The best way to refer to Christianity-- or Islam, or Buddhism, or Zoroastrianism, or whatever-- is to call it a religion. The word "religion" has a very precise meaning, and no connotation of veracity or falsehood. That way, the whole question of truth or myth is conveniently sidestepped.
On the other hand, if you're the type of person who gets his rocks off by needlessly and pointlessly challenging other people's beliefs-- this sort of person appears to make up about 25% of the Slashdot posters-- then you probably don't see the point of sidestepping that question, so never mind.
Wrong. One terabyte (what's your fascination with InterCaps?) is 2^40 bytes. One petabyte is 2^50 bytes. The difference between the decimal expansions and the binary expansions become very significant at higher powers. For example, your statement that one terabyte is 10^12 bytes is wrong by 11%, or more than 99 billion bytes. That's quite a rounding error you've got there.
Have you done any estimates of total worldwide memory production? In a discussion of 64-bit filesystems, I once saw some figures indicating that in order to build an 18,000,000 terabyte filesystem, one would have to buy every hard drive produced in the world between 1993 and 2000. I don't know how accurate those numbers are, but it raises the question: how much RAM is manufactured in the world every year? How many years of production would you have to commandeer to do what you're talking about? Have you thought about those questions at all?
Remember that MTBF stands for mean time between failures. That's mean as in average. Half of the drives will fail in less than the mean time. The more drives you have, the greater the change that one of them will fail at any given time. If the MTBF is one million hours, and you have 1,000 drives, then unscientific but useful arithmetic says that the MTBF for your entire system drops to 1,000 hours, which is just over six weeks. As I said, the math isn't entirely sound, but it is very useful for estimating the overall reliability of a system.
Let us not forget that we're talking about terrestrial television broadcasts. These are line-of-sight transmissions. While it's obviously sensible to have common standards in major regions-- the US, Europe, the Pacific Rim-- having one standard for the whole planet is just not sufficiently important to justify the nightmarish amount of work that would go into arranging it.
If accounting can't go with Linux, it's going to be a big pain...
Pardon my ignorance, but why? Linux and Windows can run on the same hardware, so you won't need to buy any special computers. The admin overhead on one or a few Windows machines should be easy to absorb, especially if you're prepared to absorb the admin overhead of Linux. Why will it be a big pain to run accounting on Windows? And, of course, the next question is, "Will it be a bigger or a smaller pain to run accounting with less well known software?"
Peachtree is robust, Win98 works and nobody other than the slashdot and Apple crowds would find fault with your decision.
Actually, speaking as a charter member of the Apple crowd, I'd like to say that I support this recommendation completely. Accounting software is the closest most businesses come to something truly mission-critical. If the accounting software fails, nobody will die, but nobody will get paid or billed either, and that's about as bad as it gets for a business.
Pardon the coarse language, but don't fuck around with your accounting software. Use the best product for your particular needs, regardless of platform or-- within reason-- cost.
You know, not all broadcasts are terrestrial, they invented satellites in the meantime.
Which means... what? I don't see your point.
Multiple standards are a ploy by manufacturers and broadcasters to control market share....
Blah, blah, blah. You know, when you hear hoofbeats, you should think "horses," not "zebras." There are three different standards for analog TV broadcasts because the three standards were developed in different places and at different times. There was basically no reason to work together on a single worldwide standard because having one wouldn't have benefitted anybody at that time.
There are different standards for terrestrial HTDV broadcasts because, yeah, the standards were developed at different times and in different places. The Japanese had an 1125-line analog format in the 80's, I believe, but when the time came for the Americans to develop their own system, they wanted to improve on that standard. Thus, the ATSC formats transmitted over 8VSB.
Not everything is a conspiracy. Some things just evolve naturally.
Maybe, but I doubt economies of scale affect things at that level. The difference in cost between producing the first one of something and the thousandth one of something is huge; that's an example of an economy of scale at work. But Sony (for example) is already producing a gazillion units each of twelve different models of camcorder, so there's probably not much savings to be had by producing three gazillion units each of only four different models.
The article doesn't really imply that the standards are in any way undefined. Those details were set in stone years ago, between 1993 and 1996. The ATSC specs for broadcast digital television are implemented as SMPTE standards, and, as I said in my other post, they've been actively in use for over five years.
It's true that there's not anywhere near as much HDTV as their is regular TV, but that's to be expected. HDTV requires more bandwidth than regular TV (6 MHz per channel as opposed to 3 MHz for NTSC TV), and production and encoding equipment is expensive. But is it accurate to say that the rollout is "many years behind schedule?" Not really. That's because there isn't really a schedule, to speak of. The FCC has mandated that the chunk of spectrum currently used for NTSC TV transmission will be reallocated in 2007, at which time broadcasters are expected to turn off their analog transmission equipment. That's a long way off, so don't expect to see changes happen in days, or even months. The process takes years.
Does anyone even consider Digital TV relevant anymore?
Yup. Last night I watched Saving Private Ryan in 720p/60 with 5.1-channel Dolby Digital sound. Unedited. For free. On ABC.
Digital TV is a mystery to most people because the equipment is still pretty pricey. But that simply means the economies of scale haven't kicked in yet.
You know, there are something like 220 million televisions in the United States, owned by about 300 million people. It'd be fair to assume that all but a very few-- maybe two or three hundred thousand-- of those people are ignorant of the difference between NTSC, PAL, and SECAM. Some 299,800,000 people in the US alone don't even know that NTSC, PAL, and SECAM exist, or what they mean. For fifty years, we've lived in a world where Asia, Europe, and the US have all had different and incompatible television standards... and yet, somehow, the sun continues to rise each morning.
The vast-- and I truly mean vast-- majority of people will never know that the United States, Europe, Asia, France, and wherever-the-heck-else have incompatible television signal formats. For obvious reasons you can't receive Asian terrestrial broadcasts in Europe anyway, so for most people the issue simply never comes up. It's just not that big a deal.
You know, I do feel obligated to point out that ATSC-standard digital television signals using the 8VSB standard have been broadcast in the US full-time since 1998. You can receive 8VSB transmissions-- for free-- with nothing more than a decent pair of rabbit-ears, or, as in my case, with an inexpensive rooftop antenna. In my house, we watch at least six or eight hours of HDTV a week, and have been since the summer. Just last night, ABC broadcast Saving Private Ryan uncut and unedited in HDTV with, yeah, Dolby Digital sound.
You can also get HDTV via satellite-- HBO and Showtime have HDTV channels-- and in some markets via cable.
It's a gross exaggeration to say that the DTV system in the US is "bungled" or "broken."
The iDVD program was restricted in order to leverage sales of their internal DVD burners.
I think you typed "restricted" when you meant to type "created." Apple wrote iDVD-- or paid to have it written, which is effectively the same thing; I'm not sure if that program was developed internally or not-- so that people would have an easy way of creating DVDs out of their home movies and whatnot. They did this because many people wouldn't have bought Macs with Superdrives otherwise.
They didnt SAVE ANY MONEY by NOT DEVELOPING THAT PART OF THE PROGRAM, because the patch to allow external DVD burners took one person 3 days to write.
They saved three man-days of effort, and a huge amount of QA time. Apple would have had to test the entire iDVD function suite for every external drive they wanted to support. That's a serious investment of time and energy. So yeah, Apple did save money by not writing drivers for third-party drives. But that's not the point. The point is that Apple wrote iDVD specifically for people who have Macs with Superdrives. It's not for people who have Macs with other DVD burners, or people who have Macs without DVD burners, or people who have Commodore 64s with floppy drives.
See this is the problem with America, everyone thinks that the most profitable approach is "Right" or "Justified" and a-o-k.
See, this is the problem with people like you. (Far be it for me to generalize. Your problem cuts across boundaries of race, class, and nationality. Idiocy is color-blind.) You think business in inherently evil just because it creates inequities. Lots of uninformed people hold that opinion, but they're mistaken.
Yeah, and I'm totally off the deep end to think that corporate lobbies are buying our rights away, right? Even if it is an exagerated take on recent events, how dare you stifle that sort of thinking?
I think you typed "stifle" when you meant to type "ignore." And I think you typed "thinking" when you meant to type... I don't know. Something else. Because "thinking" has little to do with the things you're saying here.
Apple sold 100,000 copies of Mac OS X 10.2 during the first weekend it was available. At $129 each, that's $12,900,000 in gross revenue for just that one weekend. Red Hat, on the other hand, sold 4,802 copies of their software in the entire month of July. Even if you're generous and assume all of those units were the $150 "Professional" version-- they almost certainly weren't-- that comes to $720,300 in revenues for an entire month. If Red Hat were able to do that for an entire year, they wouldn't be able to match the gross sales of Mac OS X for one weekend.
Note that this doesn't count units of Mac OS X that were bundled with a new machine; this only counts actual retail packages sold. The number of units of Mac OS X 10.2 shipped is somewhat higher than this figure, making it difficult to estimate gross revenue from all sales of Mac OS X-- bundled and retail-- until we see the annual report.
(All figures from PC Data. If you have other figures that lead to a different conclusion, we can discuss them. Until then, STFU about sales figures about which you have no information.)
Only if the user is explicitly authorized to use the "cp" or "rm" commands. Sudo lets you set superuser privilege on a command-by-command basis. For example, "adduser" (or "useradd," depending on your OS) might be "sudoable" for everybody in the "admin" group, but not every command and not for ever user.
During the day, we rsync the development areas every 15 minutes.
I have found this to be a terrible idea. It sounded neat to me at first, too, but then I managed to grab an rsync of a directory in an inconsistent state. (I don't remember all the details, but it was something like one developer opening two files, saving the first file, the sync doing its thing, then the developer saving the other file. The files depended on each other, so the sync'd copy was useless. The code wouldn't compile without either reverting the first file, or modifying the second file.)
I'm not saying that it couldn't be made to work, but in that case it really bit us in the ass. It wasn't the end of the world, but it was extremely inconvenient.
I'm not entirely sure I think this is a good idea. You're talking about combining automatic saving-- saving after every keystroke, or every n seconds, or whatever-- with automatic version control. The net result is that your document would include one version every few seconds as long as you work on it. It'd be easy to accumulate a document with tens of thousands of versions. How would that be useful? You could, in theory, go back in time to any point to recover your work, but how would you know which point in time was the right one?
I'm not saying it wouldn't be neat; you would basically have enough data to do an instant-replay of the entire document creation process. But it doesn't sound too practical to me.
Yeah, the inode thing bothered me, too. See, the inode uniquely identifies a file, but it doesn't uniquely identify the file.
What I mean is this: when you use a traditional desktop program, like Photoshop for instance, the program takes measures to make sure your data is safe at all times. Saving a big file out of Photoshop can take a really long time on a slow machine, and during that time the chance that your machine or the program itself may crash is non-trivial.
So rather than overwriting your file during a save, Photoshop saves to a new, temporary file, leaving the original file intact. It does this invisibly. When the file is successfully saved and verified, only then does Photoshop rename your old file to a temporary name, rename the new file to the old name, then unlink the old file. The net result? A crash at any point in the process will leave you with at least one perfect copy of your file on disk. But across a save, your file's inode number changes.
This fact became evident to me on one occasion when I was trying to be too clever. I wanted to keep track of files by their inode numbers rather than their names, because names can change. When everything went to hell, I realized that programs like Photoshop and Word and, doubtlessly, others use this bit of cleverness that protects your data but that completely fubars the mappings of inode numbers to file names.
I can't swear to it, because I'm not in front of an OS X Server system right now, but I'm pretty sure OS X Server and vanilla OS X are identical as far as this article is concerned. OS X Server includes many more applications, and some "helper" applications to manage things like starting and stopping daemons, but the actually web server software itself is the same.
Again, I could be wrong about this. It's possible that I'm confused, in which case I apologize in advance.
So stipulated. But to be fair, let's describe Apple's work as being interesting in the "sell millions of copies and become, in less than a year, the world's largest supplier of UNIX-based operating systems" sense, while Hurd is interesting in the "Hurd who?" sense.
Slashdot Mirror
The best way to refer to Christianity-- or Islam, or Buddhism, or Zoroastrianism, or whatever-- is to call it a religion. The word "religion" has a very precise meaning, and no connotation of veracity or falsehood. That way, the whole question of truth or myth is conveniently sidestepped.
On the other hand, if you're the type of person who gets his rocks off by needlessly and pointlessly challenging other people's beliefs-- this sort of person appears to make up about 25% of the Slashdot posters-- then you probably don't see the point of sidestepping that question, so never mind.
Wrong. One terabyte (what's your fascination with InterCaps?) is 2^40 bytes. One petabyte is 2^50 bytes. The difference between the decimal expansions and the binary expansions become very significant at higher powers. For example, your statement that one terabyte is 10^12 bytes is wrong by 11%, or more than 99 billion bytes. That's quite a rounding error you've got there.
Have you done any estimates of total worldwide memory production? In a discussion of 64-bit filesystems, I once saw some figures indicating that in order to build an 18,000,000 terabyte filesystem, one would have to buy every hard drive produced in the world between 1993 and 2000. I don't know how accurate those numbers are, but it raises the question: how much RAM is manufactured in the world every year? How many years of production would you have to commandeer to do what you're talking about? Have you thought about those questions at all?
Remember that MTBF stands for mean time between failures. That's mean as in average. Half of the drives will fail in less than the mean time. The more drives you have, the greater the change that one of them will fail at any given time. If the MTBF is one million hours, and you have 1,000 drives, then unscientific but useful arithmetic says that the MTBF for your entire system drops to 1,000 hours, which is just over six weeks. As I said, the math isn't entirely sound, but it is very useful for estimating the overall reliability of a system.
Let us not forget that we're talking about terrestrial television broadcasts. These are line-of-sight transmissions. While it's obviously sensible to have common standards in major regions-- the US, Europe, the Pacific Rim-- having one standard for the whole planet is just not sufficiently important to justify the nightmarish amount of work that would go into arranging it.
If accounting can't go with Linux, it's going to be a big pain...
Pardon my ignorance, but why? Linux and Windows can run on the same hardware, so you won't need to buy any special computers. The admin overhead on one or a few Windows machines should be easy to absorb, especially if you're prepared to absorb the admin overhead of Linux. Why will it be a big pain to run accounting on Windows? And, of course, the next question is, "Will it be a bigger or a smaller pain to run accounting with less well known software?"
Peachtree is robust, Win98 works and nobody other than the slashdot and Apple crowds would find fault with your decision.
Actually, speaking as a charter member of the Apple crowd, I'd like to say that I support this recommendation completely. Accounting software is the closest most businesses come to something truly mission-critical. If the accounting software fails, nobody will die, but nobody will get paid or billed either, and that's about as bad as it gets for a business.
Pardon the coarse language, but don't fuck around with your accounting software. Use the best product for your particular needs, regardless of platform or-- within reason-- cost.
I think I speak for all of us here on Slashdot when I say... huh?
You know, not all broadcasts are terrestrial, they invented satellites in the meantime.
Which means... what? I don't see your point.
Multiple standards are a ploy by manufacturers and broadcasters to control market share....
Blah, blah, blah. You know, when you hear hoofbeats, you should think "horses," not "zebras." There are three different standards for analog TV broadcasts because the three standards were developed in different places and at different times. There was basically no reason to work together on a single worldwide standard because having one wouldn't have benefitted anybody at that time.
There are different standards for terrestrial HTDV broadcasts because, yeah, the standards were developed at different times and in different places. The Japanese had an 1125-line analog format in the 80's, I believe, but when the time came for the Americans to develop their own system, they wanted to improve on that standard. Thus, the ATSC formats transmitted over 8VSB.
Not everything is a conspiracy. Some things just evolve naturally.
Maybe, but I doubt economies of scale affect things at that level. The difference in cost between producing the first one of something and the thousandth one of something is huge; that's an example of an economy of scale at work. But Sony (for example) is already producing a gazillion units each of twelve different models of camcorder, so there's probably not much savings to be had by producing three gazillion units each of only four different models.
The article doesn't really imply that the standards are in any way undefined. Those details were set in stone years ago, between 1993 and 1996. The ATSC specs for broadcast digital television are implemented as SMPTE standards, and, as I said in my other post, they've been actively in use for over five years.
It's true that there's not anywhere near as much HDTV as their is regular TV, but that's to be expected. HDTV requires more bandwidth than regular TV (6 MHz per channel as opposed to 3 MHz for NTSC TV), and production and encoding equipment is expensive. But is it accurate to say that the rollout is "many years behind schedule?" Not really. That's because there isn't really a schedule, to speak of. The FCC has mandated that the chunk of spectrum currently used for NTSC TV transmission will be reallocated in 2007, at which time broadcasters are expected to turn off their analog transmission equipment. That's a long way off, so don't expect to see changes happen in days, or even months. The process takes years.
Like punching someone in the dark.
Does anyone even consider Digital TV relevant anymore?
Yup. Last night I watched Saving Private Ryan in 720p/60 with 5.1-channel Dolby Digital sound. Unedited. For free. On ABC.
Digital TV is a mystery to most people because the equipment is still pretty pricey. But that simply means the economies of scale haven't kicked in yet.
You know, there are something like 220 million televisions in the United States, owned by about 300 million people. It'd be fair to assume that all but a very few-- maybe two or three hundred thousand-- of those people are ignorant of the difference between NTSC, PAL, and SECAM. Some 299,800,000 people in the US alone don't even know that NTSC, PAL, and SECAM exist, or what they mean. For fifty years, we've lived in a world where Asia, Europe, and the US have all had different and incompatible television standards... and yet, somehow, the sun continues to rise each morning.
The vast-- and I truly mean vast-- majority of people will never know that the United States, Europe, Asia, France, and wherever-the-heck-else have incompatible television signal formats. For obvious reasons you can't receive Asian terrestrial broadcasts in Europe anyway, so for most people the issue simply never comes up. It's just not that big a deal.
You know, I do feel obligated to point out that ATSC-standard digital television signals using the 8VSB standard have been broadcast in the US full-time since 1998. You can receive 8VSB transmissions-- for free-- with nothing more than a decent pair of rabbit-ears, or, as in my case, with an inexpensive rooftop antenna. In my house, we watch at least six or eight hours of HDTV a week, and have been since the summer. Just last night, ABC broadcast Saving Private Ryan uncut and unedited in HDTV with, yeah, Dolby Digital sound.
You can also get HDTV via satellite-- HBO and Showtime have HDTV channels-- and in some markets via cable.
It's a gross exaggeration to say that the DTV system in the US is "bungled" or "broken."
The iDVD program was restricted in order to leverage sales of their internal DVD burners.
I think you typed "restricted" when you meant to type "created." Apple wrote iDVD-- or paid to have it written, which is effectively the same thing; I'm not sure if that program was developed internally or not-- so that people would have an easy way of creating DVDs out of their home movies and whatnot. They did this because many people wouldn't have bought Macs with Superdrives otherwise.
They didnt SAVE ANY MONEY by NOT DEVELOPING THAT PART OF THE PROGRAM, because the patch to allow external DVD burners took one person 3 days to write.
They saved three man-days of effort, and a huge amount of QA time. Apple would have had to test the entire iDVD function suite for every external drive they wanted to support. That's a serious investment of time and energy. So yeah, Apple did save money by not writing drivers for third-party drives. But that's not the point. The point is that Apple wrote iDVD specifically for people who have Macs with Superdrives. It's not for people who have Macs with other DVD burners, or people who have Macs without DVD burners, or people who have Commodore 64s with floppy drives.
See this is the problem with America, everyone thinks that the most profitable approach is "Right" or "Justified" and a-o-k.
See, this is the problem with people like you. (Far be it for me to generalize. Your problem cuts across boundaries of race, class, and nationality. Idiocy is color-blind.) You think business in inherently evil just because it creates inequities. Lots of uninformed people hold that opinion, but they're mistaken.
Yeah, and I'm totally off the deep end to think that corporate lobbies are buying our rights away, right? Even if it is an exagerated take on recent events, how dare you stifle that sort of thinking?
I think you typed "stifle" when you meant to type "ignore." And I think you typed "thinking" when you meant to type... I don't know. Something else. Because "thinking" has little to do with the things you're saying here.
Apple sold 100,000 copies of Mac OS X 10.2 during the first weekend it was available. At $129 each, that's $12,900,000 in gross revenue for just that one weekend. Red Hat, on the other hand, sold 4,802 copies of their software in the entire month of July. Even if you're generous and assume all of those units were the $150 "Professional" version-- they almost certainly weren't-- that comes to $720,300 in revenues for an entire month. If Red Hat were able to do that for an entire year, they wouldn't be able to match the gross sales of Mac OS X for one weekend.
Note that this doesn't count units of Mac OS X that were bundled with a new machine; this only counts actual retail packages sold. The number of units of Mac OS X 10.2 shipped is somewhat higher than this figure, making it difficult to estimate gross revenue from all sales of Mac OS X-- bundled and retail-- until we see the annual report.
(All figures from PC Data. If you have other figures that lead to a different conclusion, we can discuss them. Until then, STFU about sales figures about which you have no information.)
Only if the user is explicitly authorized to use the "cp" or "rm" commands. Sudo lets you set superuser privilege on a command-by-command basis. For example, "adduser" (or "useradd," depending on your OS) might be "sudoable" for everybody in the "admin" group, but not every command and not for ever user.
Google is your friend.
The BUILD of a kernel, for example, as reported by the uname user-land tool, says operating system is redhat-pc-gnu-linux.
;-)
Except it doesn't. I have four UNIX computers in front of me. Their "uname -s" strings are as follows:
Linux
Linux
FreeBSD
Darwin
I can't find a machine that reports redhat-pc-gnu-linux.
RMS is being taken the wrong way, I think, or could I be wrong and he realy is an asshole?
Bingo.
During the day, we rsync the development areas every 15 minutes.
I have found this to be a terrible idea. It sounded neat to me at first, too, but then I managed to grab an rsync of a directory in an inconsistent state. (I don't remember all the details, but it was something like one developer opening two files, saving the first file, the sync doing its thing, then the developer saving the other file. The files depended on each other, so the sync'd copy was useless. The code wouldn't compile without either reverting the first file, or modifying the second file.)
I'm not saying that it couldn't be made to work, but in that case it really bit us in the ass. It wasn't the end of the world, but it was extremely inconvenient.
I'm not entirely sure I think this is a good idea. You're talking about combining automatic saving-- saving after every keystroke, or every n seconds, or whatever-- with automatic version control. The net result is that your document would include one version every few seconds as long as you work on it. It'd be easy to accumulate a document with tens of thousands of versions. How would that be useful? You could, in theory, go back in time to any point to recover your work, but how would you know which point in time was the right one?
I'm not saying it wouldn't be neat; you would basically have enough data to do an instant-replay of the entire document creation process. But it doesn't sound too practical to me.
There's this great thing called "sudo" that you might consider checking out....
Yeah, the inode thing bothered me, too. See, the inode uniquely identifies a file, but it doesn't uniquely identify the file.
What I mean is this: when you use a traditional desktop program, like Photoshop for instance, the program takes measures to make sure your data is safe at all times. Saving a big file out of Photoshop can take a really long time on a slow machine, and during that time the chance that your machine or the program itself may crash is non-trivial.
So rather than overwriting your file during a save, Photoshop saves to a new, temporary file, leaving the original file intact. It does this invisibly. When the file is successfully saved and verified, only then does Photoshop rename your old file to a temporary name, rename the new file to the old name, then unlink the old file. The net result? A crash at any point in the process will leave you with at least one perfect copy of your file on disk. But across a save, your file's inode number changes.
This fact became evident to me on one occasion when I was trying to be too clever. I wanted to keep track of files by their inode numbers rather than their names, because names can change. When everything went to hell, I realized that programs like Photoshop and Word and, doubtlessly, others use this bit of cleverness that protects your data but that completely fubars the mappings of inode numbers to file names.
I can't swear to it, because I'm not in front of an OS X Server system right now, but I'm pretty sure OS X Server and vanilla OS X are identical as far as this article is concerned. OS X Server includes many more applications, and some "helper" applications to manage things like starting and stopping daemons, but the actually web server software itself is the same.
Again, I could be wrong about this. It's possible that I'm confused, in which case I apologize in advance.
So stipulated. But to be fair, let's describe Apple's work as being interesting in the "sell millions of copies and become, in less than a year, the world's largest supplier of UNIX-based operating systems" sense, while Hurd is interesting in the "Hurd who?" sense.