I considered mentioning that one, but Tremont St's exceptionally confusing, whereas I wanted an intersection that's common. And Google and Mapquest both identify that intersection as being confusing, and say things like "go straight onto Shawmut Ave" (Mapquest actually says "turn right to stay on Tremont St"), while they don't seem to find the Winchester one particularly notable.
I think everybody I know uses google these days. Of the people I know, I think I'm the only one who ever got anywhere using mapquest without getting completely lost, and that was due to figuring out the ways it was likely to mess up.
Google maps doesn't give better directions, but you can actually follow the route on the map and look at the turns, so you have some chance of recognizing the places when you're there.
At least, this is true of New England, which has a lot of confusing roads (like ones which bend 90 degrees at an intersection with two other roads parallel to the turning street). Being able to identify that the left turn you're going to make involves not turning the steering wheel is very useful.
As long as they don't put junk in the results pages, it doesn't matter too much. People with modern browsers tend to put searches into the browser's box, not go to a google page, and people with obsolete browsers generally need a portal to get RSS and such.
There's nothing to figure out. I'd support the things I like and wouldn't support the things I didn't.
So are you going to support the trust to have the street in front of your house repaved when it needs it? How much are you going to contribute to snow removal? Do you want to support giving local kids something better to do than break things in your neighborhood? Which company do you want to pay for trash removal? Are you interested in funding airplane inspections so that planes don't fail and crash into your house? How about medical research, which might save your life at some point if enough people decide to fund it to let it get anywhere?
Most items on budgets at all levels, from the city to the country, are reasonably likely to be important to you in some way you don't have time to understand. Furthermore, most of them have to be funded at a certain level, or the funding they do get is wasted. The modern world is sufficiently complex that determining what things you want to fund would be more than a full-time job.
Why can't my tax dollars stay in my pocket so I can decide how to spend them?
Because then you'd have to make individual one-dollar payments to a couple hundred organizations that you're likely to consider worthwhile if you were to look into them, and they'd have to spend a lot of money advertizing their worthiness to you. Having a government you pay taxes to which pays departments is an O(n) process, while having individuals pay them directly is O(n^2). If you think your taxes are complicated now, just think how much it would suck if you had to essentially work out the federal budget yourself each year.
Hard drive manufacturers track their customers and sell them drives from matching runs. If you happen to get into a manufacturer's bad group, you'll just get lousy drives from them until you stop buying from them. Meanwhile, your friends will be perfectly happy with the drives they get from the same company, because they're getting drives from a different run. They do this because they want to dump as many bad drives as possible on anyone stupid enough to stick with a supplier whose stuff fails instead of spreading failures among satisfied customers.
Desktop users are very resistant to any change in their systems at all. In the server market, the underlying system can change significantly while nobody outside of the support staff particularly notices, so it's possible for them to just switch to whatever platform makes them most comfortable. On the desktop, the main thing preventing Linux adoption is really inertia; if GUI installers were such a big deal Windows XP would have displaced Windows 2K by now. As it is, the main thing driving XP growth is shipments of new machines that users can't get old versions of Windows on.
Linux will grow into the desktop space when Novell or IBM or Red Hat really decided to tackle that market segment, which will probably only happen when there's not enough room for growth in the server segments. Linux geeks only have reason to care about what they themselves use, and what their families (and other people who want computer help) use.
NOTE! This copyright does *not* cover user programs that use kernel services by normal system calls - this is merely considered normal use of the kernel, and does *not* fall under the heading of "derived work".
The license for glibc says:
We use this license for certain libraries in order to permit linking those libraries into non-free programs.
TiVo's obligation is to provide source so that you can put together the base system that the TiVo software runs on (meaning that taking it apart is not particularly necessary), but they only have to include changes they made to the base system, not complete programs they wrote (and they probably chose supported hardware, formats, and so forth, so that they wouldn't have to write special drivers).
The Grand Challenge rules say that you put in the route on a map, and the vehicle has to drive itself there. Obviously, this is how any Mac car would have to work, so it's a perfect fit. Really, those cars with 2 buttons (or, worse, 3) and a wheel are just too complicated for users, and just lead to people making mistakes. With an iCar, on the other hand...
Actually, make scales pretty well if you have a single make process read a bunch of files with "include", possibly recursively. It handles having different rulesets if you use static pattern rules. The only problem I haven't been able to solve effectively in my use of make is getting it to recheck dependancies after recomputing them (and I haven't figured out how to pull together the linking requirements of libraries and get each one listed once in the necessary order).
The main thing is that you really want to have a Makefile with all your rules, and have a file of variable definitions for each directory, so that the build system rules don't have to be duplicated everywhere.
Actually, a number of the top suggestions for different first letters on Google Suggest are for names of sites (including gmail, yahoo, 1800flowers, and 3m). Of course, Google doesn't actually resolve names for you, but they're still evidentally a popular thing to try...
Actually, that table is probably good for chemists as well. It's somewhat expanded from the table of the elements, because it gives properties of ions as well, which means that it has appropriate spots for a lot of information that generally ends up in other tables. It's not just the model that matters; it's what the model is of. It matters a whole lot, for example, if you've got H+ or H- in your reaction.
A chemist would probably want to include more in the section of elemental forms (having a use for elemental forms of Cl and Na, neither of which you're going to find sitting on the ground).
(In contrast, the spiral chart is a model of the same thing, with the primary difference being the addition of visual association between the ends of periods at the cost of a lot of information density)
It seems like we've actually been freed of the need to remember things, and now are asked to pay attention to things. This is, of course, much worse for actually getting things done, because it prevents continuous thought and occupies short-term memory, which is very limited rather than long-term memory, which is copious.
I think that essentially the Peter Principle applies: more demands are placed on us until we are dysfunctional. It takes a certain amount of self-importance to refuse further demands before you're completely bogged down, and further demands are certainly no less available now than they were.
Of course, there's hope for the situation. The present demands can be managed a lot more effectively than the former demands, because you can just have your phone go to voicemail, turn off automatically checking your email, hide IM, and check all of these things when you've finished a task. It was a lot harder with the technology of 1945 to disregard the need for impractical quantities of reference material on hand for complex tasks.
The point (or, at least, a point) is that the configurability is available to users. The duplication of effort between different distros is an unsolved problem in the organization of the community. The kernel development process in 2.6 was designed to resolve this issue for the kernel, such that distros would ship mainline kernels (or kernels with a small number of patches which are put into a later mainline kernel). Of course, this doesn't help the similar problems with all of the other packages.
Personally, I think all of the distros should become instances of Gentoo: Gentoo has support for getting patches and binary images from other sources, and for generating from the master meta-distribution the system that you want to distribute to other users. Adopting emerge as the build system for Red Hat, debian, etc. packages would mean that the patches applied by different distros to the packages they distribute would be represented in a compatible way.
Patents still expire, because there are too many companies that depend on other companies patents expiring. For example, the entire generic drugs industry.
I get kids ask all the time "when are we gonna use this...". It's like they have no understanding of why history matters, and then, educrats and the morons running teacher schools give them perfect out.
Do you have an understanding of why history matters? Can you explain it? Is all of the material you cover actually relevant to the aspects of history that matter?
There is little value to knowing a lot of names and dates. The only reason to know a name or date is to make connections when things correlate. And curricula are organized such that every match is made explicit, because there's a limited amount of material you can cover.
It is worth understanding how Americans were convinced that the Mexican War and the Oregon Purchase were worthwhile by the "Manifest Destiny" platform. Having an example of an American president who set out a campaign platform, did the things on it, and didn't run for re-election is worthwhile, and so might be knowing that this is a major virtue in the ideals of the Roman Republic (although rarely followed). Unfortunately, students often get tested on "#11: Polk 1845-1849", which is trivial to look up and useless for making a point.
But C0x is an entirely different language (the development version of the successor to C99). If you want to make it easier to say, arrange it as C0x++, which is obviously pronounced "spam".
This is really an argument for versioning dynamic libraries very carefully. The Linux dynamic linker has perfectly good support for avoiding problems like this. Each shared library has a "SONAME" field. Programs linked against the library should be able to use any later version of the library with the same SONAME. If the library changes in some way that breaks desireable behavior, it is supposed to get a new SONAME. The system keeps two sets of symlinks, in addition to the object files: libfoo.so is the latest version and libfoo.so.{version} is the latest version of a compatible series. When you link a program, you use the libfoo.so to find which series is newest. When you run a program, you use libfoo.so.{version}, so you get the newest compatible version.
It sounds like the problem you had was that the common library version number wasn't changed when it should have been. If it had been changed, there would have been no effect on any of the existing programs, static or dynamic. Of course, for a two-line fix to a function which only changes the behavior of the library when dealing with corrupted files, and resulted in a buffer overflow instead of an error, the change won't break anything except attacks, so keeping the same SONAME is right.
Note that you can see this by doing "ldd {program}", which will report how it looks for each library and what it finds. And you can see that zlib claims to be backwards compatible all the way through 1.x, while openssl is only to 0.9.7.
The previous section, on being able to use old hardware, makes it sound like they're also doing Linux client machines in some classrooms. They probably have some classes where the computer use is exclusively a web browser that doesn't have to be IE, and others with Windows desktops. Including hardware costs, 30k pounds is plausible for a year in which they would have needed new hardware.
That's just the kernel. You have to add the version numbers for the kernel, glibc, gcc, and firefox to get the Linux version. I'm at 8.12.16.1, for example, which seems to be sufficient.
Most likely, the IP will wind up with Novell, due to SCO's successor not wanting to get into a copyright battle with Novell based on SCO's insane legal theories and implausible interpretations of contracts. The evidence in SCO v. Novell on SCO's side is sufficiently good for SCO to try to make a case out of it, because they have nothing to lose, but it's nowhere near good enough for anyone with assets vulnerable to a countersuit. If Sun wants the UNIX copyrights, they'd do better to wait until SCO goes away and then buy them from Novell than buy SCO's position.
I'm not sure how well you could trust a P2P system like this
"This is not a trust system. Trust requires identity first."
The only thing that this does is that it lets someone who has established an identity use that identity in other places without a relationship between the sites or between the user and the new site. This would let me convince groklaw that I'm http://slashdot.org/~iabervon as effectively as I convince slashdot itself without enabling groklaw to spoof me to other sites (like if I just sent it my slashdot password). It doesn't mean that groklaw should think I'm not a troll just because I've got a slashdot account, but it could, at least, check my latest 24 comments for spelling or moderation, or whatever, and be quite sure that I'm actually the person who made them.
Everybody likes to put things on their desktop. Whether they like to have those things stay on their desktop forever is a different question. The ideal would probably be for the system to let people toss stuff on the desktop arbitrarily, but then index these files into a database that would let them find them later. They'd sit on the desktop for a limited time (unless you kept using them), after which they'd disappear and only be found through the database (at the point where you'd probably lose track of them on the desktop anyway). That way, the desktop wouldn't get cluttered, but it would act as somewhere (in the interface) to put the documents you're working on and somewhere to leave documents that you want to be indexed.
A fifth of the capacity of their system before they switched to Linux was on hardware purchased in 2001 to handle the rush of bookings when airports reopenned after September 11th. Most likely, routine upgrades in capacity and regular equipment replacement meant that the rest of their system was relatively recent as well; the savings on running a modern mainframe over running a 1970s era one (in terms of maintenance, power usage, and space occupied for the amount of computational power) would pay for buying a new one.
Slashdot Mirror
I considered mentioning that one, but Tremont St's exceptionally confusing, whereas I wanted an intersection that's common. And Google and Mapquest both identify that intersection as being confusing, and say things like "go straight onto Shawmut Ave" (Mapquest actually says "turn right to stay on Tremont St"), while they don't seem to find the Winchester one particularly notable.
I think everybody I know uses google these days. Of the people I know, I think I'm the only one who ever got anywhere using mapquest without getting completely lost, and that was due to figuring out the ways it was likely to mess up.
Google maps doesn't give better directions, but you can actually follow the route on the map and look at the turns, so you have some chance of recognizing the places when you're there.
At least, this is true of New England, which has a lot of confusing roads (like ones which bend 90 degrees at an intersection with two other roads parallel to the turning street). Being able to identify that the left turn you're going to make involves not turning the steering wheel is very useful.
As long as they don't put junk in the results pages, it doesn't matter too much. People with modern browsers tend to put searches into the browser's box, not go to a google page, and people with obsolete browsers generally need a portal to get RSS and such.
There's nothing to figure out. I'd support the things I like and wouldn't support the things I didn't.
So are you going to support the trust to have the street in front of your house repaved when it needs it? How much are you going to contribute to snow removal? Do you want to support giving local kids something better to do than break things in your neighborhood? Which company do you want to pay for trash removal? Are you interested in funding airplane inspections so that planes don't fail and crash into your house? How about medical research, which might save your life at some point if enough people decide to fund it to let it get anywhere?
Most items on budgets at all levels, from the city to the country, are reasonably likely to be important to you in some way you don't have time to understand. Furthermore, most of them have to be funded at a certain level, or the funding they do get is wasted. The modern world is sufficiently complex that determining what things you want to fund would be more than a full-time job.
Why can't my tax dollars stay in my pocket so I can decide how to spend them?
Because then you'd have to make individual one-dollar payments to a couple hundred organizations that you're likely to consider worthwhile if you were to look into them, and they'd have to spend a lot of money advertizing their worthiness to you. Having a government you pay taxes to which pays departments is an O(n) process, while having individuals pay them directly is O(n^2). If you think your taxes are complicated now, just think how much it would suck if you had to essentially work out the federal budget yourself each year.
Hard drive manufacturers track their customers and sell them drives from matching runs. If you happen to get into a manufacturer's bad group, you'll just get lousy drives from them until you stop buying from them. Meanwhile, your friends will be perfectly happy with the drives they get from the same company, because they're getting drives from a different run. They do this because they want to dump as many bad drives as possible on anyone stupid enough to stick with a supplier whose stuff fails instead of spreading failures among satisfied customers.
Desktop users are very resistant to any change in their systems at all. In the server market, the underlying system can change significantly while nobody outside of the support staff particularly notices, so it's possible for them to just switch to whatever platform makes them most comfortable. On the desktop, the main thing preventing Linux adoption is really inertia; if GUI installers were such a big deal Windows XP would have displaced Windows 2K by now. As it is, the main thing driving XP growth is shipments of new machines that users can't get old versions of Windows on.
Linux will grow into the desktop space when Novell or IBM or Red Hat really decided to tackle that market segment, which will probably only happen when there's not enough room for growth in the server segments. Linux geeks only have reason to care about what they themselves use, and what their families (and other people who want computer help) use.
The license for glibc says:
TiVo's obligation is to provide source so that you can put together the base system that the TiVo software runs on (meaning that taking it apart is not particularly necessary), but they only have to include changes they made to the base system, not complete programs they wrote (and they probably chose supported hardware, formats, and so forth, so that they wouldn't have to write special drivers).
The Grand Challenge rules say that you put in the route on a map, and the vehicle has to drive itself there. Obviously, this is how any Mac car would have to work, so it's a perfect fit. Really, those cars with 2 buttons (or, worse, 3) and a wheel are just too complicated for users, and just lead to people making mistakes. With an iCar, on the other hand...
Actually, make scales pretty well if you have a single make process read a bunch of files with "include", possibly recursively. It handles having different rulesets if you use static pattern rules. The only problem I haven't been able to solve effectively in my use of make is getting it to recheck dependancies after recomputing them (and I haven't figured out how to pull together the linking requirements of libraries and get each one listed once in the necessary order).
The main thing is that you really want to have a Makefile with all your rules, and have a file of variable definitions for each directory, so that the build system rules don't have to be duplicated everywhere.
Actually, a number of the top suggestions for different first letters on Google Suggest are for names of sites (including gmail, yahoo, 1800flowers, and 3m). Of course, Google doesn't actually resolve names for you, but they're still evidentally a popular thing to try...
Actually, that table is probably good for chemists as well. It's somewhat expanded from the table of the elements, because it gives properties of ions as well, which means that it has appropriate spots for a lot of information that generally ends up in other tables. It's not just the model that matters; it's what the model is of. It matters a whole lot, for example, if you've got H+ or H- in your reaction.
A chemist would probably want to include more in the section of elemental forms (having a use for elemental forms of Cl and Na, neither of which you're going to find sitting on the ground).
(In contrast, the spiral chart is a model of the same thing, with the primary difference being the addition of visual association between the ends of periods at the cost of a lot of information density)
It seems like we've actually been freed of the need to remember things, and now are asked to pay attention to things. This is, of course, much worse for actually getting things done, because it prevents continuous thought and occupies short-term memory, which is very limited rather than long-term memory, which is copious.
I think that essentially the Peter Principle applies: more demands are placed on us until we are dysfunctional. It takes a certain amount of self-importance to refuse further demands before you're completely bogged down, and further demands are certainly no less available now than they were.
Of course, there's hope for the situation. The present demands can be managed a lot more effectively than the former demands, because you can just have your phone go to voicemail, turn off automatically checking your email, hide IM, and check all of these things when you've finished a task. It was a lot harder with the technology of 1945 to disregard the need for impractical quantities of reference material on hand for complex tasks.
The point (or, at least, a point) is that the configurability is available to users. The duplication of effort between different distros is an unsolved problem in the organization of the community. The kernel development process in 2.6 was designed to resolve this issue for the kernel, such that distros would ship mainline kernels (or kernels with a small number of patches which are put into a later mainline kernel). Of course, this doesn't help the similar problems with all of the other packages.
Personally, I think all of the distros should become instances of Gentoo: Gentoo has support for getting patches and binary images from other sources, and for generating from the master meta-distribution the system that you want to distribute to other users. Adopting emerge as the build system for Red Hat, debian, etc. packages would mean that the patches applied by different distros to the packages they distribute would be represented in a compatible way.
Patents still expire, because there are too many companies that depend on other companies patents expiring. For example, the entire generic drugs industry.
I get kids ask all the time "when are we gonna use this...". It's like they have no understanding of why history matters, and then, educrats and the morons running teacher schools give them perfect out.
Do you have an understanding of why history matters? Can you explain it? Is all of the material you cover actually relevant to the aspects of history that matter?
There is little value to knowing a lot of names and dates. The only reason to know a name or date is to make connections when things correlate. And curricula are organized such that every match is made explicit, because there's a limited amount of material you can cover.
It is worth understanding how Americans were convinced that the Mexican War and the Oregon Purchase were worthwhile by the "Manifest Destiny" platform. Having an example of an American president who set out a campaign platform, did the things on it, and didn't run for re-election is worthwhile, and so might be knowing that this is a major virtue in the ideals of the Roman Republic (although rarely followed). Unfortunately, students often get tested on "#11: Polk 1845-1849", which is trivial to look up and useless for making a point.
But C0x is an entirely different language (the development version of the successor to C99). If you want to make it easier to say, arrange it as C0x++, which is obviously pronounced "spam".
Ladies and gentlemen please watch my assistant as amazing new Microsoft Windows Longhorn cleans tough stains like wine! Blood! Grass! Pet Stains!
You had me for the first three, but I'm pretty sure there's no OSS project called "pet stains".
This is really an argument for versioning dynamic libraries very carefully. The Linux dynamic linker has perfectly good support for avoiding problems like this. Each shared library has a "SONAME" field. Programs linked against the library should be able to use any later version of the library with the same SONAME. If the library changes in some way that breaks desireable behavior, it is supposed to get a new SONAME. The system keeps two sets of symlinks, in addition to the object files: libfoo.so is the latest version and libfoo.so.{version} is the latest version of a compatible series. When you link a program, you use the libfoo.so to find which series is newest. When you run a program, you use libfoo.so.{version}, so you get the newest compatible version.
It sounds like the problem you had was that the common library version number wasn't changed when it should have been. If it had been changed, there would have been no effect on any of the existing programs, static or dynamic. Of course, for a two-line fix to a function which only changes the behavior of the library when dealing with corrupted files, and resulted in a buffer overflow instead of an error, the change won't break anything except attacks, so keeping the same SONAME is right.
Note that you can see this by doing "ldd {program}", which will report how it looks for each library and what it finds. And you can see that zlib claims to be backwards compatible all the way through 1.x, while openssl is only to 0.9.7.
The previous section, on being able to use old hardware, makes it sound like they're also doing Linux client machines in some classrooms. They probably have some classes where the computer use is exclusively a web browser that doesn't have to be IE, and others with Windows desktops. Including hardware costs, 30k pounds is plausible for a year in which they would have needed new hardware.
That's just the kernel. You have to add the version numbers for the kernel, glibc, gcc, and firefox to get the Linux version. I'm at 8.12.16.1, for example, which seems to be sufficient.
Most likely, the IP will wind up with Novell, due to SCO's successor not wanting to get into a copyright battle with Novell based on SCO's insane legal theories and implausible interpretations of contracts. The evidence in SCO v. Novell on SCO's side is sufficiently good for SCO to try to make a case out of it, because they have nothing to lose, but it's nowhere near good enough for anyone with assets vulnerable to a countersuit. If Sun wants the UNIX copyrights, they'd do better to wait until SCO goes away and then buy them from Novell than buy SCO's position.
I'm not sure how well you could trust a P2P system like this
"This is not a trust system. Trust requires identity first."
The only thing that this does is that it lets someone who has established an identity use that identity in other places without a relationship between the sites or between the user and the new site. This would let me convince groklaw that I'm http://slashdot.org/~iabervon as effectively as I convince slashdot itself without enabling groklaw to spoof me to other sites (like if I just sent it my slashdot password). It doesn't mean that groklaw should think I'm not a troll just because I've got a slashdot account, but it could, at least, check my latest 24 comments for spelling or moderation, or whatever, and be quite sure that I'm actually the person who made them.
Everybody likes to put things on their desktop. Whether they like to have those things stay on their desktop forever is a different question. The ideal would probably be for the system to let people toss stuff on the desktop arbitrarily, but then index these files into a database that would let them find them later. They'd sit on the desktop for a limited time (unless you kept using them), after which they'd disappear and only be found through the database (at the point where you'd probably lose track of them on the desktop anyway). That way, the desktop wouldn't get cluttered, but it would act as somewhere (in the interface) to put the documents you're working on and somewhere to leave documents that you want to be indexed.
A fifth of the capacity of their system before they switched to Linux was on hardware purchased in 2001 to handle the rush of bookings when airports reopenned after September 11th. Most likely, routine upgrades in capacity and regular equipment replacement meant that the rest of their system was relatively recent as well; the savings on running a modern mainframe over running a 1970s era one (in terms of maintenance, power usage, and space occupied for the amount of computational power) would pay for buying a new one.