Indeed. (I think the trick of running words together like that comes from BASIC, though; or at least, that's the first place I saw it.)
The same line in BASIC would be rather simpler.
Indeed, you just had to write your ASM inside square brackets as if it were a BASIC program, and it was assembled into memory. (But the [ and ] rendered as arrows in the default text-only graphics mode.)
Given that you're talking about driver support, which is entirely a hardware issue and has nothing to do with the parts of the userland that GNU provide, "Linux" is correct here even if you're the sort of person who likes to list all the relevant parts of a system in its name. (That said, I suspect "GNU/Linux" is nowadays possibly not the description for systems generally used graphically; Gnome/Linux or KDE/Linux would be more accurate, as the desktop environment is a larger part of such systems than the command-line userland; although the GNU tools are still important and relevant, they would easily be replaceable with, say, the BSD equivalents.)
I just tested on an Ubuntu LTS release I had handy:
Script started on Tue 27 Sep 2011 13:24:02 BST
root@hostname:~# uname -a
Linux hostname 2.6.32-34-generic #77-Ubuntu SMP Tue Sep 13 19:40:53 UTC 2011 i686 GNU/Linux
root@hostname:~# wc -l/etc/shadow
42/etc/shadow
root@hostname:~# exit
Script done on Tue 27 Sep 2011 13:24:10 BST
So starting in single user mode gave me enough permissions to write to/root (that's where the typescript was saved), and to read/etc/shadow. (I didn't try writing/etc/shadow, but I'm pretty confident it would have worked, given that I can write to a root-only location and that I can read it./root and/etc are on the same partition on that system.)
The distro did pop up a menu in response to my addition of "single" to the command line, but one of the options was to open a root shell, and that's the one I used. So even if it didn't drop into single-user mode directly, it wouldn't have been a lot of trouble to reset a password.
On Linux, interrupt the boot process at the bootloader prompt, choose to edit the kernel command line, then add "single" to the end of it and continue the boot process. It boots into single user mode, and the same way as on the Mac, will allow you to change arbitrary users' passwords (including root) without further authentication. I'm pretty sure that this is deliberate; physical access = game over, so why not make it easy for a sysadmin to recover a lost password?
The amount of current/amperage is a measure of how quickly the charge can deplete. When you talk about the current of a battery, it's a case not of "how much current does the battery have" but "how much is it capable of providing when the contacts are shorted together", which is a useful concept.
In the case of charge buildup on a human, the most useful current-like statistic would probably be the amount of stored energy. Once you know that, you can work out the maximum current that can be sustained for a given length of time through a given material. (The voltage would affect how much current would pass through a material of a given resistance, and the amount of energy how long it would be sustained for; even thousands of amps moving over a potential difference of thousands of volts won't do much if it only lasts for a femtosecond, as there's not that much energy dissipated.)
The idea of the cards is that the customer should insert the card into the chip reader themself, and not let the store staff get their hands on it at all. I always insist on doing this, just in case (and have upset a few shop staff in the process). If other people did at well, it'd be one theft vector closed off, but it seems unlikely; people are all in the habit of giving their card to the cashier.
Antivirus software should be the norm on mailservers, regardless of what OS they run. (And it should primarily be set to check for Windows viruses.) It's worth making sure that there aren't known exploits or malware included in the mails that you're forwarding to your users, even if the server itself wouldn't be affected by them.
It's possible to make a plausible argument along these lines for webservers too, although a much less strong one.
That'd be a CSRF attack against Facebook if it worked. Imagine if sites could simulate a "Like" on themselves from users who had no intention of clicking the button, but had actually clicked something entirely different. It's a good thing that it's impossible.
(IANAL...) Indemnification is basically an insurance against being sued over a particular product; if you're indemnified to do something and get in legal trouble for doing so, the person indemnifying you had better pay up. You can be indemnified for use of open source products just the same as closed source products, but you're probably going to have to pay someone for that rather than have it included in the price of the product. (I don't know how many, or any, closed source products do indemnify their users, though. More commonly, they ask their users to indemnify the makers of the software against problems resulting from their use of it...)
What you're missing is that even if they have both the hash/checksum and the salt, they then have to start bruteforcing the password from scratch. Because they didn't know the salt in advance, they couldn't have precomputed a dictionary of hash -> password mappings like you can with unsalted passwords, so each password has to be bruteforced separately. If the passwords aren't salted, and are just plain md5 (say), it's entirely possible that they already know what the md5 reverse-maps to because they've seen it before; it only takes a quick Google to determine that 5eb63bbbe01eeed093cb22bb8f5acdc3 is the md5 hash of "hello world", for instance, whereas if it were salted you'd have to try possibilities from scratch until you guessed that the input was "hello world" (it's kind-of guessable, but it likely wouldn't be near the top of your dictionary, so it'd take a while of brute-forcing to hit it). So, a salt is useful even if it's stored with your password file.
As another issue, if you don't salt your passwords, merely by comparing hashes you can tell that two users have the same password from a list of hashes. This also means that you could register an account on the site with a common password like "password" and instantly see who else was using it, giving you a load of accounts to quickly compromise...
Just store the binary and source on the same server, surely? That way, when you're not distributing source, you're not distributing binary either, so you're still in compliance.
Pretty much every software EULA designed for the UK explicitly allows people to make one backup copy. I assume that's because the lawyers fear the entire thing will be unenforceable if it doesn't.
Basically, because governments have to get their income somehow. The higher the tax rate, the larger a proportion of the money is paid by richer people.
When governments aren't getting taxes, they get the money in less direct ways, such as by taking bribes, or by cutting down on public services. Both those options tend to hit poorer people harder (indirectly, in the first case).
I think it's because 4th edition came out, and it was basically an entirely different game from the ones that came before it, with a few references back. It's not necessarily a bad game, but it's not a continuation of the previous versions, and it rather split the playerbase (between the ones who keep playing 3rd edition or variants of it, and ones who've moved onto the new game). If Wizards hadn't called it Dungeons & Dragons, there wouldn't really be a problem. As it is, though, it's a bit like, say, Python 2 and Python 3; a mess that's hurting the game in general.
It was me who calculated the optimal number of moves to win, at 2016. (I'm not completely convinced that the calculation is correct, though.) The current human records are 2135 with bones-stuffing (i.e. deliberately suiciding an advanced character to drop a whole load of valuable items into the game where another character can pick them up), and 2596 without any bones abuse, both by Maud. (For those wondering, the realtime record holder is Sayo (an alternate account of Adeon), with 1 hour 16 minutes.) It's obviously impossible to beat 2000 turns as that's the earliest you can enter the Quest. (For more information, see http://nethackwiki.com/wiki/Speed_ascension.)
I'd say that speed-ascension is far from solved. Maud can only get runs that fast when he gets incredible amounts of luck (only around 1% of games are lucky enough for him to even bother trying to speedrun them, he generally quits the other 99%), for instance; and different ways to do things are being discovered all the time. The polyself-based strategy all speed ascenders use nowadays is only a couple of years old, and very different from the way a normal game plays.
My own variant AceHack (which has been unexpectedly Slashdotted, it seems; I'm glad I'm not hosting it myself) is designed to improve the interface, while keeping the gameplay the same or slightly easier. (Most of the other variants are indeed designed to make the game more complex/harder.) It's also something of a protest against the way that vanilla NetHack is effectively unmaintained nowadays; although it is apparently still being developed (I've got replies to bug reports as recently as last week, including descriptions of what they changed in response), the repositories aren't public, and so nobody can actually benefit from the changes. So I'm trying to make an improved version without, hopefully, offending too many people (although some of the changes have still managed to annoy subsets of players; that's typical of anything, I guess).
This tournament is quite healthy for the game's ecosystem, in a way; rather than playing vanilla over and over again, it can advertise some of the more actively and publicly developed variants, and also the public servers on which many of the community play nowadays. (In particular, telnet://nethack.alt.org, or http://alt.org/nethack, is one of the most popular public servers nowadays, and many people play there in preference to locally; it's just one of the servers on which the tournament can be played.)
So in summary, the major advantage of the forks is that (other than SporkHack) at least they're still under development in a way that people can actually play them, unlike vanilla. My own fork AceHack is unfinished and still in alpha (SporkHack and UnNetHack have both had releases, and at least Un is likely to have more releases in the future), and we've been trading code and ideas between the forks reasonably freely.
With SI prefixes, GB = gigabyte = units of a billion bytes, GiB = gibibytes = units of 0x40000000 = 1073741824 bytes. Quite a few people think that this terminology is stupid, or at least stupidly-named (I mean, "gibibyte"?), but at least it's accurate, and it's nice to know exactly whether the units are based on powers of 2 or of 10.
The law's against carrying a knife in a way that it could easily be used to stab someone. A knife with a very short blade (such as a Swiss Army Knife) is, as far as I know, OK; as is a knife in packaging that makes it difficult or impossible to quickly draw it and stab someone. There's quite a difference between knife-for-practical-purpose and knife-for-stabbing-people.
There's no obvious way to wipe that single save slot, which may be what the grandparent is confused about. You do it by holding down a particular combination of buttons on the title screen, which is listed in the manual (I can't remember what it is offhand; at least one game uses select+up+B, but I can't remember if it's Pokémon or something else).
It's a 3DS game, and those generally store the save data on the actual cartridge. So if you buy a second cartridge, it'll likely have its own save slot.
Try Fn+7/8/9/u/o/j/k/l (and Fn-i to click), that works on my (Ubuntu 10.04) laptop. I think it's being converted to numlock-based or at least keypad-based codes behind the scenes. (You have to turn "Pointer can be controlled using the keypad" on in Assistive Technologies | Keyboard Preferences first, but that can be done entirely using keyboard shortcuts; the "System" menu can be opened via Alt-F1, Right, Right.)
Oh dear, is there a rule of some sort that if you're correcting someone else's grammar/spelling/(mis-)use of words, you'll get something wrong on your own post?
There is, it's called Muphry's Law. (For bonus points, if you bring it up in an argument on the Internet, there's about a 50-50 chance that you'll be incorrectly accused of misspelling its name.)
Slashdot Mirror
Indeed. (I think the trick of running words together like that comes from BASIC, though; or at least, that's the first place I saw it.) The same line in BASIC would be rather simpler.
Indeed, you just had to write your ASM inside square brackets as if it were a BASIC program, and it was assembled into memory. (But the [ and ] rendered as arrows in the default text-only graphics mode.)
Given that you're talking about driver support, which is entirely a hardware issue and has nothing to do with the parts of the userland that GNU provide, "Linux" is correct here even if you're the sort of person who likes to list all the relevant parts of a system in its name. (That said, I suspect "GNU/Linux" is nowadays possibly not the description for systems generally used graphically; Gnome/Linux or KDE/Linux would be more accurate, as the desktop environment is a larger part of such systems than the command-line userland; although the GNU tools are still important and relevant, they would easily be replaceable with, say, the BSD equivalents.)
I just tested on an Ubuntu LTS release I had handy:
Script started on Tue 27 Sep 2011 13:24:02 BST /etc/shadow /etc/shadow
root@hostname:~# uname -a
Linux hostname 2.6.32-34-generic #77-Ubuntu SMP Tue Sep 13 19:40:53 UTC 2011 i686 GNU/Linux
root@hostname:~# wc -l
42
root@hostname:~# exit
Script done on Tue 27 Sep 2011 13:24:10 BST
So starting in single user mode gave me enough permissions to write to /root (that's where the typescript was saved), and to read /etc/shadow. (I didn't try writing /etc/shadow, but I'm pretty confident it would have worked, given that I can write to a root-only location and that I can read it. /root and /etc are on the same partition on that system.)
The distro did pop up a menu in response to my addition of "single" to the command line, but one of the options was to open a root shell, and that's the one I used. So even if it didn't drop into single-user mode directly, it wouldn't have been a lot of trouble to reset a password.
On Linux, interrupt the boot process at the bootloader prompt, choose to edit the kernel command line, then add "single" to the end of it and continue the boot process. It boots into single user mode, and the same way as on the Mac, will allow you to change arbitrary users' passwords (including root) without further authentication. I'm pretty sure that this is deliberate; physical access = game over, so why not make it easy for a sysadmin to recover a lost password?
The amount of current/amperage is a measure of how quickly the charge can deplete. When you talk about the current of a battery, it's a case not of "how much current does the battery have" but "how much is it capable of providing when the contacts are shorted together", which is a useful concept. In the case of charge buildup on a human, the most useful current-like statistic would probably be the amount of stored energy. Once you know that, you can work out the maximum current that can be sustained for a given length of time through a given material. (The voltage would affect how much current would pass through a material of a given resistance, and the amount of energy how long it would be sustained for; even thousands of amps moving over a potential difference of thousands of volts won't do much if it only lasts for a femtosecond, as there's not that much energy dissipated.)
The idea of the cards is that the customer should insert the card into the chip reader themself, and not let the store staff get their hands on it at all. I always insist on doing this, just in case (and have upset a few shop staff in the process). If other people did at well, it'd be one theft vector closed off, but it seems unlikely; people are all in the habit of giving their card to the cashier.
Antivirus software should be the norm on mailservers, regardless of what OS they run. (And it should primarily be set to check for Windows viruses.) It's worth making sure that there aren't known exploits or malware included in the mails that you're forwarding to your users, even if the server itself wouldn't be affected by them.
It's possible to make a plausible argument along these lines for webservers too, although a much less strong one.
That'd be a CSRF attack against Facebook if it worked. Imagine if sites could simulate a "Like" on themselves from users who had no intention of clicking the button, but had actually clicked something entirely different. It's a good thing that it's impossible.
(IANAL...) Indemnification is basically an insurance against being sued over a particular product; if you're indemnified to do something and get in legal trouble for doing so, the person indemnifying you had better pay up. You can be indemnified for use of open source products just the same as closed source products, but you're probably going to have to pay someone for that rather than have it included in the price of the product. (I don't know how many, or any, closed source products do indemnify their users, though. More commonly, they ask their users to indemnify the makers of the software against problems resulting from their use of it...)
What you're missing is that even if they have both the hash/checksum and the salt, they then have to start bruteforcing the password from scratch. Because they didn't know the salt in advance, they couldn't have precomputed a dictionary of hash -> password mappings like you can with unsalted passwords, so each password has to be bruteforced separately. If the passwords aren't salted, and are just plain md5 (say), it's entirely possible that they already know what the md5 reverse-maps to because they've seen it before; it only takes a quick Google to determine that 5eb63bbbe01eeed093cb22bb8f5acdc3 is the md5 hash of "hello world", for instance, whereas if it were salted you'd have to try possibilities from scratch until you guessed that the input was "hello world" (it's kind-of guessable, but it likely wouldn't be near the top of your dictionary, so it'd take a while of brute-forcing to hit it). So, a salt is useful even if it's stored with your password file.
As another issue, if you don't salt your passwords, merely by comparing hashes you can tell that two users have the same password from a list of hashes. This also means that you could register an account on the site with a common password like "password" and instantly see who else was using it, giving you a load of accounts to quickly compromise...
Just store the binary and source on the same server, surely? That way, when you're not distributing source, you're not distributing binary either, so you're still in compliance.
Pretty much every software EULA designed for the UK explicitly allows people to make one backup copy. I assume that's because the lawyers fear the entire thing will be unenforceable if it doesn't.
When governments aren't getting taxes, they get the money in less direct ways, such as by taking bribes, or by cutting down on public services. Both those options tend to hit poorer people harder (indirectly, in the first case).
I think it's because 4th edition came out, and it was basically an entirely different game from the ones that came before it, with a few references back. It's not necessarily a bad game, but it's not a continuation of the previous versions, and it rather split the playerbase (between the ones who keep playing 3rd edition or variants of it, and ones who've moved onto the new game). If Wizards hadn't called it Dungeons & Dragons, there wouldn't really be a problem. As it is, though, it's a bit like, say, Python 2 and Python 3; a mess that's hurting the game in general.
I'd say that speed-ascension is far from solved. Maud can only get runs that fast when he gets incredible amounts of luck (only around 1% of games are lucky enough for him to even bother trying to speedrun them, he generally quits the other 99%), for instance; and different ways to do things are being discovered all the time. The polyself-based strategy all speed ascenders use nowadays is only a couple of years old, and very different from the way a normal game plays.
My own variant AceHack (which has been unexpectedly Slashdotted, it seems; I'm glad I'm not hosting it myself) is designed to improve the interface, while keeping the gameplay the same or slightly easier. (Most of the other variants are indeed designed to make the game more complex/harder.) It's also something of a protest against the way that vanilla NetHack is effectively unmaintained nowadays; although it is apparently still being developed (I've got replies to bug reports as recently as last week, including descriptions of what they changed in response), the repositories aren't public, and so nobody can actually benefit from the changes. So I'm trying to make an improved version without, hopefully, offending too many people (although some of the changes have still managed to annoy subsets of players; that's typical of anything, I guess). This tournament is quite healthy for the game's ecosystem, in a way; rather than playing vanilla over and over again, it can advertise some of the more actively and publicly developed variants, and also the public servers on which many of the community play nowadays. (In particular, telnet://nethack.alt.org, or http://alt.org/nethack, is one of the most popular public servers nowadays, and many people play there in preference to locally; it's just one of the servers on which the tournament can be played.) So in summary, the major advantage of the forks is that (other than SporkHack) at least they're still under development in a way that people can actually play them, unlike vanilla. My own fork AceHack is unfinished and still in alpha (SporkHack and UnNetHack have both had releases, and at least Un is likely to have more releases in the future), and we've been trading code and ideas between the forks reasonably freely.
With SI prefixes, GB = gigabyte = units of a billion bytes, GiB = gibibytes = units of 0x40000000 = 1073741824 bytes. Quite a few people think that this terminology is stupid, or at least stupidly-named (I mean, "gibibyte"?), but at least it's accurate, and it's nice to know exactly whether the units are based on powers of 2 or of 10.
The law's against carrying a knife in a way that it could easily be used to stab someone. A knife with a very short blade (such as a Swiss Army Knife) is, as far as I know, OK; as is a knife in packaging that makes it difficult or impossible to quickly draw it and stab someone. There's quite a difference between knife-for-practical-purpose and knife-for-stabbing-people.
And if you use it without wiping the save data first, you can't save the resulting game. Did you check your assertions before you made them?
There's no obvious way to wipe that single save slot, which may be what the grandparent is confused about. You do it by holding down a particular combination of buttons on the title screen, which is listed in the manual (I can't remember what it is offhand; at least one game uses select+up+B, but I can't remember if it's Pokémon or something else).
It's a 3DS game, and those generally store the save data on the actual cartridge. So if you buy a second cartridge, it'll likely have its own save slot.
Try Fn+7/8/9/u/o/j/k/l (and Fn-i to click), that works on my (Ubuntu 10.04) laptop. I think it's being converted to numlock-based or at least keypad-based codes behind the scenes. (You have to turn "Pointer can be controlled using the keypad" on in Assistive Technologies | Keyboard Preferences first, but that can be done entirely using keyboard shortcuts; the "System" menu can be opened via Alt-F1, Right, Right.)
You can just download a browser via FTP (the ftp command-line tool comes with Windows). That's often how I get Firefox on new Windows installations.
Oh dear, is there a rule of some sort that if you're correcting someone else's grammar/spelling/(mis-)use of words, you'll get something wrong on your own post?
There is, it's called Muphry's Law. (For bonus points, if you bring it up in an argument on the Internet, there's about a 50-50 chance that you'll be incorrectly accused of misspelling its name.)