Now, you could argue any codec implementation could be downloaded automatically. But that is not practical or desirable unless 2 concerns are resolved:
1) A reliable way to make the client download the right binary version of the codec (mac? win? linux - and if so what libc? and so on).
2) Security - I don't particularly want to run "InstallJoeBloggsCodec.exe." It could install spyware, put junk in my registry, open a back door server, who knows what.
I would say this is a dead-end unless Java has an API for accessing video hardware (ala XVideo). Without integration with video playback hardware to prevent buffer copying and hardware scaling, it will never be as good as native code.
The only difference between "dmail" and minor Exchange Server deployment change is that the "dmail" scheme is proprietary and comes with vendor lock-in.
Furthermore, "Dmail" apparently goes a step *beyond* vendor lock-in by storing all messgaes on the company's servers! You think MS Exchange is bad? What if Microsoft only sold the client and all Exchange servers resided in Redmond!
Just you wait! Once the number of Magic Box users with Linux rises above a threshold, it'll happen.
So? I really don't think it matters whether linux is intrinsically better, or just targeted less.
Let's say I suggest you switch to Linux. You say, "it doesn't support my hardware and applications." I say, "yeah, but it would if it were more popular." Would you go for that?
This is not the same as a filesystem itself being implemented in an easily searchable fashion, getting rid of the static directory tree structures altogether
Ok, it's not the same. So which is better?
One correction - filesystems (at least most UNIX filesystems) are not constrained to tree structure; the leaf nodes may have any number of parents, i.e. a file may be in any number of directories simultaneously. (Use the "ln" command). And using ln -s you can practically place a directory in any number of parent directories.
I use this to organize my music collection alphabetically by artist, by genre, and by the date I got the music simultaneously. (I tend to be most interested in music I got recently, because I'm not tired of it yet).
I know people tend to organize files and directories in a tree structure anyways. If you ask me that's because people are happy to maintain the analogy of a physical item that can only be in one place at a time - so what does that mean for WinFS?
wouldn't this be the case for more than just chess? Such as checkers, chinese checkers, chineese chess, strategeo, risk, ect.
(Dare i mention the infamous GO in a chess story?)
I don't see how you could cheat in an online GO game at championship levels. I don't play, but as I understand it playbooks aren't helpful and the computer is no help (no known algorithm plays a good game).
Personally, I would have thought an MP3 player with a hard drive was an equally bad idea. That's why I own an a solid state model. But whenever this topic is raised, iPod owners come out of the woodwork to tell about using them for exercise, dropping them, whatever, without damage. I can only conclude that hard drives (especially these little low-mass ones) aren't so fragile after all.
Also note that the phone's disk drive will be deactivated almost all the time - there's no reason to run it constantly, and it would kill the standby time. This reduces the risk even more.
Then again, the phone function might work just fine without the HD. At least I hope they did it that way. I don't see why making a call would require the drive, and it would suck battery juice.
Add to that the fact that MPEG2 does a better job (more efficient) compressing progressive video than it does compressing interlaced video
This is what bothers me about interlaced video - how do you compress it? You could use a separate stream for each field, but that's wasteful since they're bound to be very similar. On the other hand, you can't treat the two fields as a single stream, since they're off by a scanline - image detail would appear in the first field, then disappear for the next field, then reappear in the first field again.
Anybody know how compression algorithms handle interlacing?
here at corperate they are BANNING cd and DVD burners.
hell they just rolled out USB thumb drives to everyone but made sure they are too small to be useable (16 meg) to keep employees for PIRATING their precious software...
WEll long story short, the Thumb drives sit unused in drawers and the sales people still use Floppies....
Email can not be used cince they block ALL attachments now cince the last virus outbreak.
Thanks for the chuckle, and I sure hope you're one of our competitors. If you'd like to consider jumping ship send me a smoke signal sometime.
I guess apple had the right idea a while back when they stopped using floppies...
There's no foresight in predicting that a computer technology will "eventually" be obsolete. I would be much more impressed if somebody predicted that some component on today's computers will still be in use 50 years from now, and turned out to be right.
CDRW drives are now obsolete, since for only a few bucks more you can get a DVD reader/writer that also reads and writes CDs. So would I be some kind of visionary for selling computers with no CDRW drive?
Granted - superficially, it's not the US that's doing something bizarre here, but Australia.
Of course the next question is what sort of pressure the US exerts on Austrialia to get de facto legal jurisdiction there, and whether it's consistent with democratic self-rule Down Under. Somehow I doubt Australia would do the same favor for its less "influential" fellow nations.
As a side note, my biggest complaints aren't the graphics but the logic behind them. How many times do you see someone's arm go into a wall when they run against it? It's a tiny thing, but fixing that (and people hair "cutting through" their shoulders, etc) will make graphics look better.
Well I certainly agree, though I think of that as a physical modeling problem rather than a graphics problem per se. Physics models in games are hardly even in their infancy. You're doing good if you can shatter a window or knock over a lamp. You can shoot off 100 rockets at a barn without making a dent. Even though I haven't played games recently I'm looking forward to see what Half Life 2 does with physics, and whether it contributes to game play.
Now that graphics are very near "good enough", perhaps sound (which in many ways hasn't changed since the SB16 as far as today's games are concerned) will catch up.
I don't think graphics are anywhere in the realm of "good enough." Everytime I get a window seat in an airplane, I look out the window as we descend over a city and wonder how computers will ever approach all that detail and complexity. And then there's the extremely narrow field of view on a computer.
SB16 audio was already far, far closer to human perceptual abilities than graphics are now. You possess 240 million rods, 12 million cones, and exactly 2 eardrums.
I would not jump to the conclusion that the 3.8GHz with 2mb cache is faster than the 4GHz P4 with 1mb of cache.
Yeah I'm aware of the "MHz myth," but look at the benchmarks please. At 3.4GHz, the "Extreme" edition bests the 3.4GHz 1MB cache version by a whopping 1.4%. Meanwhile the Non-Extreme 3.4 beats the Extreme 3.2 by over double that amount! Both are marginal differences if you ask me, but price difference is truly "extreme."
Giving window managers direct access to video memory is rather stupid, considering that one of X's primary features is network transparency.
Choosing network transparency as a primary focus of X was a mistake. Better to optimize for the most common usage scenario, and add bells and whistles afterwards.
it's rather important to recognize that X is a protocol, nothing more, and nothing less. How fast or slow it is depends on the implementation; some are clearly better than others.
It's not true that API design and feature set have no impact on ultimate performance. Two pickup trucks can differ substantially in fuel economy, but expecting either to have the fuel efficiency of an average subcompact is unrealistic.
Furthermore, there are extensions which practically do give an application direct access to the hardware, or at least, as direct as one can safely manage
It's true, and it's nice that X provides a way to work around (most of) X. But this end-run pushes functionality into the application layer, which is part of why X programs' UIs are so inconsistent.
what's the point, if all you're doing is drawing widgets? I could understand it for games or playing DVDs, where speed is important and you'd be stupid to play it over the network anyway, but for a window manager?
It's not to save time or increase productivity, but to get that responsive, "snappy" feel which is more pleasant.
Anyone that has had to administer machines remotely appreciates the network transparency of X.
A GUI for administration? No thanks.
It's the only window system that has this feature.
That's the irony; the "afterthought" methods like VNC and whatever Windows does work at least as well on the network. VNC is much better than X over high latency links, and it allows you to pull up somebody else's session and share it with them, which is fantastic if you're trying to help them do something in the GUI.
The best "feature" of X is the huge set of apps that use it. The best argument for X is momentum.
Well, that was the topic of the 2nd paragraph of the article. Not that you read it of course.
Anyways, TiVo is always going to offer restricted functionality - after all we wouldn't want to open up the possibility of piracy just so you can do whatever you want with your own hardware.
Then there's the fact that they monitor and record everything you do with your TiVo.
Oh, and reserve part of your(?) hard drive to record programs of their own choosing.
And the pesky subscription fees.
And you can't add a second (or third...) tuner to the Tivo to expand its capabilities, or even put in a bigger hard drive without voiding the warranty.
And of course the TiVo can't also be your fileserver, mailserver, webserver, WiFi access point, mp3 jukebox, and DVD player like my under-the-TV linux box is.
I am now going to make a lame computer analogy, so bear with me:)
People "multitask" the same way as a uniprocessor computer - by context switching. I'm with you, people might think they're doing 2 things at once, but mainly they're just context switching. Part of scheduler design is deciding how often to context switch - too little and the system isn't responsive, but too much and throughput is decreased due to all the context switch overhead.
So I think it's important to be aware of each context switch you make, and remember that each one is potentially wasted time.
No, I disagree. Isolation IS the cure. Set aside part of the day to be uninterrupted, and the rest to "do email." Need to plow through a difficult paper? Take a seat in the bottom floor of the library, where nobody goes and there's no WiFi. Maybe I'm the only one who's too easily distracted, but I doubt it.
Are you implying is that CUPS is AOK and printing on Linux is fine? Surely you can't mean that. Most printers aren't supported well or at all, and CUPS is one of those programs that usually doesn't work without a lot of fiddling and never gives sensible error messages. If you're lucky you get some uninformative message in the access_log which repeats every 10 seconds forever.
Binary updates are not a good fit for Gentoo! Not only because most people don't use the binary packages, but because in order to generate the diff, the server must know the exact contents of the file on your system, as well as the exact contents of the updated file. The number of different binary patches would be exponential in the number of compile switches, compiler versions, USE flags, and so on - for both "old" and "new" file versions, so square it again!
I guess if you reall wanted to be clever, you could send the server enough information (your existing package versions, make and USE flags etc), plus the desired flags for your new file. With this the server could compile a binary matching yours, then compile the new binary for you, then make a binary diff. But that, as Kramer would say, is "kooky talk."
Slashdot Mirror
Now, you could argue any codec implementation could be downloaded automatically. But that is not practical or desirable unless 2 concerns are resolved:
1) A reliable way to make the client download the right binary version of the codec (mac? win? linux - and if so what libc? and so on).
2) Security - I don't particularly want to run "InstallJoeBloggsCodec.exe." It could install spyware, put junk in my registry, open a back door server, who knows what.
Java is made for resolving just those issues.
I would say this is a dead-end unless Java has an API for accessing video hardware (ala XVideo). Without integration with video playback hardware to prevent buffer copying and hardware scaling, it will never be as good as native code.
Let's say I suggest you switch to Linux. You say, "it doesn't support my hardware and applications." I say, "yeah, but it would if it were more popular." Would you go for that?
There must be some reason frogs aren't photosynthetic.
One correction - filesystems (at least most UNIX filesystems) are not constrained to tree structure; the leaf nodes may have any number of parents, i.e. a file may be in any number of directories simultaneously. (Use the "ln" command). And using ln -s you can practically place a directory in any number of parent directories.
I use this to organize my music collection alphabetically by artist, by genre, and by the date I got the music simultaneously. (I tend to be most interested in music I got recently, because I'm not tired of it yet).
I know people tend to organize files and directories in a tree structure anyways. If you ask me that's because people are happy to maintain the analogy of a physical item that can only be in one place at a time - so what does that mean for WinFS?
And that's the point; you don't have to worry about running out of space all the time.
Also note that the phone's disk drive will be deactivated almost all the time - there's no reason to run it constantly, and it would kill the standby time. This reduces the risk even more.
Then again, the phone function might work just fine without the HD. At least I hope they did it that way. I don't see why making a call would require the drive, and it would suck battery juice.
Anybody know how compression algorithms handle interlacing?
CDRW drives are now obsolete, since for only a few bucks more you can get a DVD reader/writer that also reads and writes CDs. So would I be some kind of visionary for selling computers with no CDRW drive?
So Dr. Evil was only demanding $1K?
Of course the next question is what sort of pressure the US exerts on Austrialia to get de facto legal jurisdiction there, and whether it's consistent with democratic self-rule Down Under. Somehow I doubt Australia would do the same favor for its less "influential" fellow nations.
SB16 audio was already far, far closer to human perceptual abilities than graphics are now. You possess 240 million rods, 12 million cones, and exactly 2 eardrums.
Yeah I'm aware of the "MHz myth," but look at the benchmarks please. At 3.4GHz, the "Extreme" edition bests the 3.4GHz 1MB cache version by a whopping 1.4%. Meanwhile the Non-Extreme 3.4 beats the Extreme 3.2 by over double that amount! Both are marginal differences if you ask me, but price difference is truly "extreme."
The best "feature" of X is the huge set of apps that use it. The best argument for X is momentum.
Anyways, TiVo is always going to offer restricted functionality - after all we wouldn't want to open up the possibility of piracy just so you can do whatever you want with your own hardware.
Then there's the fact that they monitor and record everything you do with your TiVo.
Oh, and reserve part of your(?) hard drive to record programs of their own choosing.
And the pesky subscription fees.
And you can't add a second (or third...) tuner to the Tivo to expand its capabilities, or even put in a bigger hard drive without voiding the warranty.
And of course the TiVo can't also be your fileserver, mailserver, webserver, WiFi access point, mp3 jukebox, and DVD player like my under-the-TV linux box is.
People "multitask" the same way as a uniprocessor computer - by context switching. I'm with you, people might think they're doing 2 things at once, but mainly they're just context switching. Part of scheduler design is deciding how often to context switch - too little and the system isn't responsive, but too much and throughput is decreased due to all the context switch overhead.
So I think it's important to be aware of each context switch you make, and remember that each one is potentially wasted time.
No, I disagree. Isolation IS the cure. Set aside part of the day to be uninterrupted, and the rest to "do email." Need to plow through a difficult paper? Take a seat in the bottom floor of the library, where nobody goes and there's no WiFi. Maybe I'm the only one who's too easily distracted, but I doubt it.
Are you implying is that CUPS is AOK and printing on Linux is fine? Surely you can't mean that. Most printers aren't supported well or at all, and CUPS is one of those programs that usually doesn't work without a lot of fiddling and never gives sensible error messages. If you're lucky you get some uninformative message in the access_log which repeats every 10 seconds forever.
I guess if you reall wanted to be clever, you could send the server enough information (your existing package versions, make and USE flags etc), plus the desired flags for your new file. With this the server could compile a binary matching yours, then compile the new binary for you, then make a binary diff. But that, as Kramer would say, is "kooky talk."