How about if I just want to watch a video on my laptop while I'm travelling, or when I feel ill and stay in bed? Mighty good a tv card will do that I can't attach to my laptop or for which I have no cable connection available... And don't forget about all the people who are still on dial-up. Not everyone has the fortune of having a 100MB fiber connection at home. Downloadable videos provides the most freedom to watch video when someone wants to. (which is the reason VCRs were invented, remember?)
No problem on mine, but then I've not bothered upgrading from Dapper Drake.
I use the via driver, running 800x600x24 on the TV out (PAL) and only the cpu intensive H.264 codec is too much for the 1GHz cpu.
I do plan on either upgrading Ubuntu or installing FreeBSD 7.0 though, depending on how good the driver for the Terratec audio card is.
I wonder if many will complain when it's ads for Sloggi.:-P
Anyway, seems to me it's about time more and better sandbox programs appear so games like this can be cut off from the OS itself.
No probing what OS, applications and colour of my shirt is, just run the damned game for which people paid good money. Otherwise I'll just stick to good ol' Simcity, unless the development of games for *nix/*BSD skyrockets.
if a web page is put up on a server, and nobody is there to surf it -- does it make an impression?
Seems like it. Not sure how, but mine went from the 12th page to rank 4 on the search for "nystrom", (just checked) rank 8 on "setup terminal server trueview 2007" and hardly anyone looks at it (except for plenty of visits from msnbot and Yahoo! and I'm not even in the top 10 there). The only thing I can imagine is it did because of the links in my signature here and on OSnews.com. I do have a sitemap XML file, but only since a few days because I was curious if it would really get used by any bots.
I believe you wanted to say it's not as good as Japan?
Over here in my town (Eindhoven , the Netherlands) I was lucky to live in the right spot where we got free installment and one year of fiber connection. After that year we could choose to continue and had a choice of extras like telephony and tv channels through the fiber.
Last month I saw the sign put up announcing 1200 more homes connected to fiber in a neighbourhood next to mine.
I can't tell how it is in Japan, but fiber is starting to grow here and in more cities in the Netherlands, connectivity won't differ much from Japan I'd say.
..check out the wonderful little magazine The Audio Critic, it takes a science approach to audio equipment and in the downloadable back issues there's more about the myths preached by the audio reviewers and the debunking by double blind tests. Even reading some of the reactions to this by the reviewers make me think they's quite the religious group, but I have to admit that the editor can be very harsh too. (one of the best examples is a NY high-end dealer refusing to pay the reward for proving he couldn't hear the difference between his 5 figure amplifier and a 100$ Pioneer receiver)
Nvu and Kate is what I use. I really dislike sites that use flashy content but don't have an alternative access to a simplified version. Whenever I encounter one I won't even bother using it in the future.
Synergy is something completely different. That is used to control a second pc from the main pc (I use it daily to control my FreeBSD workstation from my Windows laptop). If anything Synergy is like VNC without the viewer on screen.
This "invention" splits a pc in half so by plugging in a second keyboard and mouse two people can work on 1 pc, something that has been able on Linux pc's years ago for use in schools. Only the Linux invention isn't a half-assed idea as everyone gets to use their own monitor and not have to do with just a small piece of screen real estate. (and at that time the pc could support even up to 4 people on it)
I wonder what next bit of old OSS solution will be "invented" by Microsoft and delivered with only half the features...
With simple replication, if each of the three files is different in different places, you're screwed.
Actually, then you're not. As long as the corrupt part of the file is okay in the other two that part can be retrieved (you don't check the file as a whole but block by block). So as you said yourself, as long as the three files are different in different places it doesn't matter. Only if two of the files have the same part corrupted does it become a manual job to retrieve the right part. But then I think there's much less chance of two copies being corrupt than one. And if it's still not good enough, get four or more copies.
As far as I can see about AFS is indeed the DFS structure, but on top of that would have to be a process which would check for differences between the volumes and correct any faults. The issue here wouldn't be so much as having a distributed set of data for better availability, but having a system that can detect and automatically correct data corruption. This is not something I can find as a feature in the AFS FAQ.
If there is corruption, how can an automated system be sure it's the data or the checksum that's corrupt? To make sure you'd need two undamaged copies to tell which is the third corrupt copy. With data storage getting larger and cheaper there's no reason to not have 3 copies of data. And the more copies the more reliable the consistency of the data.
Splitting the data files wouldn't be enough. You'll need at least 3 copies to compare against one another to see if there's bitrot. If there's a difference in one file compared to the other two it is most likely that part is corrupted, and it should be overwritten by the data from the other copies. I don't know how redundant.par files are, but if it can be redundant like this it should work.
Slashdot Mirror
Here's a list of the major changes and the reasons behind them.
How about if I just want to watch a video on my laptop while I'm travelling, or when I feel ill and stay in bed? Mighty good a tv card will do that I can't attach to my laptop or for which I have no cable connection available...
And don't forget about all the people who are still on dial-up. Not everyone has the fortune of having a 100MB fiber connection at home.
Downloadable videos provides the most freedom to watch video when someone wants to. (which is the reason VCRs were invented, remember?)
No problem on mine, but then I've not bothered upgrading from Dapper Drake.
I use the via driver, running 800x600x24 on the TV out (PAL) and only the cpu intensive H.264 codec is too much for the 1GHz cpu.
I do plan on either upgrading Ubuntu or installing FreeBSD 7.0 though, depending on how good the driver for the Terratec audio card is.
I wonder if many will complain when it's ads for Sloggi. :-P
Anyway, seems to me it's about time more and better sandbox programs appear so games like this can be cut off from the OS itself.
No probing what OS, applications and colour of my shirt is, just run the damned game for which people paid good money.
Otherwise I'll just stick to good ol' Simcity, unless the development of games for *nix/*BSD skyrockets.
Aren't we supposed to upgrade to OSX/Linux/BSD as /. inhabitants?
BTW, aren't you tired from running such a long time? (strike one alcohol induced bad joke from my to-do list)
if a web page is put up on a server, and nobody is there to surf it -- does it make an impression?
Seems like it. Not sure how, but mine went from the 12th page to rank 4 on the search for "nystrom", (just checked) rank 8 on "setup terminal server trueview 2007" and hardly anyone looks at it (except for plenty of visits from msnbot and Yahoo! and I'm not even in the top 10 there).
The only thing I can imagine is it did because of the links in my signature here and on OSnews.com. I do have a sitemap XML file, but only since a few days because I was curious if it would really get used by any bots.
I believe you wanted to say it's not as good as Japan?
Over here in my town (Eindhoven , the Netherlands) I was lucky to live in the right spot where we got free installment and one year of fiber connection. After that year we could choose to continue and had a choice of extras like telephony and tv channels through the fiber.
Last month I saw the sign put up announcing 1200 more homes connected to fiber in a neighbourhood next to mine.
I can't tell how it is in Japan, but fiber is starting to grow here and in more cities in the Netherlands, connectivity won't differ much from Japan I'd say.
Anybody else felt dyslexic and read Space monkey invented for space tourists?
Who the hell wants to start a zoo in orbit?
I'm surprised no other /.er posted this. There must be some of out there who love music?
*taps on your shoulder and points to page 2*
:-P
..check out the wonderful little magazine The Audio Critic, it takes a science approach to audio equipment and in the downloadable back issues there's more about the myths preached by the audio reviewers and the debunking by double blind tests.
Even reading some of the reactions to this by the reviewers make me think they's quite the religious group, but I have to admit that the editor can be very harsh too.
(one of the best examples is a NY high-end dealer refusing to pay the reward for proving he couldn't hear the difference between his 5 figure amplifier and a 100$ Pioneer receiver)
Now where's the obligatory "640MHz is enough for everybody" line?
I think it was some 2000+ years ago, some guy with an animal fetish. I believe his name was Noah.
Best. Analogy. Ever. :-)
I'll go for more /. insightful..
But, will those 775 run Linux?
:-P
Offtopic! Who cares about your rooster?
Nvu and Kate is what I use.
I really dislike sites that use flashy content but don't have an alternative access to a simplified version. Whenever I encounter one I won't even bother using it in the future.
Synergy is something completely different. That is used to control a second pc from the main pc (I use it daily to control my FreeBSD workstation from my Windows laptop). If anything Synergy is like VNC without the viewer on screen.
This "invention" splits a pc in half so by plugging in a second keyboard and mouse two people can work on 1 pc, something that has been able on Linux pc's years ago for use in schools. Only the Linux invention isn't a half-assed idea as everyone gets to use their own monitor and not have to do with just a small piece of screen real estate. (and at that time the pc could support even up to 4 people on it)
I wonder what next bit of old OSS solution will be "invented" by Microsoft and delivered with only half the features...
Be patient, they're almost done working on the monkey suit for Steve B. for the dancing monkey icon.
I'd say "You must be new here.", but then that would mean you bought that /. ID from Ebay.
Also funny, mine opens my luggage. O.O;
With simple replication, if each of the three files is different in different places, you're screwed.
Actually, then you're not. As long as the corrupt part of the file is okay in the other two that part can be retrieved (you don't check the file as a whole but block by block). So as you said yourself, as long as the three files are different in different places it doesn't matter.
Only if two of the files have the same part corrupted does it become a manual job to retrieve the right part. But then I think there's much less chance of two copies being corrupt than one. And if it's still not good enough, get four or more copies.
As far as I can see about AFS is indeed the DFS structure, but on top of that would have to be a process which would check for differences between the volumes and correct any faults.
The issue here wouldn't be so much as having a distributed set of data for better availability, but having a system that can detect and automatically correct data corruption.
This is not something I can find as a feature in the AFS FAQ.
If there is corruption, how can an automated system be sure it's the data or the checksum that's corrupt? To make sure you'd need two undamaged copies to tell which is the third corrupt copy.
With data storage getting larger and cheaper there's no reason to not have 3 copies of data. And the more copies the more reliable the consistency of the data.
Splitting the data files wouldn't be enough. You'll need at least 3 copies to compare against one another to see if there's bitrot. If there's a difference in one file compared to the other two it is most likely that part is corrupted, and it should be overwritten by the data from the other copies. .par files are, but if it can be redundant like this it should work.
I don't know how redundant
You were thinking of a clusterfuck too?