> Our industrial UPS (which is orders of magnitude more reliable than any APC > product ever made) recently exploded, burnt, and shorted out
Well, I admit, that if my building burns down, I just might have to suffer through the inconvenience of running fsck on my filesystems for ten minutes or so, while you with your journalled fs will be able to restart in less than a minute. Call me a fool, but I think I just might be willing live with such a horrible risk.
> It's also painfully obvious that you've never worked as a sysadmin before.
Of course not, and that's my point: the world does not revolve around servers. Some of us use Linux on the desktop, and if the kernel developers don't get their head out of their asses and realize that, the year of "Linux on the desktop" will never come.
> Wake me up after you've been waiting for fsck to finish on your 1TB > drive and it's been running for the last 72 hours.
If you are a sysadmin and it takes you 72 hours to fsck a 1TB drive, you should be fired.
> Yeah, because systems never kernel panic, or crash for any other reason than power outages.
You must be running Windows. Serves you right. If you were running Linux, you wouldn't be having these problems. I haven't seen a kernel panic since 1996, and you wouldn't either if you stopped running the daily build from Linus's git tree on crappy hardware from the thrift shop.
> Just search for benchmarks, something like reiserfs beats ext2 by huge margins
You mean like these ones where ext2 beats reiserfs in most cases and is at least as fast in the others?
> I hope you're joking. ext2 is nice and simple, but it's neither fast not reliable. > It uses a linear search to find directory entries, which means it's very slow on > large directories, like Maildir mailboxes.
Believe it or not, the world does not revolve around huge mail servers. Some of us actually run Linux on a desktop, and so don't really care about how well an fs handles a million maildir mailboxes. Latency is the most important criteria, and reiserfs is just too complicated to deliver it, as well as being a largely fringe fs. Especially now with Hans gone, it would become even more fringe.
> It doesn't do tail packing which means it wastes space and is slower with small files.
Yup, I'd like to have efficient small file handling. But really, it is better to avoid having many small files in the first place. Use compressed archives to store such things; it's quite a bit more efficient, and does not require exotic file systems which most normal people (i.e. your customers) will not use.
> It's not reliable because without a journal it needs a fsck after a bad shutdown
I used to do that, and then I got a UPS instead and switched back to pure ext2. The performance hit from journalling is simply too high to tolerate. A decent UPS (pretty much anything made by APC) will prevent the crashes in the first place, solving the problem completely and without any unnecessary overhead. With UPS prices being as low as they are, there is no excuse for not having one, so I think that journalling will become obsolete in some near future.
> Call me old fashioned but I should be able to grab everything under C:\Program Files\My App > and move that to another machine and run it
Old fashioned? Hell no! You're a newbie who never uses the command line or cares about how much space your applications are using. Ok, so that's what pretty much all Windows users are, but users are not the ones who should be setting the standards here. Application directories are indeed convenient in your particular scenario. Sure, manipulating a single directory is easy, but nobody should do that! A user should not even know that there is a "Program Files" directory, or have any access to it whatsoever. It is the administrator's job to take care of program installation. For home users, Windows provides the uninstaller tool under the control panel, which is much simpler than looking for the application directory, which is usually buried under the company directory (what vain idiot came up with that?), since the actual application name is displayed in the uninstall list. Copying an app to another machine is not a good idea either since configuration will likely be different and will need to be updated. That's why we have installers in the first place.
From the developer's perspective, application directories are much worse. They stop you from sharing code. All the libraries you use, you must package with the app, and while they are running, they are not shared with any other app, leading to a considerable increase in your program's memory footprint, and consequent complaints by your users about how your worthless app takes five minutes to just start. Critical updates to important libraries can not be performed easily. Say someone finds a vulnerability in a dll and issues a fix. With application directories, the updater would have to search for every instance of that particular dll and replace each and every one!
Then there is the problem of not being able to launch the programs from the command line. Windows users don't do that, but any "old-fashioned" power user can't live without it. It is, in fact, one of the things I really hate about Windows, that all the apps live under their own little app dirs, and are not in the PATH.
As for the "convenience" of manipulating directories, a proper package manager is a much better solution. Not only does it allow you to recreate the package, copy it to another machine, and reinstall like you wanted, it will also track all dependencies and install them automatically for you. If you think about it, there really isn't much difference whether the filesystem keeps track of where the package is (with app dir), or the package manager keeps track of where all the package parts are (with the install log). The power of the merged hierarchy amply compensates for preventing you from mucking directly with the app dir.
> Lets say you have 10,000 users using your program under Windows 95. > You store their files under Program Files, Program name, User
In other words, you want to be inconsiderate and make me hunt down my data files under some weird directory name under Program Files, which, by the way, is hidden by default until you tell explorer to show system files or something. There is absolutely no excuse for writing data files in the program directory. Windows 95 supports home directories. Use them.
> then XP hits. It recomends that you move those data files to My Documents. > Well Now you have 50,000 users. Do you make them move them?
Hell no! That's also inconsiderate. You have the setup program move them and tell the user what it did. No ifs ands or buts. Sure, you'll annoy your 50000 users, but that's just what you deserve for not following standards.
> Now comes Vista and it throws a fit if you store a data file under Program files > Which if they are none executable I don't see what the problem is.
The problem is that users should never write to Program Files. Period. The permissions should be set to not allow it. This is a very very good idea because it prevents viruses from mucking up your installed programs. And, of course, it prevents the user from accidentally deleting them. Mac users, in particular, tend to think that an app can be uninstalled by deleting its directory.
> So you simply tell people to turn off the UAC.
Heck no! There is no way I'm compromising my computer's security just to cover for your bad design. I'll either stop using your program, or implement an inconvenient workaround (which normal users would not be able to do), consisting of manually going to Program Files/Your App and setting permissions on the directories you are foolishly trying to write to. Old games are VERY bad about this. Fortunately, setting permissions on the saved game subdir usually works.
It would be a much better idea to force every programmer to run under a non-Administrator account (and no Administrators or even Power Users group membership either!) Anyone who complains is obviously writing bad code, since there is absolutely no friggin' reason that a regular application should require administrative privileges. Whatever you set during setup is IT! And, for God's sake, learn to open registry keys in read-only mode!
The probability of an event that has already happened is always equal to 1. Predictions are uncertain only about the future, and, of course, for past events about which there is not enough information.
Get your stinkin' 3D off my cell phone! It is yet another ridiculous piece of bloatware because of which it takes ages to turn the phone on or off. Do these developers realize that most people don't want the stuff they already implemented? How about focusing on battery life, startup time, and voice clarity instead, huh? You know, the things that actually matter in a phone, instead of making 3D games, which are cool, but most certainly do not belong on a handheld device.
> No, half of the people in this country are at or below median intelligence.
IQ is based on the assumption that intelligence distribution is normal. In the normal distribution, the average is the same as the median, and yes, half the people in the world have below-average intelligence. That does not mean that half the people in this country have below average intelligence, because IQ averages vary by country. If you look at the table of IQs by country (which are averages, BTW), you'll see that the US has the average IQ of 98, meaning that slightly more than half the people in this country have below-average intelligence: 55.3%. By comparison, in Equatorial Guinea, where the average IQ is 59, 99.7% of the population has below average intelligence. You can get the numbers by calculating the error function with this calculator, 59 is 41/15(SD)=2.73(3) standard deviations.
> what can we do with this new knowledge other than escape the bubble to realize our true freedom?
We can finally say that beer is the true meaning of the universe. This finding confirms what beer drinkers around the world have suspected for years: our universe is just a bubble in a giant glass of beer! In the beginning, the beer was flat. Then suddenly the bottle was opened, and the lowered pressure lowered carbon dioxide's solubility and enabled creation of bubbles. As the primordial beer gas accumulated in our bubble, gravity appeared (the surrounding universe is made of light beer, which does not bend space as much as the regular beer) and caused the carbon dioxide to coalesce into stars and planets, and eventually into people. Our bubble is expanding now, and floating upward in the glass. Eventually it will reach the top and become a part of the giant cosmic head, at which point we shall all be judged for our actions and be doomed to either sink back in the glass, or to fly up into the cosmos with the angels. Yup, dude, this is some heavy stuff! But don't worry, the more beer you drink, the better you understand it!
Hey, it's a good name! It's about time we stopped focusing on creating glistening new tools and started thinking about actually using them for something. But, of course, being an average Slashdot reader, you probably don't know how.
The classical example showing the error of uncritical reasoning here is the old fable about the height of the Emperor of China. Supposing that each person in China surely knows the height of the Emperor to an accuracy of at least 1 meter; if there are N=1e9 inhabitants, then it seems that we could determine his height to an accuracy at least as good as 1/sqrt(N) = 0.03mm, merely by asking each person's opinion and averaging the results.
- E.T.Jaynes, "Probability Theory, The Logic of Science"
Any encryption is better than no encryption. Besides, once people learn how to encrypt things, it is pretty easy to just forget to turn it off. Or to receive encrypted emails in case I want to send one to them.
> So we've established you're not subscribed to the LKML...
Of course not. After all, I believe that mailing lists are a hideous misuse of email. A usenet group or a web forum is a much more appropriate way of having discussions about something. It is much easier to read posts that way instead of wading through a sea of email on subjects in which you have no interest.
> You know you're free to view those patents anyway.
Only if you do not plan to have any ideas in forseeable future. You never know if you might think of something clever that is based on something from a patent you forgot you viewed. Then you will be liable for willful infringement instead of innocent one, and with triple the damages. So no, I would highly recommend you stay away from the patent database.
You obviously have not actually used a satellite connection. The quality of service is abysmal and you'll be lucky if you can get speeds faster than dialup. Oh, and latency totally sucks. And it costs an enormous amount of money for what you get. Satellite companies have every incentive to cram as many users as possible onto their satellite(s), and so they do, with the result being speeds just fast enough to prevent their offices being razed by angry customers.
Slashdot Mirror
> Our industrial UPS (which is orders of magnitude more reliable than any APC
> product ever made) recently exploded, burnt, and shorted out
Well, I admit, that if my building burns down, I just might have to suffer through the inconvenience of running fsck on my filesystems for ten minutes or so, while you with your journalled fs will be able to restart in less than a minute. Call me a fool, but I think I just might be willing live with such a horrible risk.
> It's also painfully obvious that you've never worked as a sysadmin before.
Of course not, and that's my point: the world does not revolve around servers. Some of us use Linux on the desktop, and if the kernel developers don't get their head out of their asses and realize that, the year of "Linux on the desktop" will never come.
> Wake me up after you've been waiting for fsck to finish on your 1TB
> drive and it's been running for the last 72 hours.
If you are a sysadmin and it takes you 72 hours to fsck a 1TB drive, you should be fired.
> Yeah, because systems never kernel panic, or crash for any other reason than power outages.
You must be running Windows. Serves you right.
If you were running Linux, you wouldn't be having these problems. I haven't seen a kernel panic since 1996, and you wouldn't either if you stopped running the daily build from Linus's git tree on crappy hardware from the thrift shop.
> Just search for benchmarks, something like reiserfs beats ext2 by huge margins
You mean like these ones where ext2 beats reiserfs in most cases and is at least as fast in the others?
> I hope you're joking. ext2 is nice and simple, but it's neither fast not reliable.
> It uses a linear search to find directory entries, which means it's very slow on
> large directories, like Maildir mailboxes.
Believe it or not, the world does not revolve around huge mail servers. Some of us actually run Linux on a desktop, and so don't really care about how well an fs handles a million maildir mailboxes. Latency is the most important criteria, and reiserfs is just too complicated to deliver it, as well as being a largely fringe fs. Especially now with Hans gone, it would become even more fringe.
> It doesn't do tail packing which means it wastes space and is slower with small files.
Yup, I'd like to have efficient small file handling. But really, it is better to avoid having many small files in the first place. Use compressed archives to store such things; it's quite a bit more efficient, and does not require exotic file systems which most normal people (i.e. your customers) will not use.
> It's not reliable because without a journal it needs a fsck after a bad shutdown
I used to do that, and then I got a UPS instead and switched back to pure ext2. The performance hit from journalling is simply too high to tolerate. A decent UPS (pretty much anything made by APC) will prevent the crashes in the first place, solving the problem completely and without any unnecessary overhead. With UPS prices being as low as they are, there is no excuse for not having one, so I think that journalling will become obsolete in some near future.
Nah, it stands for Blue Screen of Death.
Ah, yes. My lovely creatures lounging on IKEA furniture and sipping Starbucks coffee. Oh, drat, that expansion pack broke the nude patch again...
> Call me old fashioned but I should be able to grab everything under C:\Program Files\My App
> and move that to another machine and run it
Old fashioned? Hell no! You're a newbie who never uses the command line or cares about how much
space your applications are using. Ok, so that's what pretty much all Windows users are, but
users are not the ones who should be setting the standards here. Application directories are
indeed convenient in your particular scenario. Sure, manipulating a single directory is easy,
but nobody should do that! A user should not even know that there is a "Program Files"
directory, or have any access to it whatsoever. It is the administrator's job to take care of
program installation. For home users, Windows provides the uninstaller tool under the control
panel, which is much simpler than looking for the application directory, which is usually
buried under the company directory (what vain idiot came up with that?), since the actual
application name is displayed in the uninstall list. Copying an app to another machine is not
a good idea either since configuration will likely be different and will need to be updated.
That's why we have installers in the first place.
From the developer's perspective, application directories are much worse. They stop you from
sharing code. All the libraries you use, you must package with the app, and while they are
running, they are not shared with any other app, leading to a considerable increase in your
program's memory footprint, and consequent complaints by your users about how your worthless
app takes five minutes to just start. Critical updates to important libraries can not be
performed easily. Say someone finds a vulnerability in a dll and issues a fix. With application
directories, the updater would have to search for every instance
of that particular dll and replace each and every one!
Then there is the problem of not being able to launch the programs from the command line.
Windows users don't do that, but any "old-fashioned" power user can't live without it. It is,
in fact, one of the things I really hate about Windows, that all the apps live under their
own little app dirs, and are not in the PATH.
As for the "convenience" of manipulating directories, a proper package manager is a much
better solution. Not only does it allow you to recreate the package, copy it to another
machine, and reinstall like you wanted, it will also track all dependencies and install
them automatically for you. If you think about it, there really isn't much difference
whether the filesystem keeps track of where the package is (with app dir), or the
package manager keeps track of where all the package parts are (with the install log).
The power of the merged hierarchy amply compensates for preventing you from mucking
directly with the app dir.
> Lets say you have 10,000 users using your program under Windows 95.
> You store their files under Program Files, Program name, User
In other words, you want to be inconsiderate and make me hunt down my data files under some weird directory name under Program Files, which, by the way, is hidden by default until you tell explorer to show system files or something. There is absolutely no excuse for writing data files in the program directory. Windows 95 supports home directories. Use them.
> then XP hits. It recomends that you move those data files to My Documents.
> Well Now you have 50,000 users. Do you make them move them?
Hell no! That's also inconsiderate. You have the setup program move them and tell the user what it did. No ifs ands or buts. Sure, you'll annoy your 50000 users, but that's just what you deserve for not following standards.
> Now comes Vista and it throws a fit if you store a data file under Program files
> Which if they are none executable I don't see what the problem is.
The problem is that users should never write to Program Files. Period. The permissions should be set to not allow it. This is a very very good idea because it prevents viruses from mucking up your installed programs. And, of course, it prevents the user from accidentally deleting them. Mac users, in particular, tend to think that an app can be uninstalled by deleting its directory.
> So you simply tell people to turn off the UAC.
Heck no! There is no way I'm compromising my computer's security just to cover for your bad design. I'll either stop using your program, or implement an inconvenient workaround (which normal users would not be able to do), consisting of manually going to Program Files/Your App and setting permissions on the directories you are foolishly trying to write to. Old games are VERY bad about this. Fortunately, setting permissions on the saved game subdir usually works.
It would be a much better idea to force every programmer to run under a non-Administrator account (and no Administrators or even Power Users group membership either!) Anyone who complains is obviously writing bad code, since there is absolutely no friggin' reason that a regular application should require administrative privileges. Whatever you set during setup is IT! And, for God's sake, learn to open registry keys in read-only mode!
The probability of an event that has already happened is always equal to 1. Predictions are uncertain only about the future, and, of course, for past events about which there is not enough information.
Get your stinkin' 3D off my cell phone! It is yet another ridiculous piece of bloatware because of which it takes ages to turn the phone on or off. Do these developers realize that most people don't want the stuff they already implemented? How about focusing on battery life, startup time, and voice clarity instead, huh? You know, the things that actually matter in a phone, instead of making 3D games, which are cool, but most certainly do not belong on a handheld device.
> No, half of the people in this country are at or below median intelligence.
IQ is based on the assumption that intelligence distribution is normal. In the normal distribution, the average is the same as the median, and yes, half the people in the world have below-average intelligence. That does not mean that half the people in this country have below average intelligence, because IQ averages vary by country. If you look at the table of IQs by country (which are averages, BTW), you'll see that the US has the average IQ of 98, meaning that slightly more than half the people in this country have below-average intelligence: 55.3%. By comparison, in Equatorial Guinea, where the average IQ is 59, 99.7% of the population has below average intelligence. You can get the numbers by calculating the error function with this calculator, 59 is 41/15(SD)=2.73(3) standard deviations.
"Now, kids, can you say 'doubly-linked list'? I knew you could!"
No, the Falling Blue Sky Of Death.
> what can we do with this new knowledge other than escape the bubble to realize our true freedom?
We can finally say that beer is the true meaning of the universe. This finding confirms what beer drinkers around the world have suspected for years: our universe is just a bubble in a giant glass of beer! In the beginning, the beer was flat. Then suddenly the bottle was opened, and the lowered pressure lowered carbon dioxide's solubility and enabled creation of bubbles. As the primordial beer gas accumulated in our bubble, gravity appeared (the surrounding universe is made of light beer, which does not bend space as much as the regular beer) and caused the carbon dioxide to coalesce into stars and planets, and eventually into people. Our bubble is expanding now, and floating upward in the glass. Eventually it will reach the top and become a part of the giant cosmic head, at which point we shall all be judged for our actions and be doomed to either sink back in the glass, or to fly up into the cosmos with the angels. Yup, dude, this is some heavy stuff! But don't worry, the more beer you drink, the better you understand it!
Hey, it's a good name! It's about time we stopped focusing on creating glistening new tools and started thinking about actually using them for something. But, of course, being an average Slashdot reader, you probably don't know how.
China is sending men into space. We are scrapping our only way of getting into space. Talk about progress...
Alpha: it doesn't work.
Beta: it still doesn't work.
- E.T.Jaynes, "Probability Theory, The Logic of Science"
> It's main color is no longer blue, it's brown
Windows: the biodegradable edition!
Any encryption is better than no encryption. Besides, once people learn how to encrypt things, it is pretty easy to just forget to turn it off. Or to receive encrypted emails in case I want to send one to them.
> So we've established you're not subscribed to the LKML...
Of course not. After all, I believe that mailing lists are a hideous misuse of email. A usenet group or a web forum is a much more appropriate way of having discussions about something. It is much easier to read posts that way instead of wading through a sea of email on subjects in which you have no interest.
How about this one?
That's two orders of magnitude more email than I have received in my entire life!
> You know you're free to view those patents anyway.
Only if you do not plan to have any ideas in forseeable future. You never know if you might think of something clever that is based on something from a patent you forgot you viewed. Then you will be liable for willful infringement instead of innocent one, and with triple the damages. So no, I would highly recommend you stay away from the patent database.
> Stop whining and get a satellite connection.
You obviously have not actually used a satellite connection. The quality of service is abysmal and you'll be lucky if you can get speeds faster than dialup. Oh, and latency totally sucks. And it costs an enormous amount of money for what you get. Satellite companies have every incentive to cram as many users as possible onto their satellite(s), and so they do, with the result being speeds just fast enough to prevent their offices being razed by angry customers.