put European companies at a competitive disadvantage with every other company around the world who does have access to these new technologies.
It could also put them at a competitive advantage by using stable technologies while their foreign competitors play with new thingies. By the time it gets introduced in Europe, the others have already found the first bugs and Microsoft may have fixed some of them.
Here in the Netherlands a static IP is standard for almost all DSL lines, and the dynamic addresses used on cable are almost always "fixed".
It is convenient for the provider and law enforcement, because they can easily track a subscriber's behaviour by his fixed address. Also, it would not be possible to save addresses by using dynamic assignment, as these services are always-on. Usually the modem/router keeps the connection, even when the connected PC is off. So you would have to reserve an address for every subscriber anyway.
But IPv6 is far from being universally used. So, Vista will also support the current IPv4. The side effect, according to Mockapetris, is that a Vista PC will make two DNS requests, one for each IP version, instead of just one.
"It is going to try a DNS lookup for the IPv6 address and then a DNS lookup for the IPv4 address," Mockapetris said. "It just uses more DNS, and until we increase the supply, things are going to go slower."
Is this really true, or is it just FUD? I mean, the DNS protocol supports putting an A and an AAAA question in the same request, doesn't it? Does he mean that Vista doesn't do that? That would adversely affect Vista users as well, as they have to wait two RTTs before they get the A answer they probably want.
Because visitors don't know the difference between bugs in the software they use, and problems with the website they visit. So when they have visited 10 sites that displayed ok, then visit your site and it is a mess, it must be a problem of your site.
Now, this may change with IE7. There will be a lot of sites that people who newly installed IE7 will visit, and will render incorrectly. That may wake up some people.
It is probably because of this that Microsoft does not allow you to install IE6 and 7 alongside. You would be able to compare. That is unwanted, because it would show up problems that they prefer to deny. But, you can still install Firefox and Opera and compare with those...
They cannot install IE7 if they aren't running XP SP2 or 2003 server. It has always been possible to install another browser, but looking at the number of IE5 and 5.5 visits I still see, I think it will be 5 years before IE6 shows any sign of disappearing.
It is a step, but it is very questionable if it is a good thing to make a step when they are still so far of the target of rendering common CSS constructs that all other browsers render without problem. (I do not mean the ACID2 test!)
Now we will have yet another browser to make special exceptions for, different from IE5 and IE6, and we still cannot feed IE7 the same CSS as Firefox, Opera or Konqueror. That is a step, but is it the right direction? I don't know.
The site at work (http://www.uw.nl/) outputs standard html/css to everyone, and uses "IE conditional comments" to feed IE5 and IE6 specialized CSS items to work around their bugs. A workaround sheet for IE7 has not yet been written, but it is very apparent (at least in beta3) that it is not up to the quality in standard CSS handling that the other browsers (Opera, Firefox, Konqueror) are. There are still positioning and stacking bugs.
I hope they fix them before release, but I'm afraid they won't. So this will introduce yet another class of broken browser workarounds: not as broken as IE6, but still broken.
In a country where people behave like that on a really large scale, it is tough to keep a "western civilization" in place. Probably the country will fall down to an antisocial state where everyone only does what he deems good for himself and does not care what the effect on overall society is.
Not that this would be too amazing, but that is already what its government does in world politics. Together with the USA. And the effects are clearly visible.
Of course, as also explained in another posting about such a system operating in Belgium, the tariff structure includes a fixed fee additional to the per-kilogram tariff.
There are several cities here in the Netherlands that already have such a system in operation. Indeed, the cans are weighed and identified by number, and the total amount used to calculate the waste disposal bill.
It is not a bad thing, in principle. I would not know why a family with 3 babies that fills two cans a week would have to pay the same as an individual who does not even have a full can after 3 weeks. Differentiation of payment makes people more aware about what they do.
However, there are problems due to antisocial behaviour. What is frequently seen in those areas, is that people dump their waste in other people's cans when they have been put outside. Also, people take garbage from their home to their work place, and dump it there, especially when such a system is not yet implemented in the area where they work. One would expect such problems to disappear over time (especially when the system is widely implemented).
Note that the advanced video processing offered by PureVideo is supported only on Windows. When you build your MythTV box based on Linux, your nVidia card actually is only an accellerated video card with some MPEG decoding support, but the de-interlacing and filtering is very primitive compared to what is offered in Windows and tested in the review.
So, the "Upconvert and de-interlace DVD content for display on HDTV quality screens and do so better than dedicated players with inexpensive NVIDIA or ATI cards" does not apply in the case of a Linux system.
Service Temporarily Unavailable The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later. Apache/1.3.33 Server at www.dragonsteelmods.com Port 80
This is great when you have unlimited development resources. In a world where developers are a scarce resource, it may be better to unify some projects and develop the remaining project at a better pace (even when it requires more thought).
Do those BSD versions actually share something? Different Linux distributions at least use the same kernel. From what I read in the article, it seems that he is discussing a lot of deficiencies in the NetBSD kernel. That would not be an issue when all BSD versions had the same kernel (or it would be an issue for all of them, but that is not how it is described).
Maybe there is no place for so many different kernels, that all need to be maintained by (groups of) people.
Notice that your simple script requires "some" temporary space. AT LEAST two times the amount of data you want to backup plus the size of one medium, but as written above even 3 times the amount of data. This makes it (and similar solutions that create intermediate files that are handed over to another program) a bit unpractical.
Problem with that is that after a while it will exit with "NO SPACE LEFT ON DEVICE", and when you have inserted the next disk there is no way to start from where it left off.
I don't get it. What is different between your solution that requires all kinds of rewrites, and using a journalling filesystem that is existing technology? Blocklevel update handling also exists already, as part of LVM. People use it to make backups of live data.
You clearly know nothing about tape. Or you only have used primitive programs like "tar".
Tape drives can fast-forward to the right point, when the proper index marks and directory information are available. Files can be removed in minutes from tapes that take hours to write sequentially.
Will you be able to restore files from a single DVD (not number 1)? Are there tools available that burn the parts on DVD on-the-fly while creating them, so you won't need to first generate 7 4GB files on the disk and then burn those?
This is not a problem in many raid systems. You can configure a RAID-1 array with one or more hot spares. Then on day one you can take drive 1 offline and replace it with hotspare 1. The RAID subsystem will copy your active disk to this spare. You can take the offlined disk out of the system (easiest when you have a hotpluggable drive mounting), and continue to work in RAID-1 mode with a safe offsite backup. The next day you can put a blank disk in the system, tell the system it is a spare, take another one offline, activate the spare, etc.
The advantage is that all your disks are mirror copies of another, so you can start a replacement system from them without any problem. (assuming that your replacement system is hardware compatible)
When using (S)ATA drives, it also is not that uneconomical. Look at the prices of 250GB drives and at the drive and media cost of an SDLT, and you know why.
Well, my Linux workstation behaves exactly like that (e.g. when I switch on the display at work after the system has been on during the night, making a backup etc, I have to sweep the mouse across all windows to force the apps to page-in all their lost memory). However, I do not see this as much in Windows. Maybe it depends on the version, we use Windows 2000 SP4. I think older versions did not even have this "the entire memory can be used as a disk cache" feature, so it wasn't there to bite you.
Unfortunately the Linux system has a hard time determining what are "infrequently accessed pages" and what are useful pages to keep in the disk cache.
This is most obvious when you are copying large amounts of data, e.g. during a backup. Say you have a 250GB disk and you copy it to another one. The system will continously try to keep the files you have read in disk cache (because you may read them again) and try to keep room for many dirty pages that still have to be written to the destination disk (because you may change them again before final writing). All of this "(because)" is never going to happen as everything is read once and written once and then no longer needed. But still, it will swap out running processes to make room for the above.
The net effect you see is that the source and swap disks are very busy, the destination disks sits idle long times until the kernel feels like flushing out some dirty buffers, and the other programs slow down to a crawl fighting for the swapspace.
It can be tuned with the "swappiness" variable but it remains a tough thing to control. It looks like Windows does a better job in this (not so hypothetical) case.
There should be some "file copy mode" (used during backups and other large tree copies) where it: - discards all disk USERDATA caches immediately after use (directory and other filesystem allocation data may be kept) - immediately writes out any written USERDATA to the destination disk, not having it populate the dirty pages until bdflush comes around to write them - keeps re-using the same small set of buffers to pump the data from source to destination, without stealing memory from others
Issue is of course: how could this mode be enabled. It could be a special systemcall, but who would call it and where? Personally, I would already be happy with a program like "nice" or "ionice" that would run a commandline in a special mode (e.g. with a very small buffer quota) to force such behaviour. But the world at large would of course be better serviced if this would happen automatically when lots of data are copied sequentially.
Slashdot Mirror
put European companies at a competitive disadvantage with every other company around the world who does have access to these new technologies.
It could also put them at a competitive advantage by using stable technologies while their foreign competitors play with new thingies.
By the time it gets introduced in Europe, the others have already found the first bugs and Microsoft may have fixed some of them.
Here in the Netherlands a static IP is standard for almost all DSL lines, and the dynamic addresses used on cable are almost always "fixed".
It is convenient for the provider and law enforcement, because they can easily track a subscriber's behaviour by his fixed address.
Also, it would not be possible to save addresses by using dynamic assignment, as these services are always-on. Usually the modem/router keeps the connection, even when the connected PC is off. So you would have to reserve an address for every subscriber anyway.
But IPv6 is far from being universally used. So, Vista will also support the current IPv4. The side effect, according to Mockapetris, is that a Vista PC will make two DNS requests, one for each IP version, instead of just one.
"It is going to try a DNS lookup for the IPv6 address and then a DNS lookup for the IPv4 address," Mockapetris said. "It just uses more DNS, and until we increase the supply, things are going to go slower."
Is this really true, or is it just FUD?
I mean, the DNS protocol supports putting an A and an AAAA question in the same request, doesn't it?
Does he mean that Vista doesn't do that? That would adversely affect Vista users as well, as they have to wait two RTTs before they get the A answer they probably want.
Because visitors don't know the difference between bugs in the software they use, and problems with the website they visit.
So when they have visited 10 sites that displayed ok, then visit your site and it is a mess, it must be a problem of your site.
Now, this may change with IE7. There will be a lot of sites that people who newly installed IE7 will visit, and will render incorrectly.
That may wake up some people.
It is probably because of this that Microsoft does not allow you to install IE6 and 7 alongside. You would be able to compare. That is unwanted, because it would show up problems that they prefer to deny.
But, you can still install Firefox and Opera and compare with those...
They cannot install IE7 if they aren't running XP SP2 or 2003 server.
It has always been possible to install another browser, but looking at the number of IE5 and 5.5 visits I still see, I think it will be 5 years before IE6 shows any sign of disappearing.
It is a step, but it is very questionable if it is a good thing to make a step when they are still so far of the target of rendering common CSS constructs that all other browsers render without problem.
(I do not mean the ACID2 test!)
Now we will have yet another browser to make special exceptions for, different from IE5 and IE6, and we still cannot feed IE7 the same CSS as Firefox, Opera or Konqueror.
That is a step, but is it the right direction? I don't know.
The site at work (http://www.uw.nl/) outputs standard html/css to everyone, and uses "IE conditional comments" to feed IE5 and IE6 specialized CSS items to work around their bugs.
A workaround sheet for IE7 has not yet been written, but it is very apparent (at least in beta3) that it is not up to the quality in standard CSS handling that the other browsers (Opera, Firefox, Konqueror) are. There are still positioning and stacking bugs.
I hope they fix them before release, but I'm afraid they won't. So this will introduce yet another class of broken browser workarounds: not as broken as IE6, but still broken.
In a country where people behave like that on a really large scale, it is tough to keep a "western civilization" in place.
Probably the country will fall down to an antisocial state where everyone only does what he deems good for himself and does not care what the effect on overall society is.
Not that this would be too amazing, but that is already what its government does in world politics. Together with the USA. And the effects are clearly visible.
Of course, as also explained in another posting about such a system operating in Belgium, the tariff structure includes a fixed fee additional to the per-kilogram tariff.
There are several cities here in the Netherlands that already have such a system in operation. Indeed, the cans are weighed and identified by number, and the total amount used to calculate the waste disposal bill.
It is not a bad thing, in principle. I would not know why a family with 3 babies that fills two cans a week would have to pay the same as an individual who does not even have a full can after 3 weeks. Differentiation of payment makes people more aware about what they do.
However, there are problems due to antisocial behaviour. What is frequently seen in those areas, is that people dump their waste in other people's cans when they have been put outside. Also, people take garbage from their home to their work place, and dump it there, especially when such a system is not yet implemented in the area where they work.
One would expect such problems to disappear over time (especially when the system is widely implemented).
But that is CPU-based filtering!
The "PureVideo" filtering is supposed to happen on the GPU.
This currently only works in Windows, not in Linux.
What resolution and refreshrate are you looking for?
Try one of these:
Modeline "1280x720p50" 74.25 1280 1724 1772 1980 720 723 725 750 +Hsync +Vsync
Modeline "1280x720p60" 74.11 1280 1392 1440 1648 720 723 725 750 +Hsync +Vsync
Modeline "1920x1080i50" 74.25 1920 2440 2456 2640 1080 1083 1085 1125 +Hsync+Vsync Interlace
Modeline "1920x1080i60" 74.25 1920 1976 2008 2200 1080 1083 1085 1125 +Hsync+Vsync Interlace
Note that the advanced video processing offered by PureVideo is supported only on Windows.
When you build your MythTV box based on Linux, your nVidia card actually is only an accellerated video card with some MPEG decoding support, but the de-interlacing and filtering is very primitive compared to what is offered in Windows and tested in the review.
So, the "Upconvert and de-interlace DVD content for display on HDTV quality screens and do so better than dedicated players with inexpensive NVIDIA or ATI cards" does not apply in the case of a Linux system.
(I think the situation is similar for ATI)
100 millibits per second??? That is one bit in 10 seconds, I presume.
Over 10 meters of distance. That smells like IP-over-Avian-carriers...
Service Temporarily Unavailable
The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later.
Apache/1.3.33 Server at www.dragonsteelmods.com Port 80
This is great when you have unlimited development resources.
In a world where developers are a scarce resource, it may be better to unify some projects and develop the remaining project at a better pace (even when it requires more thought).
Do those BSD versions actually share something?
Different Linux distributions at least use the same kernel. From what I read in the article, it seems that he is discussing a lot of deficiencies in the NetBSD kernel. That would not be an issue when all BSD versions had the same kernel (or it would be an issue for all of them, but that is not how it is described).
Maybe there is no place for so many different kernels, that all need to be maintained by (groups of) people.
Notice that your simple script requires "some" temporary space. AT LEAST two times the amount of data you want to backup plus the size of one medium, but as written above even 3 times the amount of data.
This makes it (and similar solutions that create intermediate files that are handed over to another program) a bit unpractical.
Problem with that is that after a while it will exit with "NO SPACE LEFT ON DEVICE", and when you have inserted the next disk there is no way to start from where it left off.
I don't get it. What is different between your solution that requires all kinds of rewrites, and using a journalling filesystem that is existing technology?
Blocklevel update handling also exists already, as part of LVM. People use it to make backups of live data.
You clearly know nothing about tape. Or you only have used primitive programs like "tar".
Tape drives can fast-forward to the right point, when the proper index marks and directory information are available.
Files can be removed in minutes from tapes that take hours to write sequentially.
Will you be able to restore files from a single DVD (not number 1)?
Are there tools available that burn the parts on DVD on-the-fly while creating them, so you won't need to first generate 7 4GB files on the disk and then burn those?
This is not a problem in many raid systems.
You can configure a RAID-1 array with one or more hot spares.
Then on day one you can take drive 1 offline and replace it with hotspare 1. The RAID subsystem will copy your active disk to this spare.
You can take the offlined disk out of the system (easiest when you have a hotpluggable drive mounting), and continue to work in RAID-1 mode with a safe offsite backup.
The next day you can put a blank disk in the system, tell the system it is a spare, take another one offline, activate the spare, etc.
The advantage is that all your disks are mirror copies of another, so you can start a replacement system from them without any problem.
(assuming that your replacement system is hardware compatible)
When using (S)ATA drives, it also is not that uneconomical. Look at the prices of 250GB drives and at the drive and media cost of an SDLT, and you know why.
Well, my Linux workstation behaves exactly like that (e.g. when I switch on the display at work after the system has been on during the night, making a backup etc, I have to sweep the mouse across all windows to force the apps to page-in all their lost memory).
However, I do not see this as much in Windows. Maybe it depends on the version, we use Windows 2000 SP4. I think older versions did not even have this "the entire memory can be used as a disk cache" feature, so it wasn't there to bite you.
Unfortunately the Linux system has a hard time determining what are "infrequently accessed pages" and what are useful pages to keep in the disk cache.
This is most obvious when you are copying large amounts of data, e.g. during a backup.
Say you have a 250GB disk and you copy it to another one. The system will continously try to keep the files you have read in disk cache (because you may read them again) and try to keep room for many dirty pages that still have to be written to the destination disk (because you may change them again before final writing).
All of this "(because)" is never going to happen as everything is read once and written once and then no longer needed.
But still, it will swap out running processes to make room for the above.
The net effect you see is that the source and swap disks are very busy, the destination disks sits idle long times until the kernel feels like flushing out some dirty buffers, and the other programs slow down to a crawl fighting for the swapspace.
It can be tuned with the "swappiness" variable but it remains a tough thing to control. It looks like Windows does a better job in this (not so hypothetical) case.
There should be some "file copy mode" (used during backups and other large tree copies) where it:
- discards all disk USERDATA caches immediately after use (directory and other filesystem allocation data may be kept)
- immediately writes out any written USERDATA to the destination disk, not having it populate the dirty pages until bdflush comes around to write them
- keeps re-using the same small set of buffers to pump the data from source to destination, without stealing memory from others
Issue is of course: how could this mode be enabled. It could be a special systemcall, but who would call it and where?
Personally, I would already be happy with a program like "nice" or "ionice" that would run a commandline in a special mode (e.g. with a very small buffer quota) to force such behaviour. But the world at large would of course be better serviced if this would happen automatically when lots of data are copied sequentially.