If rhnsd was leading file descriptors, then no filesystem could fix that, because file descriptors only exist in memory.
One bone I have to pick with ext2 is how the swap partition cannot be adjusted on the fly.
This has nothing to do with ext2. In general, you can't safely resize partitions at runtime. Swap files are probably more flexible, but I haven't used them under Linux.
The max capacity of a fiber may be essentially fixed (as other posters pointed out), but most people aren't using that capacity because they can't switch it fast enough. Since switches are made out of ICs, Moore's Law applies to them, giving an increase in the effective bandwidth over time.
I don't think it would be that easy; the graphics and sound APIs in Linux are totally different from the ones in OS X. Since QuickTime mostly deals with graphics and sound, you're looking at some serious rewriting.
Sorry if I wasn't clear; I don't expect to spend mojo at Wal-Mart. What I meant was right now there aren't many files you can buy with mojo, but in the future I expect there will be.
Stuff like that has been sort of a holy grail for years, usually under names like amorphous computing or cellular computing or RAW, but writing compilers that can turn C or Java code into something that these chips can run efficiently is reallyhard.
I didn't see anything about the die size of the chip, but considering all the stuff that's on it and the fact that it's made on a brand-new.15 micron process, I'm betting it's going to be expensive.
Transmeta doesn't design systems; it's up to other companies to decide what to do with their chips. Those stereo component MP3 players will probably be using Cirrus Logic chips that are cheaper than Crusoes because they use a much slower ARM core.
The other replies were pretty vague, so I'll say it more plainly: Mojo Nation doesn't let anyone run code on your machine. It's no different than Napster or Gnutella or Apache in that regard; other people send requests to your computer and it either answers them or it doesn't. Of course the software might have bugs just like any software might have bugs; that's why it's a good idea to run it under a separate account.
I doubt it. However, a lot of people (like me) have disk space just sitting around unused, so why not rent it out?
Right now there's not much you can buy with mojo, but because of the design I think it will be much easier to get hard-to-find or large files through Mojo Nation than through other file-sharing systems.
The JFS code that IBM released is actually from OS/2 (yeah, that makes no sense to me either) and apparently it's a different codebase than AIX.
Re:Nobody needs basic any more
on
KBasic
·
· Score: 1
Even people in category A will probably start switching to C# pretty soon; it has most of the advantages of Java but also integrates with that Windows thing that so many people are addicted to. So long-term I think a gc# compiler that is supported by KDE Studio might be a better choice.
You're right that there's some FUD in that article, but it wasn't written by Be. The Be engineers know what they're talking about, but some of the fans go overboard a bit.
They's planning to "certify" apps to run on their boxes, and they charge title developers a fee (percent?) to get certified. Presumably apps that aren't certified won't run.
I would love to see a computer which you can just snap together from parts and increase in ram, processors, storage, and such without any limitation, and without turning off machines.
Like the SGI 3000? Granted, the modules are pretty big...
SourceForge has adopted the Trove taxonomy, so e.g. you could easily find out if there are any mature graphical Java apps under the X license that frob frobnitzes.
Besides the fact that the PowerPC 604 actually came after the PowerPC 601 that was in the original Power Macs, many apps actually ran faster under emulation because of the huge jump in raw MHz (the '040s were running at 25-40 MHz, but the early 601s were 60-80 MHz).
People will find ways to use up the bandwidth. Can you say personal TV stations?
But I agree that Net access does not need to be "earned". If someone is willing to sell a service at a certain price and people are willing to pay for it, then what's the problem?
MPEG-4 is open, but it's just a spec. If you want an implementation of it, you have to code one yourself. (ISO provides some reference code, but I don't know if it's any good.) You also have to take care of the patent licensing if you want to use it.
Slashdot Mirror
If rhnsd was leading file descriptors, then no filesystem could fix that, because file descriptors only exist in memory.
One bone I have to pick with ext2 is how the swap partition cannot be adjusted on the fly.
This has nothing to do with ext2. In general, you can't safely resize partitions at runtime. Swap files are probably more flexible, but I haven't used them under Linux.
I got the same impression you did, but what does that have to do with new TLDs or disputes about who owns a particular domain?
The max capacity of a fiber may be essentially fixed (as other posters pointed out), but most people aren't using that capacity because they can't switch it fast enough. Since switches are made out of ICs, Moore's Law applies to them, giving an increase in the effective bandwidth over time.
I don't think it would be that easy; the graphics and sound APIs in Linux are totally different from the ones in OS X. Since QuickTime mostly deals with graphics and sound, you're looking at some serious rewriting.
Sorry if I wasn't clear; I don't expect to spend mojo at Wal-Mart. What I meant was right now there aren't many files you can buy with mojo, but in the future I expect there will be.
Stuff like that has been sort of a holy grail for years, usually under names like amorphous computing or cellular computing or RAW, but writing compilers that can turn C or Java code into something that these chips can run efficiently is reallyhard.
I didn't see anything about the die size of the chip, but considering all the stuff that's on it and the fact that it's made on a brand-new .15 micron process, I'm betting it's going to be expensive.
Transmeta doesn't design systems; it's up to other companies to decide what to do with their chips. Those stereo component MP3 players will probably be using Cirrus Logic chips that are cheaper than Crusoes because they use a much slower ARM core.
The other replies were pretty vague, so I'll say it more plainly: Mojo Nation doesn't let anyone run code on your machine. It's no different than Napster or Gnutella or Apache in that regard; other people send requests to your computer and it either answers them or it doesn't. Of course the software might have bugs just like any software might have bugs; that's why it's a good idea to run it under a separate account.
I doubt it. However, a lot of people (like me) have disk space just sitting around unused, so why not rent it out?
Right now there's not much you can buy with mojo, but because of the design I think it will be much easier to get hard-to-find or large files through Mojo Nation than through other file-sharing systems.
AFAIK, Napster connected their servers recently, so you should be able to see everything.
The JFS code that IBM released is actually from OS/2 (yeah, that makes no sense to me either) and apparently it's a different codebase than AIX.
Even people in category A will probably start switching to C# pretty soon; it has most of the advantages of Java but also integrates with that Windows thing that so many people are addicted to. So long-term I think a gc# compiler that is supported by KDE Studio might be a better choice.
You're right that there's some FUD in that article, but it wasn't written by Be. The Be engineers know what they're talking about, but some of the fans go overboard a bit.
They's planning to "certify" apps to run on their boxes, and they charge title developers a fee (percent?) to get certified. Presumably apps that aren't certified won't run.
If it's really open, then you'll be able to write any app, and you can bet that Napster/Gnutella clients will be some of the first ones.
If it's not open, then they shold stop saying that it is.
I would love to see a computer which you can just snap together from parts and increase in ram, processors, storage, and such without any limitation, and without turning off machines.
Like the SGI 3000? Granted, the modules are pretty big...
It provides RAID 1, but not necessarily 320 GB when using RAID 1. I think they're quoting the raw capacity.
SourceForge has adopted the Trove taxonomy, so e.g. you could easily find out if there are any mature graphical Java apps under the X license that frob frobnitzes.
And on an Internet whose use is driven by games, Napster, and VoIP, almost all "workstations" are also servers.
BeOS is not based on Mach.
Do you have any evidence of that?
Besides the fact that the PowerPC 604 actually came after the PowerPC 601 that was in the original Power Macs, many apps actually ran faster under emulation because of the huge jump in raw MHz (the '040s were running at 25-40 MHz, but the early 601s were 60-80 MHz).
People will find ways to use up the bandwidth. Can you say personal TV stations?
But I agree that Net access does not need to be "earned". If someone is willing to sell a service at a certain price and people are willing to pay for it, then what's the problem?
ASFRecorder is similar to Streambox VCR, except it works with Windows Media. I wonder when MS will try to get all those mirrors taken down...
MPEG-4 is open, but it's just a spec. If you want an implementation of it, you have to code one yourself. (ISO provides some reference code, but I don't know if it's any good.) You also have to take care of the patent licensing if you want to use it.