I used to be a Lilo fan, until I ran into trouble with it on my IBM laptop -- it took WAY longer at the part where it loads the kernel to memory than it had any business to, a stage that goes by so fast it's easy to miss on most computers.
That's funny - I had an almost identical problem using grub with my ASUS laptop (used to take yonks to load stage 1.5), which forced me to switch to lilo. Lilo's more painful (have to re-run it each time you install a new kernel, doesn't have a nice command-line option) but at least it works and works consistently.
Because of your Google cookie, Google in theory knows exactly what you search for, what you shop for, who you IM, and your entire email records.
Which is why I search through google.com.au, not google.com, and I block cookies set by google.com.au.
It shocks me a bit that google likes to promote things like search history as a feature! Scary stuff... mind you, I still use Gmail so I guess I'm not that paranoid just yet:-)
Well, me for one, and I'm sure there's heaps of others out there too. I don't want kids, but I do want to share my life with someone. I most certainly don't need chicks for sexual gratification - I've got a hand and I'm not afraid to use it:) But love is far more than sex...
That easy, are you sure? If the police can intercept your e-mail, then most likely it will become forbidden to encrypt it - or the allowed encryption level will be far too weak to be usable.
The obvious way around this is for as many people as possible to start using encryption - once too many are doing it, the law becomes unenforcible. Unfortunately, one of the problems with the popularity of web-based email accounts is that using PGP or a similar equivalent scheme becomes impossible for most email users:(
Of course, all this stuff is nothing new, is it? Hasn't echelon been around for years, scanning emails?
Python gives you the power and expressiveness of perl, but with actual design principles behind it.
Just as Perl gives you the power and expressiveness of Python, but with more freedom to code the way you want, rather than the way some whitespace fanatic wanted?;-)
I don't think I've ever met someone who liked coding in both Python and Perl. Thankfully, having both around doing fabulously doesn't hurt anyone:)
It may not even be real. The sample may simply be skewed for some reason.
Well, it probably is skewed: there'll be a lot less academic usage this time of year, and if academics are adopting firefox faster than normal users (which you'd probably expect) then you'll see a dip over the Northern Hemisphere summer break.
Hiding a SSID in some ways is anti-social as it makes it more difficult for your neighbours to find your network if it interferes with their's.
What?!? All it does is remove the SSID information - you can still see that there's a network on a particular channel if you do a scan! You just can't connect to it...
If you can't rely on SSID broadcasts, then how can anyone know when a network is public and when it's private? Perhaps, as you say, there needs to be a more unambiguous code for this...
Geeks need to get out of the habit of assuming that a default configuration amounts to "permission to use". It doesn't. Only permission to use is permission to use.
Except that SSID broadcast without encryption has been assumed to be permission to use, by hardware manufacturers and even by certain OS software companies! Perhaps that's unfortunate, and perhaps it shouldn't be like that. I take your point, though, that there's no formal definition that a broadcast SSID equates to an open network, and that it's wrong to assume that that's so.
As has been mentioned on/. on several times before when this particular case came up, this guy didn't accidentally or "automagically" attach to his neighbour's wifi network: he sat outside their house, in his car, and acted very suspiciously when they walked past (e.g. snapping his laptop shut). He'd been doing this over a three month period. To my mind his punishment was more a result of his behaviour than mere connection to some idiot's wide open wireless network.
And as has also been previously mentioned on/. this guy did nothing technically wrong. The wireless network was inviting him to connect to it, and he obliged. Or is it now illegal for me to use the wireless network at my university?? Perhaps I'll get arrested for suspiciously sitting in the library with my laptop open?? It's exactly the same situation: an unsecured access point is actively broadcasting it's presence and letting people know that it's available.
His behaviour was immoral - obviously, it's fairly unlikely that anyone is purposely sharing their wireless network - but not illegal. The fact that he could be punished for this just goes to prove what a bunch of IT-ignorant people make/apply the laws...
Yeah? You should see some of the emails I send my PhD supervisor...;)
Seriously, I agree completely with your second point - IM is useless time wasting chatter for the masses. Only email has the potential to be eloquent and erudite, simply because you have more time to consider what you're going to write.
Just imagine how much worse/. would be if it was IM based, rather than posting...
Nope. If I don't lock my door, it doesn't imply that I'm giving someone permission to walk through my house. Unless I tell you you can access my wireless network, you aren't authorized to use it.
You've just shown how little you know about wireless networks. By default, your wireless network is doing exactly that. Unless setup otherwise, a wireless network is broadcasting the fact that it is there and is open to all comers. It's like you standing by your open front door yelling out to the world "Hey, guys, come right on in, the water's fine!"
If you want only a certain subset of computers in this world to use it, then setup MAC address filtering - that's a bit like having a couple of hefty bouncers standing along side your door pushing people away. If you only want a certain subset of people to use it, then set up encryption - that's like having a lock on the door, and only people who have the key can enter. But in both of these cases you're still standing outside inviting anyone and everyone to come in.
However, if you don't want to yell out to the world to enter your house, the equivalent in wireless terms is to simply not broadcast your ESSID. It's not hard to do...
This type of law enforcement is ignorant, stupid and just plain wrong. The guys weren't doing anything illegal, even if it might have been immoral. A far better solution would be to let the guys keep on using the unsecured, open networks and publicise the fact. Maybe that way people might learn a bit about how to set up a wireless network correctly.
Think about it for a minute. You only have time to get a single version of the client ready so which OS do you support first? You could release a Linux or Mac client and reach a fraction of your users or a Windows client and reach a large majority. Hmmm, let's see...
Ever heard of Java?? Obviously not...
Let me guess - you were the guy our Gov't employed to write the eTax software, weren't you?
I understand the governments position in the issue, that by providing for windows they are providing for the larger portion of citizens with computers, whether we like it or not.
But you see, a government should aim to provide for all the people, not just a proportion of them, even if that proportion happens to be a majority. And - should a government decide to introduce an etax system - it should design a system that is as user-friendly and extensible as possible. Have you seen the crap that they use?? That's truly a system designed by the lowest bidder.
How hard would it have been to code in java? How obvious would that decision have been?? For that matter, why on earth wouldn't you have a server-side solution which any user can connect to from anywhere only needing a web-browser that supports HTTPS?
But no - our brilliant government decided to go with a solution that is limited to one OS only. Having a comment about using Windows emulator software just makes it worse, as it shows that they were aware of the fact that not everyone uses Windows, but decided to go with a Windows-only solution anyway.
Now I can finally spell check my reply. I do not need to highlight my blog and paste it into my open office writer to do the spell check then correct any misspelled words and do the reverse by pasting it back into this page.
Yeah, because nobody ever created a spellchecker extension for firefox before, such as SpellBound... lucky Google came to the rescue, huh?
And to think, there wasn't a dedicated Google search bar in firefox before, either...
I'm trying to find what year these cards initially launched.
Mid 90's, IIRC - my 1996 Cyrix bomb had a Diamond Stealth card in it. It wasn't any good, mind you - it was the only thing in that box that made the Cyrix processor look good in comparison...
I may be missing something here but what does it matter what he said? Bittorrent is what it is independently of his expressed beliefs. Is he going to be prosecuted for what he did or what he thought?
Especially when there's no link between the writing in question and bittorrent! Just because he says...
I further my goals with technology. I build systems to disseminate information, commit digital piracy, synthesize drugs, maintain untrusted contacts, purchase anonymously, and secure machines and homes. I release my code and writings freely, and publish all of my ideas early to make them unpatentable.
... that does not mean that bittorrent was designed to commit digital piracy!
Personally, I've never used bittorrent, but I've considered it several times - not because I pirate music or videos or software or anything else over the web: I don't! But rather because it's used as a legitmate distribution tool for so many large open-source products, especially distribution CDs/DVDs. I've never seen any other file sharing tool used like this, and surely that must count strongly in the software's favour.
Does anyone know how much bittorrent is used for illegitimate purposes? I'd be interested to know...
here it's highlighting/middle button, there it's ctrl-c/ctrl-v
In general, both functions will work with just about any X application that you're likely to use. I have yet to come across any modern app that doesn't support Ctrl-C/Ctrl-V... and why's that? Probably because every single current toolkit for X-windows has those shortcuts built in as standard...
The only hard example that you can give is image copy and paste, which is completely different to text copy&paste and which does not work at all under X-windows AFAIK. But that's probably because you'd need to have a default format for image data and I don't think anyone's bothered to worry about it. Maybe you could suggest that FreeDesktop.org look into it - it's more in their area. (Maybe they already are, for all I know...)
But I think most of us here would happily trade image copy&paste for middle-mouse paste any day of the week - I know I work with text far more than I do images!
If it's just a illusion caused by the brain, then howcome the moon is different sizes in photos too?
Has anyone ever taken photos of the moon near the horizon and in the centre of the sky with a digital camera and a tripod? I've always wondered if it's ever been done, to conclusively show whether or not the moon illusion is in fact an illusion...
Of course, I imagine that you're thinking of telephoto shots of the moon near the horizon, where the moon is extremely large as it's been taken at high magnification. Your brain, OTOH, doesn't see the surrounding landscape as magnified, it just assumes your point of reference is closer. Thus the telephoto trick.
GoboLinux tries to do something like this, albeit not with stow. It puts each package in its own directory, and then symlinks everything back into better recognised locations... one of the GoboLinux philosophies is that the file system itself is the package manager. I presume that to uninstall a package, one would essentially delete the directory, and then do a file system sweep every so often of orphaned symlinks.
Nice to know there's a sensible distro out there! I imagine that you need to run some tool - it's probably stow, anyway - that creates the symlinks to install. And so, just like stow, there'd be another command to remove the symlinks when uninstalling..
For whatever reason, the main reaction here seems to be rideculing the idea, claiming that package managers have already solved all those problems, and claiming that accessing everything through symlinks will slow the system down.
That's a pretty dumb argument - if there's a performance decrease at all it's going to be incredibly marginal and only affect application startup times. I doubt that you could even measure it.
I don't use it, and go for Debian instead, if only because I prefer to have the much bigger support community. I do use Stow a lot with accounts on other people's systems, when I want to build my own miniature installation hierarchy and need a simple way to organise it. What Gobo's playing with doesn't seem like such a bad idea to me.
I generally use a simple distro as a base and then build and stow all the important software I use. Mind you, I've just started playing around with Ubuntu and I'm rather taken with the simplicity of apt-get...:-) I only wish Debian had adopted a Stow approach to packages:-(
First, all programs get dumped together under/usr, making it nearly impossible to cleanly uninstall if it was compiled from source
Try GNU Stow - it solves this problem very nicely, keeping each software package in a separate directory and symlinking everything to the relevant PATH directories. Pity there isn't a linux distro that uses it, or even comes with it - it would make things *so* much easier...:-(
True, but under/etc they are placed almost randomly. If you don't know the exact name of the configuration file you need (which may or may not be under a subdirectory...), you're out of luck.
At the end of an application's man page, there should be a section entitled "Files" which lists the location and names of the configuration files. That's not too difficult, is it? And there's always find or grep if that fails.
Becuase someone posts a comment in source questioning something doesn't seem to me to be a problem, it's there for a reason... a good reason would be to make other hackers take a closer look at it and make decisions based on it.
Agreed. If not having questioning comments in the kernel source is a reason for using *BSD, then I'm sticking to Linux.
It's funny, isn't it - two articles on the Linux vs. BSD war in as many days, one an interview with Linus, one an interview with *BSD lead developers. Interesting that Linus chose to be diplomatic and praise BSD, whereas the BSD guys decided to take the flamebait route. Guess it says a lot about the respective maturities of the OSes...;-)
Slashdot Mirror
I used to be a Lilo fan, until I ran into trouble with it on my IBM laptop -- it took WAY longer at the part where it loads the kernel to memory than it had any business to, a stage that goes by so fast it's easy to miss on most computers.
... I just can't :(
That's funny - I had an almost identical problem using grub with my ASUS laptop (used to take yonks to load stage 1.5), which forced me to switch to lilo. Lilo's more painful (have to re-run it each time you install a new kernel, doesn't have a nice command-line option) but at least it works and works consistently.
I'd like to like grub
Because of your Google cookie, Google in theory knows exactly what you search for, what you shop for, who you IM, and your entire email records.
... mind you, I still use Gmail so I guess I'm not that paranoid just yet :-)
Which is why I search through google.com.au, not google.com, and I block cookies set by google.com.au.
It shocks me a bit that google likes to promote things like search history as a feature! Scary stuff
Who needs a 'partner'?
:) But love is far more than sex ...
Well, me for one, and I'm sure there's heaps of others out there too. I don't want kids, but I do want to share my life with someone. I most certainly don't need chicks for sexual gratification - I've got a hand and I'm not afraid to use it
That easy, are you sure? If the police can intercept your e-mail, then most likely it will become forbidden to encrypt it - or the allowed encryption level will be far too weak to be usable.
:(
The obvious way around this is for as many people as possible to start using encryption - once too many are doing it, the law becomes unenforcible. Unfortunately, one of the problems with the popularity of web-based email accounts is that using PGP or a similar equivalent scheme becomes impossible for most email users
Of course, all this stuff is nothing new, is it? Hasn't echelon been around for years, scanning emails?
I do both.
:) Sure, most are going to have to play with both, but it doesn't mean that you're going to enjoy it!
Note: I said like, not do
Python gives you the power and expressiveness of perl, but with actual design principles behind it.
;-)
:)
Just as Perl gives you the power and expressiveness of Python, but with more freedom to code the way you want, rather than the way some whitespace fanatic wanted?
I don't think I've ever met someone who liked coding in both Python and Perl. Thankfully, having both around doing fabulously doesn't hurt anyone
It may not even be real. The sample may simply be skewed for some reason.
Well, it probably is skewed: there'll be a lot less academic usage this time of year, and if academics are adopting firefox faster than normal users (which you'd probably expect) then you'll see a dip over the Northern Hemisphere summer break.
Hiding a SSID in some ways is anti-social as it makes it more difficult for your neighbours to find your network if it interferes with their's.
...
...
What?!? All it does is remove the SSID information - you can still see that there's a network on a particular channel if you do a scan! You just can't connect to it
If you can't rely on SSID broadcasts, then how can anyone know when a network is public and when it's private? Perhaps, as you say, there needs to be a more unambiguous code for this
Geeks need to get out of the habit of assuming that a default configuration amounts to "permission to use". It doesn't. Only permission to use is permission to use.
Except that SSID broadcast without encryption has been assumed to be permission to use, by hardware manufacturers and even by certain OS software companies! Perhaps that's unfortunate, and perhaps it shouldn't be like that. I take your point, though, that there's no formal definition that a broadcast SSID equates to an open network, and that it's wrong to assume that that's so.
As has been mentioned on /. on several times before when this particular case came up, this guy didn't accidentally or "automagically" attach to his neighbour's wifi network: he sat outside their house, in his car, and acted very suspiciously when they walked past (e.g. snapping his laptop shut). He'd been doing this over a three month period. To my mind his punishment was more a result of his behaviour than mere connection to some idiot's wide open wireless network.
/. this guy did nothing technically wrong. The wireless network was inviting him to connect to it, and he obliged. Or is it now illegal for me to use the wireless network at my university?? Perhaps I'll get arrested for suspiciously sitting in the library with my laptop open?? It's exactly the same situation: an unsecured access point is actively broadcasting it's presence and letting people know that it's available.
...
And as has also been previously mentioned on
His behaviour was immoral - obviously, it's fairly unlikely that anyone is purposely sharing their wireless network - but not illegal. The fact that he could be punished for this just goes to prove what a bunch of IT-ignorant people make/apply the laws
After glancing over the screenshots and reading some of the comments the author had, the appearance to firefox is remarkable.
Well, sure, but can your great big so-called fire-thingy install spyware for you, automatically, without you even noticing, huh?
Beat that, you Open Source geeks - only IE7 is fully compatible with spyware straight out of the box!
It is hard to write an email about nothing.
... ;)
/. would be if it was IM based, rather than posting ...
Yeah? You should see some of the emails I send my PhD supervisor
Seriously, I agree completely with your second point - IM is useless time wasting chatter for the masses. Only email has the potential to be eloquent and erudite, simply because you have more time to consider what you're going to write.
Just imagine how much worse
Nope. If I don't lock my door, it doesn't imply that I'm giving someone permission to walk through my house. Unless I tell you you can access my wireless network, you aren't authorized to use it.
...
You've just shown how little you know about wireless networks. By default, your wireless network is doing exactly that. Unless setup otherwise, a wireless network is broadcasting the fact that it is there and is open to all comers. It's like you standing by your open front door yelling out to the world "Hey, guys, come right on in, the water's fine!"
If you want only a certain subset of computers in this world to use it, then setup MAC address filtering - that's a bit like having a couple of hefty bouncers standing along side your door pushing people away. If you only want a certain subset of people to use it, then set up encryption - that's like having a lock on the door, and only people who have the key can enter. But in both of these cases you're still standing outside inviting anyone and everyone to come in.
However, if you don't want to yell out to the world to enter your house, the equivalent in wireless terms is to simply not broadcast your ESSID. It's not hard to do
This type of law enforcement is ignorant, stupid and just plain wrong. The guys weren't doing anything illegal, even if it might have been immoral. A far better solution would be to let the guys keep on using the unsecured, open networks and publicise the fact. Maybe that way people might learn a bit about how to set up a wireless network correctly.
The Longhorn GUI will be based on IE, meaning that there will be zero difference between finding a directory and typing in a website.
... how, exactly? explorer.exe == iexplore.exe, and has been since Windows 98, iirc.
As opposed to the current situation with Windows
Think about it for a minute. You only have time to get a single version of the client ready so which OS do you support first? You could release a Linux or Mac client and reach a fraction of your users or a Windows client and reach a large majority. Hmmm, let's see...
...
Ever heard of Java?? Obviously not
Let me guess - you were the guy our Gov't employed to write the eTax software, weren't you?
I understand the governments position in the issue, that by providing for windows they are providing for the larger portion of citizens with computers, whether we like it or not.
But you see, a government should aim to provide for all the people, not just a proportion of them, even if that proportion happens to be a majority. And - should a government decide to introduce an etax system - it should design a system that is as user-friendly and extensible as possible. Have you seen the crap that they use?? That's truly a system designed by the lowest bidder.
How hard would it have been to code in java? How obvious would that decision have been?? For that matter, why on earth wouldn't you have a server-side solution which any user can connect to from anywhere only needing a web-browser that supports HTTPS?
But no - our brilliant government decided to go with a solution that is limited to one OS only. Having a comment about using Windows emulator software just makes it worse, as it shows that they were aware of the fact that not everyone uses Windows, but decided to go with a Windows-only solution anyway.
like typing slashdot and then Shift-Enter (for .net) instead of Ctrl-Shift Enter (for .org) and go to some other site
Eh? Does anyone actually use these shortcuts?? Why not just type the first few letters and then tab complete it?
Now I can finally spell check my reply. I do not need to highlight my blog and paste it into my open office writer to do the spell check then correct any misspelled words and do the reverse by pasting it back into this page.
... lucky Google came to the rescue, huh?
...
Yeah, because nobody ever created a spellchecker extension for firefox before, such as SpellBound
And to think, there wasn't a dedicated Google search bar in firefox before, either
I'm trying to find what year these cards initially launched.
...
Mid 90's, IIRC - my 1996 Cyrix bomb had a Diamond Stealth card in it. It wasn't any good, mind you - it was the only thing in that box that made the Cyrix processor look good in comparison
Especially when there's no link between the writing in question and bittorrent! Just because he says
Personally, I've never used bittorrent, but I've considered it several times - not because I pirate music or videos or software or anything else over the web: I don't! But rather because it's used as a legitmate distribution tool for so many large open-source products, especially distribution CDs/DVDs. I've never seen any other file sharing tool used like this, and surely that must count strongly in the software's favour.
Does anyone know how much bittorrent is used for illegitimate purposes? I'd be interested to know
here it's highlighting/middle button, there it's ctrl-c/ctrl-v
... and why's that? Probably because every single current toolkit for X-windows has those shortcuts built in as standard ...
...)
In general, both functions will work with just about any X application that you're likely to use. I have yet to come across any modern app that doesn't support Ctrl-C/Ctrl-V
The only hard example that you can give is image copy and paste, which is completely different to text copy&paste and which does not work at all under X-windows AFAIK. But that's probably because you'd need to have a default format for image data and I don't think anyone's bothered to worry about it. Maybe you could suggest that FreeDesktop.org look into it - it's more in their area. (Maybe they already are, for all I know
But I think most of us here would happily trade image copy&paste for middle-mouse paste any day of the week - I know I work with text far more than I do images!
It is like NASA World Wind, but better.
...
Except for one tiny little issue that I'm surprised no one on this site has picked up on: World Wind is open source, whereas Google Earth is not.
I've been playing around with World Wind for a while, and it's just fantastic stuff. I'll take the OSS solution any day
If it's just a illusion caused by the brain, then howcome the moon is different sizes in photos too?
...
Has anyone ever taken photos of the moon near the horizon and in the centre of the sky with a digital camera and a tripod? I've always wondered if it's ever been done, to conclusively show whether or not the moon illusion is in fact an illusion
Of course, I imagine that you're thinking of telephoto shots of the moon near the horizon, where the moon is extremely large as it's been taken at high magnification. Your brain, OTOH, doesn't see the surrounding landscape as magnified, it just assumes your point of reference is closer. Thus the telephoto trick.
GoboLinux tries to do something like this, albeit not with stow. It puts each package in its own directory, and then symlinks everything back into better recognised locations... one of the GoboLinux philosophies is that the file system itself is the package manager. I presume that to uninstall a package, one would essentially delete the directory, and then do a file system sweep every so often of orphaned symlinks.
..
... :-) I only wish Debian had adopted a Stow approach to packages :-(
Nice to know there's a sensible distro out there! I imagine that you need to run some tool - it's probably stow, anyway - that creates the symlinks to install. And so, just like stow, there'd be another command to remove the symlinks when uninstalling
For whatever reason, the main reaction here seems to be rideculing the idea, claiming that package managers have already solved all those problems, and claiming that accessing everything through symlinks will slow the system down.
That's a pretty dumb argument - if there's a performance decrease at all it's going to be incredibly marginal and only affect application startup times. I doubt that you could even measure it.
I don't use it, and go for Debian instead, if only because I prefer to have the much bigger support community. I do use Stow a lot with accounts on other people's systems, when I want to build my own miniature installation hierarchy and need a simple way to organise it. What Gobo's playing with doesn't seem like such a bad idea to me.
I generally use a simple distro as a base and then build and stow all the important software I use. Mind you, I've just started playing around with Ubuntu and I'm rather taken with the simplicity of apt-get
First, all programs get dumped together under /usr, making it nearly impossible to cleanly uninstall if it was compiled from source
... :-(
/etc they are placed almost randomly. If you don't know the exact name of the configuration file you need (which may or may not be under a subdirectory...), you're out of luck.
Try GNU Stow - it solves this problem very nicely, keeping each software package in a separate directory and symlinking everything to the relevant PATH directories. Pity there isn't a linux distro that uses it, or even comes with it - it would make things *so* much easier
True, but under
At the end of an application's man page, there should be a section entitled "Files" which lists the location and names of the configuration files. That's not too difficult, is it? And there's always find or grep if that fails.
Becuase someone posts a comment in source questioning something doesn't seem to me to be a problem, it's there for a reason... a good reason would be to make other hackers take a closer look at it and make decisions based on it.
... ;-)
Agreed. If not having questioning comments in the kernel source is a reason for using *BSD, then I'm sticking to Linux.
It's funny, isn't it - two articles on the Linux vs. BSD war in as many days, one an interview with Linus, one an interview with *BSD lead developers. Interesting that Linus chose to be diplomatic and praise BSD, whereas the BSD guys decided to take the flamebait route. Guess it says a lot about the respective maturities of the OSes