptable.
2) The sad fact is that for most people it doesn't cause mental anguish. Our standards for phones are so low that we just accept things like removing the battery to find a hidden button and then resetting the phone to factory spec as par for the course.
I do not think the iPhone will do things like replace oxygen or solve global warming. I am excited that phones will start to be seriously evaluated in terms of usability as well as features. We take phone software's awfulness for granted, but we'd never accept the poor usability and low software quality on our home computers or from a webapp.
If nothing else, the iPhone really brings that aspect to the fore of public awareness. The products we've worked with up until now have been usable as you say, but I don't think anyone would say their usability is "good".
Oh. So if my treo is acting up because of my third party apps, all I need to do is remove the battery, find the button behind it, and press and hold that button in unison with another button on the opposite side of the case? What could be simpler?
For Xenu's sake, listen to yourself! You're so desensitized to the awful experience and software quality of modern phones that you think the battery-removal dance is a positive experience!
Sssh. Too rational. You'll break slashdot.
Also, the notion of "no accountability" is suspect as well. With the massive array of lawsuits pointed at Sony/BMG, they're going to get creamed. Sure, they're litigating, but that's because of their fiduciary responsibility and the way our legal system is constructed. If the board said, "We're going to settle out of court for the terms named." then I'm reasonably certain there would be a new board in within 24 hours.
Fortunately both HD-DVD and Blu-ray have had their DRM been broken (they use the exact same DRM scheme). And it's staying broken. So we can start to factor the DRM out the way we did with DVDs once DeCSS started happening.
The recordable data disk market is in dire need, imho. A factor of 10 (or 5 if you're picky) reduction (4 -> 40gb) is a huge deal. That is the only real interest I have in the format war, I want the better data format to win. That means Blu-ray, because it has better per-disc capacity and a higher minimum data throughput.
I hope that's rational enough to cut the mustard.
DVD isn't going anywhere, man.
:)
I'm talking about data formats here, not for movie distribution. I'd really like to be storing/burning 10x less disks. Wouldn't you?
I'm not saying which one you should prefer, but lots of people either hate HD-DVD or Blu-ray on irrational basis. For example, "HD-DVD players break too much! (even though I don't own them and the current generation is just fine)" or "I hate Sony/BMG, therefore I will boycott all of Sony (even though the connections between various divisions of such a large company are extremely slim)." Some people are even so foolish as to have decided-retroactively, of course-that the format they purchased is the superior one because, well... they spent a lot of money!
There is no reason to hope both lose. I'd really hate to be suck with DVDs for several years while the next-next-gen media gets its act together, and probably does the exact same thing all over again.
Me, I prefer Blu-ray because Sony takes their recordable-data business seriously and they're getting that stuff to market much faster. You might prefer something else, like HD-DVD because the hardware is a bit cheaper. Either way, there are plenty of rational non-fanboy reasons to prefer formats. The most irrational view I can think of is your position. How would the completely failure of the new media types benefit the market or consumers?
Well sure it seems chaotic and different because you're not used to the idea. Assuming you're working with just 1-5 other people, it's a fairly simple cognative load. Heck, you'd probably even script it, so it'd just happen.
I submit to you, the reader, that the subversion method is pretty chaotic too. Because of the "Thunderdome" style of launching all patches into trunk without regard for if the build works, it can be really unclear if your checkout works, has passing tests, or any other thing. All you can do is hope the logs are accurate. And this assumes you have the patience to wait for SVN to tell you these things, given how slow it can be.
To me, that's one of the worst case scenarios. Because responsibilities are often delegated in this kind of situation, you seldom have any idea about the code that's being worked on "over there." So if it breaks in a reasonably-sized project, you're somewhat screwed.
There is almost exactly the same amount of integration work. The difference is that you can defer it, or foist it off on other people. These other people may find a merge that baffles you to be utterly trivial.
Actually, it works fine. What you do naturally with such a large group is that you begin to delegate. You say, "It's Alice's job to get everyone's patches for this component, and she, Bob and Carlyle with be working on that part." Their patches feed up through her, and then you pull from Alice, confident that she's dealing with that part of the software.
Obviously git scales to the large-delegated-group solution, that's where it's being used to greatest effect (i.e., the linux kernel).
No, no, you have it wrong. It's because of "To Each Their Own" that I can think your tired FPS rehashes are not just trite and boring, but an utter ripoff. But it also means that if you want to keep paying to play the same game over and over, you can feel free to do so. And you can even ridicule me for finding new games instead of playing the same thing over and over.
It also means that if you think that a 24" monitor, some nice speakers and a comfy chair approximates a home theater, I am free to ridicule your seclusion.
Maybe I'm not
But hey, if you want to hook me up with one or more, that'd be great.
I own one. It's "powerful" in that execution can be powerful on it.
Honestly, after all the trouble I went through getting it, it's kinda a letdown. My PS3 gets a lot more use.
It seems disturbingly often that my post starts some kind of cascade effect of people comparing UIDs.
I can't speak for you or the grandparent post's author, but I like to leave the house occasionally. A laptop is a good decision for people who occasionally stand, walk, or otherwise engage in self-locomotion.
I try to stick to the simple things, like World of Warcraft, Guild Wars, Quake 4, etc.
All that snarkiness aside, I am really into games, so I did exactly what you said. Powerful system, peripherals, and huge monitor. It's called my Wii, PS3, and my HD television.
Git works perfectly well with a centralized repository. This use case is fully supported and actually has some unique support in the git-hooks. However, its discouraged socially.
I am not sure what kind of model git, darcs, mercuirial, bazaar, or monotone couldn't satisfy. They all scale from one person to many (albeit speed concerns with some). They can be used in a variety of ways. The only really annoying part I've ever found is the lack of support for empty directories, but that's what .anchor files are for, I guess.
Git actually comes stock with some of the best UI tools I've ever had included with my version control system. Seriously. Go take a look at gitk and check out git-gui.
As for IDE integration, a google search will show you that's already falling into place, and it took me all of 2 days to get work to play nice with TextMate (my editor of choice).
Don't get me wrong, a lot of people (myself included) would love to use Darcs. It's done a lot of thing right.
We can't.
Darcs is just too slow. You don't need a huge project to reveal this slowness, just a medium-sized one. It's merge algorithm freaks out when two patches appear at the same time that are identical (which happens a lot more than you might thing). It's also too freakin' slow! I'm using git until Darcs can un-slow-ify. Then I'll go back to it (unless, of course, git can simulate what darcs does, in particular interactive pulls).
Okay, I'll bite. Which part is worse?
/proc.
No really. What toolkit does Linux have that's better than Cocoa? Certainly the only thing that's even _close_ is Qt, and their tech for GUI applications is still a few years behind. The way you build out an app in InterfaceBuilder as a serialized collection of objects that "wakes up" into an application state is absolutely brilliant, and mirror some of the (engineering-wise) best platforms ever devised.
Sure, there are some bugs and undocumented edges, but Qt isn't really different in that regard. So I would be hard pressed to believe anyone who knows both Cocoa and Qt could express some kind of longing for Qt.
Maybe you're upset about Objective-C? Sometimes people seem to think that Objective-C is bad, and this is an opinion that's not directly refutable... but in general Objective-C (and in particular Apple's implemention of it) is a pretty frikkin' awesome. I'm not sure what there is to complain about. So I'm going to rule that out.
Maybe you're pissed about Carbon. Here you have a legit and common complaint. Fortunately, every year (and more importantly, every major OS release) Carbon is driven back by fire into the dank hole of history from whence it came. So it does suck a little, but it's going away. I'm sure you can find at least one important API on linux that sucks to use, so this is a gimme.
What's left? CoreFoundation is a little weird, but not bad. Apple's IOKit is actually a really big step in the right direction. Nearly every library you could want from linux works on a mac already. Like, the only time I really pine for the Fjords-of-Linux is when I wish I had
So why is it so much more awsome on linux? And what apps have you developed, anyways?
What confuses me most about that article (besides the ugly picture of the inverse hydra) was that they sort of implied that it's impossible for any small company looking to oppose google to make any kind of capital, but nothing could be further from the truth.
What google has done to that space is remove all the BS, not all the "oxygen" as the article quotes. Your product has to be good, your plan has to be merciless, your people have to be dedicated not just to making a new product, but also to actually taking on the giant reverse-hydra on its home turf and its core competency. It's not impossible, it's not even unreasonable, it's just very hard.
As for the salary issue they mention, that confuses me. Who goes to a startup to make a higher salary? People who join startups join for the stock, not the awesome monthly compensation.
Powerset, the company I work for, is a real-life david-size phenomenon. It confuses me that the SF chronicle didn't even mention the SF-based startup that isn't dealing in billions of dollars to do what Microsoft is apparently struggling to do.
1. AAC is as "open" a standard as you are going to get in the audio world. You can get the spec, you can implement it. It's a controlled spec, but it's not like people are making audio codecs every day, nor is there much incentive to update a codec that is in deployment. The failure of widespread HE-AAC (aka AAC+) adoption should demonstrate this more eloquently than words can.
2. There is a licensing fee to sell an encoder, which is pretty small. There are no distribution fees. MP3, on the other hand, is a quagmire of submarine patents, royalties, and fees. Really, it's time for MP3's funeral barge to be let loose. The amount of legal restriction on that format is staggering, the history of it is pretty amazing.
AAC is the future, in a lot of ways. We might wish it were Ogg, but at least the format that will emerge is technically similar with what we wanted, if not legally similar.
Mp3 is a very old technology, and delivers poorer quality sound per a given bitrate. AAC at 256kbps is incredibly good, and AAC @ 128kb is very good (although Apple hasn't moved to support the extended support for low-bandwidth stuff, so it's not as good as it could be).
Most portable players bought within the past 3 years play AAC. AAC also is less encumbered by patents than MP3, and even better the laws are clear and there don't appear to be any submarine patents/copyrights waiting to trip people up. The legal world of MP3 was a quagmire, and there is no reason to plunge into that world if you don't have to. Apple has neatly sidestepped the MP3 patent issues by using the newer AAC, and gotten a technical advantage to boot.
Since when is AAC ipod and only ipod? "M4P" files are ipod-and-only-ipod because of DRM, not because of AAC. The new iTunes store offerings will be playable on nearly every modern player, and if encoded properly will be about as close to CD quality as you get without hitting lossless. Moreover, there will be a surge of cheap players that play AAC, as if there aren't a ton already.
Please, remove the tinfoil hat. The only hidden agenda Apple has here is that they will sell even more music this way, and will charge people for upgrade info. But I'd definitely like the ability to move my music around, and I'd even pay more for it since it is not the norm in online sales world.
I hope that, now that we have the promise of downloadable iTMS 720p video via the AppleTV, Apple will get their asses in gear and upgrade Quicktime's H.264 support. Forget about DivX, that's a waste of time. What's meaningful is if the new, modern Codecs can be played easily.
The problem with Apple's H.264 support is that it only supports the Baseline and Main Profiles, and lacks the popular B-pyramids feature which x264 Main-profile encoders use. None of the High or Extended profile features are supported. These features exist for a reason, and they're there to improve your quality/datarate. iTunes Music Store movies are already a fairly long download. To increase them to 720p would make it unacceptably longer.
If Apple can get QT7's H.264 decode (and hopefully encode) support up to par with x264's, then the iTunes Music Store's video downloads can increase in resolution without a linearly proportional increase in file size.
NASA is sorely outnumbered on this issue. I don't claim to know why they decided on the G4s, but I can only assume it's because they have very wise mathmaticians who have decided that they don't need double-precision floating point. Even this presents problems, since C stdlib math functions all implicitly cast to double before doing their work. It's really annoying.
I can only assume that NASA had the resources to work around that problem. I sure didn't when I worked at Lockheed.
And the HPC page you linked is a software description, is that supposed to mean anything other than that yes, the software existed?
Have you heard something that the general public hasn't? As far as I can tell, the only limiting factor so far is how much Yellow Dog has managed to do with it.
Because it was, for a long time. In fact, it still seems to be somewhat faster than Intels SIMD instructions. The Altivec gets its amazing speed by being a very, very simple piece of hardware. Using it properly is not easy at all, and in some cases it simply can't provide the precision you want.
But Apple and Adobe seem to have discovered that it's even faster to have a dedicated GPU do this work. And so the important use cases for SIMD have evolved in Mac OS X, edging more towards a scientific and gaming bias. But anyone can tell you that the Altivec, with its lack of double precision floating point support, is not well suited to scientific applications. The difficulty of using it properly is also a major limiting factor in games. Intel chips have other advantages which end up giving a net win to things like CoreImage, but that assumes that you can farm out work to the GPU.
People like to say Apple pulled a 180 on this, and I suppose there is some merit to that. But it's also true that Apple Engineers decided they found a better way outright, and the marketing engine didn't fight them on that, unlike in some companies we've seen (Intel with the P4 architecture, for example).