Interesting. We got these crappy new HP machines at work with integrated Intel graphics that aren't supported by Xorg X11 at all.
My team's requirements from IT were that these machines must support two monitors and must run Linux. Obviously without X11 the first fails. The integrated graphics only support a single monitor, so the second requirement also fails.
Adding an nVidia card, which is what I would do for two monitors, turns out to be impossible since it only has a PCI-e 1x slot and the only cards out there are ATI or Matrox. I spent several days trying to get the ATI driver to install, and it's still quite buggy. The Radeon X1300 framebuffer driver was unusable since it was so slow. The dual monitor support from this card also sucks, with one of the outputs having poor impedance matching causing all sorts of ringing on the image.
Cadmium? NiMH batteries usually don't contain cadmium. You're thinking of NiCd batteries, which are not the same, and not widely used any more except in some power tools due to the low internal impedance allowing for very high current output.
I haven't had much trouble with hills either with my 2006 Prius and I just got back from driving up to Lake Tahoe last weekend via CA-88. The only time I had a problem was last year when I drove over Sonora Pass, but that's 9400 feet and a very long 26% grade. I was limited to about 30-35MPH going uphill, but then you don't want to drive much over that anyway since there are very few straight stretches unless you have a sports car that handles it well. I also didn't want to push it too hard since the car was still pretty new at the time.
From the reports I have read, the battery pack on the Prius typically lasts far longer than 150K miles. That is just the minimum guaranteed (at least in CA). Also, if the battery pack does fail, it is recycled. NiMH batteries are not really all that nasty for the environment. The 2005 model is estimated to last 150K miles, though I have read numerous reports of them lasting longer. Replacement cost is reportedly around $2K, even though a new battery pack lists for close to $6K. My guess is that this is due to the fact that Toyota is able to rebuild defective batteries at their factory, thus avoiding the high cost of replacing the nickel. Reports of the classic Prius being used as taxis in New York show problems with batteries, even at 250K miles.
I have a couple of games I play on Linux and while the built-in nVidia CK804 sound on my motherboard is fine for general applications, I ran into problems when trying to run games. The big problem was that I could not get ALSA to support mixing sound from more than one application at a time. After doing some research, I found that the Audigy drivers supported this and went out and bought one of the cards. Generally it works well, but once in a while it goes into a mode where it makes an annoying click every few seconds where only a reboot will correct it (even reloading the kernel module and restarting ALSA doesn't stop it). My other computer had no built-in sound on the motherboard so I bought a cheap SB 128. It works, but periodically the sound dies and I have to restart ALSA (ES1370 driver). I wish ALSA could perform mixing in software, but at least with the 2.6.16 Linux kernel I couldn't get it working.
In general, the parts of the Linux kernel I have looked at are far better documented than Solaris. As for file systems, I took a quick glance through the EXT3, XFS, NFS and Reiserfs code in the Linux kernel and found that generally it is far better documented than ZFS, though Reiserfs and NFS seem worse than the others. Even where not well documented, the function names are often much more meaningful, making it much easier to pick up the code. Similarly, the networking code in Linux is generally very well documented compared to Solaris, where very little of the internals are documented. Trying to find out how to interface a streams protocol stack to UDP was rather difficult to figure out, with comments being generally non-existent anywhere I looked other than a couple of articles on the web which did not go into detail. Thankfully Sun was quite responsive to my questions and I was able to sort it out relatively quickly. If I were to rate the quality of documentation on what I have seen from 1 to 10 with both kernels, I would give Linux a 6 and Solaris a 3.
As for D-trace, sadly I have not been able to play with it since the system I work on has to run Solaris 8 due to using an older version of ClearCase for version control. It would certainly make life easier. (Note that the code I looked at was resent OpenSolaris code, though it's not that much different than the Solaris 8 code).
I can't comment much on BSD since my only experience with it is the old BSD TCP/IP stack used in VxWorks, which is a mess, though fairly well documented.
In my case, what I list as "good" documentation is relative. IMHO I should be able to glance at the beginning of a file to see what functionality it provides, and similarly at the beginning of each function. I also believe describing inputs, outputs and return values is important, since it makes it easier for other programmers to interface to an API. It also helps when I suddenly have to return to code I haven't worked on for several years and can immediately figure out why I did things the way I did. If something is rather complicated, I will try and describe the thought process as well so I don't come back later with WTF was I thinking. It also makes code reviews much easier since other people who are completely unfamiliar with the code can quickly follow what it does and why. It may take longer to write well documented code, but invariably saves time when debugging at some later point or when somebody else has to pick up the project.
As for the control/caps lock keys, I prefer where Sun put them and in fact use Sun keyboards on my Linux PCs. Java is well documented, in part because all of the API documentation comes from using javadoc. I have done the same thing with DOxygen, which is also an excellent tool.
I'm not saying everything about Solaris is bad. It generally is quite stable, but is missing a lot of drivers and features available on Linux. I often get a Linux crash every month or so on my new desktop. My older Linux desktop went 6 months without a reboot. I think I've only had a couple of crashes I can attribute to Solaris on that desktop (though every time we reboot our NFS servers we have to reboot all of our Solaris desktops).
I would love to see D-trace ported to Linux and some areas cleaned up. The ZFS filesystem sounds interesting with the snapshot and other features.
I am running KDE 3.5.6 on Solaris at work. It was painful to compile due to all the missing libraries, but I have most stuff working, including sound (video doesn't work well since the Ultrasparc system is way too slow). There are binary versions available for download for Solaris which are actively maintained.
The Solaris kernel needs a *lot* of work. It has some cool features like D-trace, but don't expect anybody to be able to jump in and write stuff for it since it is very poorly documented. I don't think most Sun engineers know what comments are for. Also, device driver support is poor at best. Opensolaris kernel development looks like it is moving very slowly, with little traffic on their mailing lists. ZFS also sounds like it needs work and may be a bit overhyped. The ZFS code is rather difficult to follow, again due to the lack of any comments or meaningful variable or function names.
With the Linux kernel, I can easily jump in and find what I'm looking for and can easily make changes. The code is fairly well organized and generally well documented.
The command-line tools often are missing many of the features one finds in the Linux tools, like decent help. Manytimes very useful features are just plain missing.
Sun's X Windows also leaves a lot to be desired. At least on Sparc, Xorg is not supported and Sun doesn't have proper working render support as far as I can tell, let alone font support. Also, any open source libraries that Sun does provide are often years out of date, and they don't make it easy to download various source packages. If you want just the kernel code, good luck. Everything is in one huge repository.
For servers, Solaris is generally rock solid. For a desktop system, it sucks badly. Linux generally works, unless you're stuck, like I am, with a POS ATI card on a POS HP desktop computer. The desktop computer only has PCI-e 1x and the only cards that will fit it and drive two monitors are ATI and Matrox. I'd throw it off a cliff if I could and replace it with a computer half the speed with an nVidia card any day. This is better than Solaris, though, which has much worse driver support than Linux.
KDE offers some of this, though nothing like dropping Blender into Gaim. In KDE, most applications are also components and can be easily embedded inside other applications. For example, Konqueror is not so much a web and file browser as a container. I.e. I click on a Word document and it opens it in the browser using kword, or I click on a MP3 and it can use Amarok, or a photo brings up my preferred photo viewer inside the browser.
As far as not requiring metadata for MP3s, Amarok already supports this (another KDE application). It calculates a sound fingerprint of the file and uses the Musicbrainz database to try and figure out the song. Not only that, but I can bring up lyrics, the CD cover and even Wikipedia entries on the band in question. It's pretty amazing.
As far as grouping photographs, I don't know anything open source that does that based on picture content.
-Aaron
I wonder if they took Linus's patches?
on
Gnome 2.18 Released
·
· Score: 0, Redundant
Anyone know if they accepted the patches Linus Torvalds gave for Gnome?
I picked up a FotoChute from Amazon for around $60 (it looks like they had it mismarked). It works, though I think better backup solutions are available for the same price range. I've had some issues with it being rather flakey and it requires an external flash reader if you don't plug the camera into it (and it's *very* picky about those). It basically behaves like a small external USB drive but can also back up your camera by plugging the USB cable from your camera into the drive.
For the price range, a Wolverine might be better, though I haven't used one of those. Do some research before buying.
Most people would recharge their vehicles at night, when there is often a significant excess capacity on the grid. Generators can't quickly ramp up and down the amount of power they generate so there's usually a lot of excess capacity at night. Many articles I have read on the web say that our current infrastructure could easily handle millions of electric or plug-in hybrid vehicles. One advantage is if designed right, the vehicles could help provide additional power to the grid during peak demand times while charging during periods of lower demand since currently there are not many ways to store the significant amount of electricity.
1. Private schools are able to pick and choose which students they accept. They do not have to deal with ESL students or special education. Special ed students can be very expensive to deal with, often requiring an aid for every 2-4 students. Public schools have to educate all children.
2. Private schools most likely have parents who take a much more active role than the average parent at a public school, after all, they push education more, hence are willing to spend the extra money.
One can get a very good education at a public school, though there will be a high percentage of students who don't care. It also depends on how much the student is willing to put into their education.
I went to a private school up until high school and most of the kids I know from then ended up as failures. I got a good education, but many other kids just screwed around and would have likely flunked in a public school. All the parents had to do was pay enough for their kid to make it to the next grade level.
I found public school was actually much more challenging if I chose the right class. In my initial history class, for example, I found the class boring and a waste of time. I transferred to a different teacher and the class was much more challenging than even the honors class.
I would have done even better in math had I gone to a public school rather than private school because in the private school they were not able to offer decent algebra classes (they basically handed us books and said go study, no explanation of anything and my mother was completely clueless about algebra and thus no help). If I had an actual teacher discussing algebra I would have finished high school calculus. As it was, I sailed through it in college.
Granted, some private schools were better than others, but where I lived there were only two schools available and neither one was very good. When I was first put into private school, though, the public schools were in the toilet because they had just lost a huge percentage of their funding due to Proposition 13, and half the kids in my class could not speak English.
It depends on which test is used. For example, the No Child Left Behind (NCLB) test is different for each state. Some states have more difficult tests than others. California has one of the hardest tests. By that standard, California does pretty poorly compared to states that have much easier tests. If, on the other hand, you look at state SAT scores, California doesn't do too badly and is fairly close to the national average, despite the fact that California has a huge percentage of students who's native language is not English. Also, some of the requirements of the NCLB testing are causing problems, like the fact that it includes special education students (i.e. the mentally disabled) as well as English Second Language (ESL).
California set a high bar for their tests, hence a much larger percentage of schools and students will be labeled as failing. I recall at one point the high school in Cupertino, CA, one of the best high schools in the nation, was labeled as failing because it didn't improve enough compared to the previous year, even though the school had one of the best test scores in the state.
No. Since it is a journaling filesystem, as long as the journal is intact, all pending writes will be completed the next time the system is powered up. Journaling filesystems typically write data out to the journal first, before writing it again to the proper location on the hard drive and updating all the inodes. The idea is that if power is lost that fsck will play back the journal and complete all pending operations. The problem with journals is that two write operations are required. Some filesystems can speed things up by storing the journal on a different hard drive than the data. By using a ram disk, this should significantly speed things up. This does not work for all filesystems, though. Some, like Reiser 4, do not have the traditional journal and hence only need one write. EXT3 also has the ability to write both data and metadata to the journal first for the best recovery, but it makes writes much slower. Most just write metadata only to the journal for performance so after recovery the filesystem is in a consistent state but more actual data will be lost.
There are RAM drives available that use DRAM, but due to the refresh circuitry and whatnot it takes a bit of power so the battery will only supply power to the RAM for a limited amount of time.
Also, if the flash were removable (i.e. SD card, compact flash) then it could be possible to move to another machine.
I would like to see a battery-backed RAM drive with FLASH as well. I think that for journaling filesystems it would be great for performance since the journal could be written into RAM and then later written to disk. The drawback of the RAM based drives I saw was that the battery is only good for a limited amount of time. The way to fix it is to provide less battery time but use that time to write the RAM out to FLASH when the power is cut. The advantage of combining RAM and FLASH is that RAM is very fast to write to and has an unlimited number of write cycles. Of course, I'd really like to see one of these new memory technologies come out that combines the best of DRAM and FLASH.
I've had no problems running Flash in a 64-bit version of Konqueror, though I think Konqueror takes a much saner approach to running plug-ins by using a separate process to run the plugin than the browser. That way if the plug-in goes haywire it doesn't take out the browser, or if I see a plug-in consuming 500MB of memory and 90% of the CPU I can just kill it and leave the browser running just fine. The other advantage is that the plug-in process can be 32-bit while the browser itself is 64 bit. I would think this is also good for security.
It wasn't divide by zero, but just plain floating-point division. After I learned of the bug I wrote a simple program to just brute force test division and it very quickly detected the problem. Certain values would cause errors. See http://en.wikipedia.org/wiki/Pentium_FDIV_bug for some more information. As far as I know, there never was a problem with the division by zero logic.
All it takes is for a field within the range of the bee hive to use a pesticide which can harm the bees and all the bees will suffer, since workers carry it back to the hive.
White LEDs are actually not very good for color in many cases. Most white LEDs actually only produce blue, and through the use of a scintillator, mostly yellow light. The yellow is the wavelength which affects the eye by stimulating both red and green since the wavelength is in-between. The result of this is that when using it to look at various objects, the reflected colors are often pretty bad. See http://en.wikipedia.org/wiki/LED#Disadvantages_of_ using_LEDs.
That's because you bought daylight bulbs. They are supposed to look like that since their color temperature is typically 6500K which approximates the color temperature of the sun. If your eyes have not adjusted to it and there are regular warm bulbs nearby then it will look bluish. Next time buy one with a lower color temperature, like 2800-3500K. See http://en.wikipedia.org/wiki/Color_temperature for more information.
According to TFA the new bulbs won't be available until 2010 and who knows how long for the 4x efficient bulbs (which is still a bit less than a decent CFL). -Aaron
That's not exactly fair since XP came out over 5 years ago. How about trying a Linux distribution from back then?
I have a server chugging away that is happily running SuSE 8.2 on a 333MHz PII with 512MB of RAM. It's been running as a file, SMTP (with spam filter), IMAP, DNS and web server and it hasn't been rebooted in over 3 1/2 years. I also use it for some remote access and even fire up Mozilla on it from time to time from a remote terminal. The only reason I see to upgrade it is so I can move to modern hardware and a more up to date Linux distro since a lot of current apps won't compile on such an old distro (2.4.20 based).
It is not just a childhood inconvenience. I didn't catch it as a young child despite my parents making sure to expose me to the neighbor kids when they had t, but later when I was in high school. I still have many scars to prove it. Not every child catches it, and for adults, chicken pox can be far more serious. Also, it can come back later in life in the form of shingles. If a pregnant woman gets it the baby could get birth defects, including brain damage, damage to the eye, neurological disorders among other things. Also, the older you are, the more severe it is. Even for kids, it's no fun and I wouldn't wish it on anyone. Since I was older when I got it, I got hit very hard by it.
Just saying that you'll keep your kid inside when they get it doesn't help, since they're contagious two days before any symptoms show.
Here is some more information on that "inconvenience". My grandfather suffers from shingles caused by chickenpox and it's no fun.
Besides, if your kid gets it, is it fair for your kid to pass it on to everyone else who hasn't had it? How about an adult who never got it?
I have been using Bibble Pro on Linux and have been very happy with it. It has great workflow support as well as being multithreaded and able to take advantage of multiple cores. It does all its processing in 16bits per color and has excellent raw support for Canon, Nikon, Pentax and others. The workflow support works quite well for me. There are also numerous plug-ins available and they provide the API to 3rd party developers.
They also are fairly good about releasing new versions with new features and support for the latest cameras and lenses. Usually they release a new version every 2-3 months.
It runs on Linux, Mac OSX and Windows, which makes sense since it was based on the cross-platform QT library.
The raw converter in Bibble is very good, being based on dcraw. Similarly, it has many other plug-ins like a single click lens distortion correction based on Panarama, Noise Ninja and many more, all being very easy to work with. Of course it has all the tools for manipulating color, white balance, contrast, curves, shadow and highlight recovery, sharpening and many other features. The evaluation version is free to download.
As far as features, the only feature that I know of that does not work on Linux at this time is teathered shooting. All of the other features now work. Earlier versions did have issues with some features not working on Linux, but they have addressed that.
I did have issues with printing a while back, but it looks like it has been addressed.
Slashdot Mirror
Interesting. We got these crappy new HP machines at work with integrated Intel graphics that aren't supported by Xorg X11 at all.
My team's requirements from IT were that these machines must support two monitors and must run Linux. Obviously without X11 the first fails. The integrated graphics only support a single monitor, so the second requirement also fails.
Adding an nVidia card, which is what I would do for two monitors, turns out to be impossible since it only has a PCI-e 1x slot and the only cards out there are ATI or Matrox. I spent several days trying to get the ATI driver to install, and it's still quite buggy. The Radeon X1300 framebuffer driver was unusable since it was so slow. The dual monitor support from this card also sucks, with one of the outputs having poor impedance matching causing all sorts of ringing on the image.
Cadmium? NiMH batteries usually don't contain cadmium. You're thinking of NiCd batteries, which are not the same, and not widely used any more except in some power tools due to the low internal impedance allowing for very high current output.
I haven't had much trouble with hills either with my 2006 Prius and I just got back from driving up to Lake Tahoe last weekend via CA-88. The only time I had a problem was last year when I drove over Sonora Pass, but that's 9400 feet and a very long 26% grade. I was limited to about 30-35MPH going uphill, but then you don't want to drive much over that anyway since there are very few straight stretches unless you have a sports car that handles it well. I also didn't want to push it too hard since the car was still pretty new at the time.
From the reports I have read, the battery pack on the Prius typically lasts far longer than 150K miles. That is just the minimum guaranteed (at least in CA). Also, if the battery pack does fail, it is recycled. NiMH batteries are not really all that nasty for the environment. The 2005 model is estimated to last 150K miles, though I have read numerous reports of them lasting longer. Replacement cost is reportedly around $2K, even though a new battery pack lists for close to $6K. My guess is that this is due to the fact that Toyota is able to rebuild defective batteries at their factory, thus avoiding the high cost of replacing the nickel. Reports of the classic Prius being used as taxis in New York show problems with batteries, even at 250K miles.
I have a couple of games I play on Linux and while the built-in nVidia CK804 sound on my motherboard is fine for general applications, I ran into problems when trying to run games. The big problem was that I could not get ALSA to support mixing sound from more than one application at a time. After doing some research, I found that the Audigy drivers supported this and went out and bought one of the cards. Generally it works well, but once in a while it goes into a mode where it makes an annoying click every few seconds where only a reboot will correct it (even reloading the kernel module and restarting ALSA doesn't stop it). My other computer had no built-in sound on the motherboard so I bought a cheap SB 128. It works, but periodically the sound dies and I have to restart ALSA (ES1370 driver). I wish ALSA could perform mixing in software, but at least with the 2.6.16 Linux kernel I couldn't get it working.
In general, the parts of the Linux kernel I have looked at are far better documented than Solaris. As for file systems, I took a quick glance through the EXT3, XFS, NFS and Reiserfs code in the Linux kernel and found that generally it is far better documented than ZFS, though Reiserfs and NFS seem worse than the others. Even where not well documented, the function names are often much more meaningful, making it much easier to pick up the code. Similarly, the networking code in Linux is generally very well documented compared to Solaris, where very little of the internals are documented. Trying to find out how to interface a streams protocol stack to UDP was rather difficult to figure out, with comments being generally non-existent anywhere I looked other than a couple of articles on the web which did not go into detail. Thankfully Sun was quite responsive to my questions and I was able to sort it out relatively quickly. If I were to rate the quality of documentation on what I have seen from 1 to 10 with both kernels, I would give Linux a 6 and Solaris a 3.
As for D-trace, sadly I have not been able to play with it since the system I work on has to run Solaris 8 due to using an older version of ClearCase for version control. It would certainly make life easier. (Note that the code I looked at was resent OpenSolaris code, though it's not that much different than the Solaris 8 code).
I can't comment much on BSD since my only experience with it is the old BSD TCP/IP stack used in VxWorks, which is a mess, though fairly well documented.
In my case, what I list as "good" documentation is relative. IMHO I should be able to glance at the beginning of a file to see what functionality it provides, and similarly at the beginning of each function. I also believe describing inputs, outputs and return values is important, since it makes it easier for other programmers to interface to an API. It also helps when I suddenly have to return to code I haven't worked on for several years and can immediately figure out why I did things the way I did. If something is rather complicated, I will try and describe the thought process as well so I don't come back later with WTF was I thinking. It also makes code reviews much easier since other people who are completely unfamiliar with the code can quickly follow what it does and why. It may take longer to write well documented code, but invariably saves time when debugging at some later point or when somebody else has to pick up the project.
As for the control/caps lock keys, I prefer where Sun put them and in fact use Sun keyboards on my Linux PCs. Java is well documented, in part because all of the API documentation comes from using javadoc. I have done the same thing with DOxygen, which is also an excellent tool.
I'm not saying everything about Solaris is bad. It generally is quite stable, but is missing a lot of drivers and features available on Linux. I often get a Linux crash every month or so on my new desktop. My older Linux desktop went 6 months without a reboot. I think I've only had a couple of crashes I can attribute to Solaris on that desktop (though every time we reboot our NFS servers we have to reboot all of our Solaris desktops).
I would love to see D-trace ported to Linux and some areas cleaned up. The ZFS filesystem sounds interesting with the snapshot and other features.
I am running KDE 3.5.6 on Solaris at work. It was painful to compile due to all the missing libraries, but I have most stuff working, including sound (video doesn't work well since the Ultrasparc system is way too slow). There are binary versions available for download for Solaris which are actively maintained.
The Solaris kernel needs a *lot* of work. It has some cool features like D-trace, but don't expect anybody to be able to jump in and write stuff for it since it is very poorly documented. I don't think most Sun engineers know what comments are for. Also, device driver support is poor at best. Opensolaris kernel development looks like it is moving very slowly, with little traffic on their mailing lists. ZFS also sounds like it needs work and may be a bit overhyped. The ZFS code is rather difficult to follow, again due to the lack of any comments or meaningful variable or function names.
With the Linux kernel, I can easily jump in and find what I'm looking for and can easily make changes. The code is fairly well organized and generally well documented.
The command-line tools often are missing many of the features one finds in the Linux tools, like decent help. Manytimes very useful features are just plain missing.
Sun's X Windows also leaves a lot to be desired. At least on Sparc, Xorg is not supported and Sun doesn't have proper working render support as far as I can tell, let alone font support. Also, any open source libraries that Sun does provide are often years out of date, and they don't make it easy to download various source packages. If you want just the kernel code, good luck. Everything is in one huge repository.
For servers, Solaris is generally rock solid. For a desktop system, it sucks badly. Linux generally works, unless you're stuck, like I am, with a POS ATI card on a POS HP desktop computer. The desktop computer only has PCI-e 1x and the only cards that will fit it and drive two monitors are ATI and Matrox. I'd throw it off a cliff if I could and replace it with a computer half the speed with an nVidia card any day. This is better than Solaris, though, which has much worse driver support than Linux.
KDE offers some of this, though nothing like dropping Blender into Gaim. In KDE, most applications are also components and can be easily embedded inside other applications. For example, Konqueror is not so much a web and file browser as a container. I.e. I click on a Word document and it opens it in the browser using kword, or I click on a MP3 and it can use Amarok, or a photo brings up my preferred photo viewer inside the browser.
As far as not requiring metadata for MP3s, Amarok already supports this (another KDE application). It calculates a sound fingerprint of the file and uses the Musicbrainz database to try and figure out the song. Not only that, but I can bring up lyrics, the CD cover and even Wikipedia entries on the band in question. It's pretty amazing.
As far as grouping photographs, I don't know anything open source that does that based on picture content.
-Aaron
Anyone know if they accepted the patches Linus Torvalds gave for Gnome?
I picked up a FotoChute from Amazon for around $60 (it looks like they had it mismarked). It works, though I think better backup solutions are available for the same price range. I've had some issues with it being rather flakey and it requires an external flash reader if you don't plug the camera into it (and it's *very* picky about those). It basically behaves like a small external USB drive but can also back up your camera by plugging the USB cable from your camera into the drive.
For the price range, a Wolverine might be better, though I haven't used one of those. Do some research before buying.
Most people would recharge their vehicles at night, when there is often a significant excess capacity on the grid. Generators can't quickly ramp up and down the amount of power they generate so there's usually a lot of excess capacity at night. Many articles I have read on the web say that our current infrastructure could easily handle millions of electric or plug-in hybrid vehicles. One advantage is if designed right, the vehicles could help provide additional power to the grid during peak demand times while charging during periods of lower demand since currently there are not many ways to store the significant amount of electricity.
This is true, but:
1. Private schools are able to pick and choose which students they accept. They do not have to deal with ESL students or special education. Special ed students can be very expensive to deal with, often requiring an aid for every 2-4 students. Public schools have to educate all children.
2. Private schools most likely have parents who take a much more active role than the average parent at a public school, after all, they push education more, hence are willing to spend the extra money.
One can get a very good education at a public school, though there will be a high percentage of students who don't care. It also depends on how much the student is willing to put into their education.
I went to a private school up until high school and most of the kids I know from then ended up as failures. I got a good education, but many other kids just screwed around and would have likely flunked in a public school. All the parents had to do was pay enough for their kid to make it to the next grade level.
I found public school was actually much more challenging if I chose the right class. In my initial history class, for example, I found the class boring and a waste of time. I transferred to a different teacher and the class was much more challenging than even the honors class.
I would have done even better in math had I gone to a public school rather than private school because in the private school they were not able to offer decent algebra classes (they basically handed us books and said go study, no explanation of anything and my mother was completely clueless about algebra and thus no help). If I had an actual teacher discussing algebra I would have finished high school calculus. As it was, I sailed through it in college.
Granted, some private schools were better than others, but where I lived there were only two schools available and neither one was very good. When I was first put into private school, though, the public schools were in the toilet because they had just lost a huge percentage of their funding due to Proposition 13, and half the kids in my class could not speak English.
It depends on which test is used. For example, the No Child Left Behind (NCLB) test is different for each state. Some states have more difficult tests than others. California has one of the hardest tests. By that standard, California does pretty poorly compared to states that have much easier tests. If, on the other hand, you look at state SAT scores, California doesn't do too badly and is fairly close to the national average, despite the fact that California has a huge percentage of students who's native language is not English. Also, some of the requirements of the NCLB testing are causing problems, like the fact that it includes special education students (i.e. the mentally disabled) as well as English Second Language (ESL).
s how.htm?doc_id=373044
California set a high bar for their tests, hence a much larger percentage of schools and students will be labeled as failing. I recall at one point the high school in Cupertino, CA, one of the best high schools in the nation, was labeled as failing because it didn't improve enough compared to the previous year, even though the school had one of the best test scores in the state.
For example, reading on fairtest.org it claims that Ohio's 4th grade test is the equivalent to the 8th grade test in Texas. See http://www.educationsector.org/analysis/analysis_
No. Since it is a journaling filesystem, as long as the journal is intact, all pending writes will be completed the next time the system is powered up. Journaling filesystems typically write data out to the journal first, before writing it again to the proper location on the hard drive and updating all the inodes. The idea is that if power is lost that fsck will play back the journal and complete all pending operations. The problem with journals is that two write operations are required. Some filesystems can speed things up by storing the journal on a different hard drive than the data. By using a ram disk, this should significantly speed things up. This does not work for all filesystems, though. Some, like Reiser 4, do not have the traditional journal and hence only need one write. EXT3 also has the ability to write both data and metadata to the journal first for the best recovery, but it makes writes much slower. Most just write metadata only to the journal for performance so after recovery the filesystem is in a consistent state but more actual data will be lost.
There are RAM drives available that use DRAM, but due to the refresh circuitry and whatnot it takes a bit of power so the battery will only supply power to the RAM for a limited amount of time.
Also, if the flash were removable (i.e. SD card, compact flash) then it could be possible to move to another machine.
I would like to see a battery-backed RAM drive with FLASH as well. I think that for journaling filesystems it would be great for performance since the journal could be written into RAM and then later written to disk. The drawback of the RAM based drives I saw was that the battery is only good for a limited amount of time. The way to fix it is to provide less battery time but use that time to write the RAM out to FLASH when the power is cut. The advantage of combining RAM and FLASH is that RAM is very fast to write to and has an unlimited number of write cycles. Of course, I'd really like to see one of these new memory technologies come out that combines the best of DRAM and FLASH.
I've had no problems running Flash in a 64-bit version of Konqueror, though I think Konqueror takes a much saner approach to running plug-ins by using a separate process to run the plugin than the browser. That way if the plug-in goes haywire it doesn't take out the browser, or if I see a plug-in consuming 500MB of memory and 90% of the CPU I can just kill it and leave the browser running just fine. The other advantage is that the plug-in process can be 32-bit while the browser itself is 64 bit. I would think this is also good for security.
I wish Firefox would adopt this feature.
It wasn't divide by zero, but just plain floating-point division. After I learned of the bug I wrote a simple program to just brute force test division and it very quickly detected the problem. Certain values would cause errors. See http://en.wikipedia.org/wiki/Pentium_FDIV_bug for some more information. As far as I know, there never was a problem with the division by zero logic.
I came across an interesting article at http://www.everythingabout.net/articles/biology/an imals/arthropods/insects/bees/aa/vanishing_part1.s html which discusses pesticide use and bees.
All it takes is for a field within the range of the bee hive to use a pesticide which can harm the bees and all the bees will suffer, since workers carry it back to the hive.
White LEDs are actually not very good for color in many cases. Most white LEDs actually only produce blue, and through the use of a scintillator, mostly yellow light. The yellow is the wavelength which affects the eye by stimulating both red and green since the wavelength is in-between. The result of this is that when using it to look at various objects, the reflected colors are often pretty bad. See http://en.wikipedia.org/wiki/LED#Disadvantages_of_ using_LEDs.
-Aaron
That's because you bought daylight bulbs. They are supposed to look like that since their color temperature is typically 6500K which approximates the color temperature of the sun. If your eyes have not adjusted to it and there are regular warm bulbs nearby then it will look bluish. Next time buy one with a lower color temperature, like 2800-3500K. See http://en.wikipedia.org/wiki/Color_temperature for more information.
-Aaron
According to TFA the new bulbs won't be available until 2010 and who knows how long for the 4x efficient bulbs (which is still a bit less than a decent CFL).
-Aaron
That's not exactly fair since XP came out over 5 years ago. How about trying a Linux distribution from back then?
I have a server chugging away that is happily running SuSE 8.2 on a 333MHz PII with 512MB of RAM. It's been running as a file, SMTP (with spam filter), IMAP, DNS and web server and it hasn't been rebooted in over 3 1/2 years. I also use it for some remote access and even fire up Mozilla on it from time to time from a remote terminal. The only reason I see to upgrade it is so I can move to modern hardware and a more up to date Linux distro since a lot of current apps won't compile on such an old distro (2.4.20 based).
-Aaron
It is not just a childhood inconvenience. I didn't catch it as a young child despite my parents making sure to expose me to the neighbor kids when they had t, but later when I was in high school. I still have many scars to prove it. Not every child catches it, and for adults, chicken pox can be far more serious. Also, it can come back later in life in the form of shingles. If a pregnant woman gets it the baby could get birth defects, including brain damage, damage to the eye, neurological disorders among other things. Also, the older you are, the more severe it is. Even for kids, it's no fun and I wouldn't wish it on anyone. Since I was older when I got it, I got hit very hard by it.
Just saying that you'll keep your kid inside when they get it doesn't help, since they're contagious two days before any symptoms show.
Here is some more information on that "inconvenience". My grandfather suffers from shingles caused by chickenpox and it's no fun.
Besides, if your kid gets it, is it fair for your kid to pass it on to everyone else who hasn't had it? How about an adult who never got it?
I have been using Bibble Pro on Linux and have been very happy with it. It has great workflow support as well as being multithreaded and able to take advantage of multiple cores. It does all its processing in 16bits per color and has excellent raw support for Canon, Nikon, Pentax and others. The workflow support works quite well for me. There are also numerous plug-ins available and they provide the API to 3rd party developers.
They also are fairly good about releasing new versions with new features and support for the latest cameras and lenses. Usually they release a new version every 2-3 months.
It runs on Linux, Mac OSX and Windows, which makes sense since it was based on the cross-platform QT library.
The raw converter in Bibble is very good, being based on dcraw. Similarly, it has many other plug-ins like a single click lens distortion correction based on Panarama, Noise Ninja and many more, all being very easy to work with. Of course it has all the tools for manipulating color, white balance, contrast, curves, shadow and highlight recovery, sharpening and many other features. The evaluation version is free to download.
As far as features, the only feature that I know of that does not work on Linux at this time is teathered shooting. All of the other features now work. Earlier versions did have issues with some features not working on Linux, but they have addressed that.
I did have issues with printing a while back, but it looks like it has been addressed.