Problem: if you write your C# compiler (let's call it csharpc) in C#, how do you compile csharpc itself? You can't use csharpc:-)
You have to use the Microsoft compiler (from Visual Studio.Net) to compile a version of csharpc which will run under the Windows C# interpreter, then somehow make csharpc output Linux native code (because it can't output.Net bytecode and expect that to run on Linux until there's something to compile the interpreter for Linux, and we don't have a Linux version of csharpc to do that with yet). That's messy, to say the least.
On the other hand, if you're DotGNU and write your C# compiler in C, you can use existing C compilers like gcc to compile csharpc. Once that's done you can write the interpreter and runtime libraries in C# if you like, since you now have a Linux copy of csharpc to compile them with.
(Replace Linux with your non-Windows OS of choice)
It'd seem sensible for Mono to use DotGNU's compiler to bootstrap stuff now it's working, although there are probably obscure reasons why they can't.
Note the.gov.uk domain; that page is a quick summary of British data protection law. This is Britain's implementation of a European Union law (I posted the British one because it's in English:-)
Theft of something as insubstantial as bandwidth and CPU time is difficult to build a case around, but what would happen to spammers if the USA had this sort of law? Never mind the spam, they obviously have a large pile of personally identifiable information - if selling your CDs of e-mail addresses is illegal (because they're being used for purposes other than the one they were collected for), there goes the address sharing for a start.
Now, I'm not American, and not a lawyer... but aren't copyright and trade secrets different concepts?
I thought the idea was that copyright prevented people from copying things you've published, trade secrets prevent your employees leaking things you haven't published, and it's only the employee's fault (i.e. not the recipient's) when it gets leaked?
So, um, if you had a Numeric Decade Trade Secrets Act:-) that'd possibly be relevant, but I fail to see where the DMCA comes into this.
I was wondering whether to get a second domain (hey,.uk domains are cheap:-) to do this sort of thing with. However:
- Do you trust your friends and family not to give your e-mail address to other less trusted friends?
- Do you trust your friends and family not to put you in the To: or Cc: list of a mail going to several less trusted people?
- Do you trust your friends and family not to forward mails you sent them, or multi-recipient mails others sent them that also went to you, with your address still visible?
- Do you trust your friends and family not to get Klez and pass your address on to just about anyone?
For me, until I convince more people that they're doing things wrong, the answers to all these are "no".
Having said that, I'm pretty much doing this already; I get a small amount of spam to my main address, but I don't think it gets harvested often, since I've managed to remove it from most web pages (at least the ones Google finds).
That was because his contract with his employer had a "all your copyright are belong to us" clause in it, so he'd been open-sourcing stuff he didn't actually hold the copyright on.
If this anonymous poster's contract isn't like that (at my last job anything I wrote in my own time that was related to my job ended up under my employer's copyright, which isn't as bad) then his/her employer can't claim ownership of the code, or of the employee's free time.
To the article poster: read your contract, carefully.
Nope, you wouldn't get energy out
on
Science Askew
·
· Score: 2
now.. if we can figure a way of generating power from these spinning cats...
According to the laws of thermodynamics, that can't happen (perpetual motion and all that).
(The reason for this, of course, is that the buttered-toast effect is stronger on an expensive carpet; as the carpet gets covered in cat hair and toast crumbs, its value decreases until the buttered side of the toast isn't attracted enough. So you have to put energy in by cleaning the carpet, so there is no net energy gain:-)
(Sorry about the cryptic subject line, there's not space)
ZoneAlarm's niche doesn't seem to exist on Linux. The assumption is that you just don't run programs you don't trust - if you have firewall-config access, a sufficiently malicious program can always reconfigure it anyway (feeding keystrokes to your logged-in-as-root terminal? inserting a trojaned su or sudo binary into your $PATH?) and presumably the idea is that if a solution is fundamentally flawed, it's not worth implementing in the first place.
Yes, in the Real World(tm) where companies are willing to be extremely unethical but unwilling to actually break the law or suffer the backlash from taking over people's computers, ZoneAlarm has its uses, but you can't really rely on it that heavily.
iptables on the client doesn't have any specific protection against malicious apps, but you can constrain individual users' network access, and if you're running programs you're that paranoid about, you should probably be using a separate user ID for them anyway. (I keep meaning to set up one or more separate uids for WINEified games).
Incidentally, I've heard Explorer/Internet Explorer is a bad thing to run under WINE, since it has been known to damage the fake Windows folder WINE uses (that, and it probably uses more undocumented API calls than most third-party Windows apps). Anyone care to confirm or deny this?
Everyone on the university network here has a full Internet connection, apart from the Windows Networking (NetBIOS/SMB/CIFS) ports, which work locally but are firewalled at the edge of the university. There's no data rate limiting other than the limitations of the hardware (my college was wired up with 10MBit hubs last year, but they've upgraded to 100MBit switches this year; I've had the full 100MBit download rate while ftp'ing Mandrake ISOs from another college's mirror, so there's certainly no artificial cap there).
If anyone uses significant amounts of bandwidth (there's no formal limit, but it seems to be measured in gigabytes a day), they're told to reduce that for the benefit of other users (on a "please stop before we have to force you to" basis).
This is great, because when you want to download something big (a CD image for instance), you get a huge data rate and don't have to wait long, but the network admins can still prevent people from downloading stuff constantly and overloading the network.
I suppose a more automated equivalent would be to give everyone the full 10/100 bandwidth to start with, then automagically reduce priority for people who've used too much in the last week/month/whatever.
Until about a month or two ago, I thought laptops were pretty much the ideal desktop machine if you could afford them - OK, their expandability tends to suck, but they're nearly as fast as desktops, quiet, have un-flickery screens etc., and they're portable into the bargain.
About a month or two ago, I joined the semi-official student tech support team at my college, and I've dealt with two overheating laptops since then. Is it just bad luck, or do modern laptops have trouble running their CPUs at full speed?
When gcc 3.2 becomes the default. gcc 3.2 breaks binary compatibility for C++, KDE is C++, so the maintainers don't want to have to re-release KDE 3 to be compatible with gcc3.2-compiled libraries if they could just delay it a bit and upload a working version.
By the sound of it, sarge (the next stable version) will have dselect as an optional alternative package and aptitude as the one you get on installation.
"Oh, IE is attempting to send 5374 mail messages!"
(use of IE in this message may not be the best example program, but I'm continuing your example)
Yes, and there's no reason we can't check for that now (no extra tech required, I would have thought something like WINE or (for ultimate paranoia) a pure x86 emulator like Bochs could sandbox things nicely) and that's fine for a pre-"Trusted" version.
However, when you get IE 2010 with "Trusted" Digital Restriction Mechanisms, will you be able to run that in a sandbox? Or will it insist on having verifiable direct hardware access so it can talk to (say) your anti-piracy, screenshots-can-be-disallowed graphics card? (OK, so Windows Media Player 2010 might have been a better example)
When my CD-burner fails in Linux (I don't entirely blame it, I'm using cheap no-name media), cdrecord will sometimes complain "scsi sendcmd: [...some hex...]: no error" as it spits out a coaster.
The BBC Micro had some great (if rather terse) error messages, like "Mistake" (triggered by BASIC it couldn't parse) and "Silly" (triggered by renumbering lines and requesting a gap between line numbers that wasn't between 1 and 255).
I was always zealous about the Mac because it pretty much worked as advertised. If that ain't the kinda shit you preach from the mountain tops, I don't know what is.
It's disturbing that the computer industry seems to have convinced most people that working as advertised is exceptionally good. For most products, working as advertised is merely acceptable.
Then again, I suppose MS and Apple both advertise their OS as easy to use, and Apple is just the one that can back it up with actual usability:-)
[ObDisclaimer: I'm not a Mac user, I use Debian and Windows]
You get annoyed and click the unsubscribe link (if you're stupid enough), they know your e-mail address is actually read by a real person, they sell your e-mail address on one of those CDs a fair proportion of spam advertises, profit!!!.
If it's HTML mail with images in, and your mail client knows too much HTML (::coughOutlookcough::), downloading the image will confirm that your e-mail address is real, if the spam mail's set up right.
My current cooling (from quietpc's UK site and overclockers.co.uk) is:
- Athlon 1.4: Zalman flower, 92mm fan @ approx 5V, 80mm case fan (right next to the CPU) designed for 12V but connected to the PSU's 5V line - Geforce 2 Pro: huge Zalman heatsink (occupies the top PCI slot!), spare 80mm case fan mounted in the general vicinity (again, designed for 12V and running on 5V) - Northbridge: Zalman heatsink - PSU: the silent 300W one from quietpc
The whole system seems stable (although it gets rather warm with the case-fans running that slowly), and the noisiest components are the hard disks:-)
If you think that, fair enough. Your words, my emphasis:
And part of providing a good moral foundation is preventing your children from playing with immoral toys.
Indeed, I wouldn't give GTA3 or most FPSs to an impressionable child. (possibly making an exception for Nerf Arena:-)
On the other hand, I have no problem with someone who can tell the difference between reality and a game playing violent games.
(Disclaimer: I enjoy violent games. I'm also non-violent IRL to the extent that I avoid killing insects. Draw your own conclusions.)
Actually, there is a Linux OS that's not GNU/Linux
on
Roll Your Own Browser
·
· Score: 2, Interesting
From the FAQ for "Revol", a distribution of Linux suitable for use on a Psion Revo (a.k.a. Diamond Mako) electronic organiser:
Shouldnt this be called GNU/Linux?
Actually, no. The argument for GNU/Linux is that most linux systems are a modified version of the GNU system which has been around for longer than linux has. However, Revol uses embedded versions of the standard parts of the operating system normally provided by GNU tools (uclibc instead of glibc, busybox instead of the GNU fileutils etc). So Revol is a non-GNU linux system.
The GNU GPL forces you to GPL anything you derive from it. That's what the authors of GPLed software chose. Their decision. If you don't like that, don't use GPLed libraries. [Example: if you link with GNU readline (like bash does), the program linked to it has to be GPLed.]
The LGPL forces you to GPL or LGPL changes to LGPLed stuff, but you can link anything you like to it. [Example: GNU libc, the standard GNU/Linux C library, is LGPLed. If you make your own C library based on glibc, you have to place that under the LGPL, but if you just link a proprietary product like Netscape Navigator with glibc, you can keep it as proprietary as you like.]
Just using a GPLed/LGPLed program isn't a problem. You can write proprietary software in a GPLed text editor (e.g. GNU Emacs) and compile it with a GPLed C compiler (e.g. gcc), and you're not required to open-source it at all.
Most open-source licenses are considerably less restrictive than the GPL and you can incorporate them into proprietary software quite happily. A couple of examples:
- Xiphophorus use a BSD-style license for libogg, libvorbis and libvorbisfile (the Ogg Vorbis libraries). Unreal Tournament 2003 uses Ogg Vorbis for its in-game music, and sure enough, its System folder contains ogg.dll, vorbis.dll and vorbisfile.dll. Hmm, I wonder what those could be? Epic Games are entirely free to do that; it would also be legal for them to compile the same code into UT2003.exe, if they so wished. (They don't, because the command-line dedicated server, a separate executable, also needs to access them for some obscure technical reason.)
- Many, many commercial products incorporate zlib.
If an author decides to open-source their code, they get to choose which is more important to them - ensuring that users' freedom is preserved in the license of modified/derived versions (the GPL is a good choice if that's important), or giving proprietary software authors the freedom to incorporate your code in their work (the zlib license is pretty good if that's important).
The LGPL is actually a pretty good compromise, although its originators (the Free Software Foundation) discourage its use, because they value the first of the freedoms I mentioned over the second.
The GPL is only a subset of open source.
The GNU GPL forces you to GPL anything you derive from it. That's what the authors of GPLed software chose. Their decision. If you don't like that, don't use GPLed libraries.
[Example: if you link with GNU readline (like bash does), the program linked to it has to be GPLed.]
The LGPL forces you to GPL or LGPL changes to LGPLed stuff, but you can link anything you like to it.
[Example: GNU libc, the standard GNU/Linux C library, is LGPLed. If you make your own C library based on glibc, you have to place that under the LGPL, but if you just link a proprietary product like Netscape Navigator with glibc, you can keep it as proprietary as you like.]
Just using a GPLed/LGPLed program isn't a problem. You can write proprietary software in a GPLed text editor (e.g. GNU Emacs) and compile it with a GPLed C compiler (e.g. gcc), and you're not required to open-source it at all.
Most open-source licenses are considerably less restrictive than the GPL and you can incorporate them into proprietary software quite happily. A couple of examples:
- Xiphophorus use a BSD-style license for libogg, libvorbis and libvorbisfile (the Ogg Vorbis libraries). Unreal Tournament 2003 uses Ogg Vorbis for its in-game music, and sure enough, its System folder contains ogg.dll, vorbis.dll and vorbisfile.dll. Hmm, I wonder what those could be? Epic Games are entirely free to do that; it would also be legal for them to compile the same code into UT2003.exe, if they so wished. (They don't, because the command-line dedicated server, a separate executable, also needs to access them for some obscure technical reason.)
- Many, many commercial products incorporate zlib.
If an author decides to open-source their code, they get to choose which is more important to them - ensuring that users' freedom is preserved in the license of modified/derived versions (the GPL is a good choice if that's important), or giving proprietary software authors the freedom to incorporate your code in their work (the zlib license is pretty good if that's important).
The LGPL is actually a pretty good compromise, although its originators (the Free Software Foundation) discourage its use, because they value the first of the freedoms I mentioned over the second.
If I remember correctly, the original Unreal Tournament release (or possibly it was just the demo) was fairly glitchy on everything except Glide; it was obvious what sort of video cards the staff at Epic had:-)
The later patches upgraded it to good Glide and D3D, and reputed-to-be-slightly-wobbly OpenGL (which wasn't officially supported); once Loki Games made progress with the Linux port, Epic released an improved GL driver heavily based on Loki's work! (obviously, non-3dfx cards on Linux don't have the option of D3D, so Loki had a rather large incentive to make a decent OpenGL driver.)
If UT2003's anything like UT, the graphics drivers will work as a swappable plugin system. UT has software, "pure" OpenGL, Glide, D3D and S3 Metal on Windows; software, Glide, "pure" OpenGL and SDL OpenGL on Linux; and presumably much the same options as Linux on MacOS, although possibly without the SDL GL driver.
Incidentally, SDL is the default choice of renderer for new Linux UT installs, and all you open-source game hackers have Loki Games (they of the Linux UT port) to thank for SDL - they developed it specifically to make Windows to Linux ports easier.
Back on topic, this is great news for the state of Linux gaming, and Linux mass-marketness in general; and purely from my own point of view, I was wondering whether to bother with UT2003 before, and I probably will now. ++EpicSales:-)
I'm currently running Gaim (a Linux/X/Gtk client originally developed for AIM compatibility) which uses a plugin system to connect to several competing IM servers (ICQ, AIM, MSN, Napster etc.) - as far as I can see, Trillian is its Windows equivalent (I don't know which came first).
The Gaim developers have stopped working on their ICQ plugin, because the same protocol ("Oscar") and server (login.oscar.aol.com port 5190) will work for both services, and their AIM plugin has expanded to have full ICQ functionality - you just fill in an ICQ number and password rather than an AIM screenname and password.
AIM and ICQ still don't seem to interoperate - I'm not sure whether this is a Gaim-ism or an AOL problem, but sending a message from my ICQ account to my AIM account (or vice versa) fails.
SFS (the Self-Certifying Filesystem, http://www.fs.net) is like NFS, but encrypted and more secure. I use it to mount/home from a server on another computer, and it seems to work (home directories in a SFS share are a little unreliable, though, so I'm looking for an alternative). I'll probably stick with it for my "files" (music, archived downloads, installers, etc.) directory though.
If Lucas wanted to make more films, it'd be difficult to fit anything less than 15-20 years after SW4:ANH without completely breaking continuity with the books.
Having said that, some of the books could make great films; Heir to the Empire/Dark Force Rising/Last Command, anyone? (They've already made pretty good graphic novels, if the one I read is anything to go by). They introduce some great characters (former Imperial assassin Mara Jade, and of course the inimitable Grand Admiral Thrawn), and even portray Imperial leaders in a more ambiguous light than in the films (particularly Thrawn and his second-in-command Pellaeon). Outside that trilogy, I, Jedi also rules.
Slashdot Mirror
Problem: if you write your C# compiler (let's call it csharpc) in C#, how do you compile csharpc itself? You can't use csharpc :-)
.Net) to compile a version of csharpc which will run under the Windows C# interpreter, then somehow make csharpc output Linux native code (because it can't output .Net bytecode and expect that to run on Linux until there's something to compile the interpreter for Linux, and we don't have a Linux version of csharpc to do that with yet). That's messy, to say the least.
You have to use the Microsoft compiler (from Visual Studio
On the other hand, if you're DotGNU and write your C# compiler in C, you can use existing C compilers like gcc to compile csharpc. Once that's done you can write the interpreter and runtime libraries in C# if you like, since you now have a Linux copy of csharpc to compile them with.
(Replace Linux with your non-Windows OS of choice)
It'd seem sensible for Mono to use DotGNU's compiler to bootstrap stuff now it's working, although there are probably obscure reasons why they can't.
You don't see a whole lot of European spam, do you? This sort of thing could be why:
.gov.uk domain; that page is a quick summary of British data protection law. This is Britain's implementation of a European Union law (I posted the British one because it's in English :-)
http://www.dataprotection.gov.uk/principl.htm
Note the
Theft of something as insubstantial as bandwidth and CPU time is difficult to build a case around, but what would happen to spammers if the USA had this sort of law? Never mind the spam, they obviously have a large pile of personally identifiable information - if selling your CDs of e-mail addresses is illegal (because they're being used for purposes other than the one they were collected for), there goes the address sharing for a start.
Now, I'm not American, and not a lawyer... but aren't copyright and trade secrets different concepts?
:-) that'd possibly be relevant, but I fail to see where the DMCA comes into this.
I thought the idea was that copyright prevented people from copying things you've published, trade secrets prevent your employees leaking things you haven't published, and it's only the employee's fault (i.e. not the recipient's) when it gets leaked?
So, um, if you had a Numeric Decade Trade Secrets Act
I was wondering whether to get a second domain (hey, .uk domains are cheap :-) to do this sort of thing with. However:
- Do you trust your friends and family not to give your e-mail address to other less trusted friends?
- Do you trust your friends and family not to put you in the To: or Cc: list of a mail going to several less trusted people?
- Do you trust your friends and family not to forward mails you sent them, or multi-recipient mails others sent them that also went to you, with your address still visible?
- Do you trust your friends and family not to get Klez and pass your address on to just about anyone?
For me, until I convince more people that they're doing things wrong, the answers to all these are "no".
Having said that, I'm pretty much doing this already; I get a small amount of spam to my main address, but I don't think it gets harvested often, since I've managed to remove it from most web pages (at least the ones Google finds).
That was because his contract with his employer had a "all your copyright are belong to us" clause in it, so he'd been open-sourcing stuff he didn't actually hold the copyright on.
If this anonymous poster's contract isn't like that (at my last job anything I wrote in my own time that was related to my job ended up under my employer's copyright, which isn't as bad) then his/her employer can't claim ownership of the code, or of the employee's free time.
To the article poster: read your contract, carefully.
now .. if we can figure a way of generating power from these spinning cats...
:-)
According to the laws of thermodynamics, that can't happen (perpetual motion and all that).
(The reason for this, of course, is that the buttered-toast effect is stronger on an expensive carpet; as the carpet gets covered in cat hair and toast crumbs, its value decreases until the buttered side of the toast isn't attracted enough. So you have to put energy in by cleaning the carpet, so there is no net energy gain
(Sorry about the cryptic subject line, there's not space)
ZoneAlarm's niche doesn't seem to exist on Linux. The assumption is that you just don't run programs you don't trust - if you have firewall-config access, a sufficiently malicious program can always reconfigure it anyway (feeding keystrokes to your logged-in-as-root terminal? inserting a trojaned su or sudo binary into your $PATH?) and presumably the idea is that if a solution is fundamentally flawed, it's not worth implementing in the first place.
Yes, in the Real World(tm) where companies are willing to be extremely unethical but unwilling to actually break the law or suffer the backlash from taking over people's computers, ZoneAlarm has its uses, but you can't really rely on it that heavily.
iptables on the client doesn't have any specific protection against malicious apps, but you can constrain individual users' network access, and if you're running programs you're that paranoid about, you should probably be using a separate user ID for them anyway. (I keep meaning to set up one or more separate uids for WINEified games).
Incidentally, I've heard Explorer/Internet Explorer is a bad thing to run under WINE, since it has been known to damage the fake Windows folder WINE uses (that, and it probably uses more undocumented API calls than most third-party Windows apps). Anyone care to confirm or deny this?
Everyone on the university network here has a full Internet connection, apart from the Windows Networking (NetBIOS/SMB/CIFS) ports, which work locally but are firewalled at the edge of the university. There's no data rate limiting other than the limitations of the hardware (my college was wired up with 10MBit hubs last year, but they've upgraded to 100MBit switches this year; I've had the full 100MBit download rate while ftp'ing Mandrake ISOs from another college's mirror, so there's certainly no artificial cap there).
If anyone uses significant amounts of bandwidth (there's no formal limit, but it seems to be measured in gigabytes a day), they're told to reduce that for the benefit of other users (on a "please stop before we have to force you to" basis).
This is great, because when you want to download something big (a CD image for instance), you get a huge data rate and don't have to wait long, but the network admins can still prevent people from downloading stuff constantly and overloading the network.
I suppose a more automated equivalent would be to give everyone the full 10/100 bandwidth to start with, then automagically reduce priority for people who've used too much in the last week/month/whatever.
Until about a month or two ago, I thought laptops were pretty much the ideal desktop machine if you could afford them - OK, their expandability tends to suck, but they're nearly as fast as desktops, quiet, have un-flickery screens etc., and they're portable into the bargain. About a month or two ago, I joined the semi-official student tech support team at my college, and I've dealt with two overheating laptops since then. Is it just bad luck, or do modern laptops have trouble running their CPUs at full speed?
when will KDE 3.x get into sid?
When gcc 3.2 becomes the default. gcc 3.2 breaks binary compatibility for C++, KDE is C++, so the maintainers don't want to have to re-release KDE 3 to be compatible with gcc3.2-compiled libraries if they could just delay it a bit and upload a working version.
By the sound of it, sarge (the next stable version) will have dselect as an optional alternative package and aptitude as the one you get on installation.
"Oh, IE is attempting to send 5374 mail messages!"
(use of IE in this message may not be the best example program, but I'm continuing your example)
Yes, and there's no reason we can't check for that now (no extra tech required, I would have thought something like WINE or (for ultimate paranoia) a pure x86 emulator like Bochs could sandbox things nicely) and that's fine for a pre-"Trusted" version.
However, when you get IE 2010 with "Trusted" Digital Restriction Mechanisms, will you be able to run that in a sandbox? Or will it insist on having verifiable direct hardware access so it can talk to (say) your anti-piracy, screenshots-can-be-disallowed graphics card? (OK, so Windows Media Player 2010 might have been a better example)
If Debian's apt-get can't find your nameserver (i.e. the network's down or your DNS daemon's died), it will still respond:
n trib/source/Release
Something wicked happened resolving 'ftp.uk.debian.org:http' (-3)
Failed to fetch http://ftp.uk.debian.org/debian/dists/unstable/co
Some others:
When my CD-burner fails in Linux (I don't entirely blame it, I'm using cheap no-name media), cdrecord will sometimes complain "scsi sendcmd: [...some hex...]: no error" as it spits out a coaster.
The BBC Micro had some great (if rather terse) error messages, like "Mistake" (triggered by BASIC it couldn't parse) and "Silly" (triggered by renumbering lines and requesting a gap between line numbers that wasn't between 1 and 255).
I was always zealous about the Mac because it pretty much worked as advertised. If that ain't the kinda shit you preach from the mountain tops, I don't know what is.
:-)
It's disturbing that the computer industry seems to have convinced most people that working as advertised is exceptionally good. For most products, working as advertised is merely acceptable.
Then again, I suppose MS and Apple both advertise their OS as easy to use, and Apple is just the one that can back it up with actual usability
[ObDisclaimer: I'm not a Mac user, I use Debian and Windows]
You get annoyed and click the unsubscribe link (if you're stupid enough), they know your e-mail address is actually read by a real person, they sell your e-mail address on one of those CDs a fair proportion of spam advertises, profit!!!.
If it's HTML mail with images in, and your mail client knows too much HTML (::coughOutlookcough::), downloading the image will confirm that your e-mail address is real, if the spam mail's set up right.
My current cooling (from quietpc's UK site and overclockers.co.uk) is:
:-)
- Athlon 1.4: Zalman flower, 92mm fan @ approx 5V, 80mm case fan (right next to the CPU) designed for 12V but connected to the PSU's 5V line
- Geforce 2 Pro: huge Zalman heatsink (occupies the top PCI slot!), spare 80mm case fan mounted in the general vicinity (again, designed for 12V and running on 5V)
- Northbridge: Zalman heatsink
- PSU: the silent 300W one from quietpc
The whole system seems stable (although it gets rather warm with the case-fans running that slowly), and the noisiest components are the hard disks
If you think that, fair enough. Your words, my emphasis:
:-)
And part of providing a good moral foundation is preventing your children from playing with immoral toys.
Indeed, I wouldn't give GTA3 or most FPSs to an impressionable child. (possibly making an exception for Nerf Arena
On the other hand, I have no problem with someone who can tell the difference between reality and a game playing violent games.
(Disclaimer: I enjoy violent games. I'm also non-violent IRL to the extent that I avoid killing insects. Draw your own conclusions.)
From the FAQ for "Revol", a distribution of Linux suitable for use on a Psion Revo (a.k.a. Diamond Mako) electronic organiser:
Shouldnt this be called GNU/Linux?
Actually, no. The argument for GNU/Linux is that most linux systems are a modified version of the GNU system which has been around for longer than linux has. However, Revol uses embedded versions of the standard parts of the operating system normally provided by GNU tools (uclibc instead of glibc, busybox instead of the GNU fileutils etc). So Revol is a non-GNU linux system.
The GPL is only a subset of open source.
The GNU GPL forces you to GPL anything you derive from it. That's what the authors of GPLed software chose. Their decision. If you don't like that, don't use GPLed libraries.
[Example: if you link with GNU readline (like bash does), the program linked to it has to be GPLed.]
The LGPL forces you to GPL or LGPL changes to LGPLed stuff, but you can link anything you like to it.
[Example: GNU libc, the standard GNU/Linux C library, is LGPLed. If you make your own C library based on glibc, you have to place that under the LGPL, but if you just link a proprietary product like Netscape Navigator with glibc, you can keep it as proprietary as you like.]
Just using a GPLed/LGPLed program isn't a problem. You can write proprietary software in a GPLed text editor (e.g. GNU Emacs) and compile it with a GPLed C compiler (e.g. gcc), and you're not required to open-source it at all.
Most open-source licenses are considerably less restrictive than the GPL and you can incorporate them into proprietary software quite happily. A couple of examples:
- Xiphophorus use a BSD-style license for libogg, libvorbis and libvorbisfile (the Ogg Vorbis libraries). Unreal Tournament 2003 uses Ogg Vorbis for its in-game music, and sure enough, its System folder contains ogg.dll, vorbis.dll and vorbisfile.dll. Hmm, I wonder what those could be? Epic Games are entirely free to do that; it would also be legal for them to compile the same code into UT2003.exe, if they so wished. (They don't, because the command-line dedicated server, a separate executable, also needs to access them for some obscure technical reason.)
- Many, many commercial products incorporate zlib.
If an author decides to open-source their code, they get to choose which is more important to them - ensuring that users' freedom is preserved in the license of modified/derived versions (the GPL is a good choice if that's important), or giving proprietary software authors the freedom to incorporate your code in their work (the zlib license is pretty good if that's important).
The LGPL is actually a pretty good compromise, although its originators (the Free Software Foundation) discourage its use, because they value the first of the freedoms I mentioned over the second.
The GPL is only a subset of open source. The GNU GPL forces you to GPL anything you derive from it. That's what the authors of GPLed software chose. Their decision. If you don't like that, don't use GPLed libraries. [Example: if you link with GNU readline (like bash does), the program linked to it has to be GPLed.] The LGPL forces you to GPL or LGPL changes to LGPLed stuff, but you can link anything you like to it. [Example: GNU libc, the standard GNU/Linux C library, is LGPLed. If you make your own C library based on glibc, you have to place that under the LGPL, but if you just link a proprietary product like Netscape Navigator with glibc, you can keep it as proprietary as you like.] Just using a GPLed/LGPLed program isn't a problem. You can write proprietary software in a GPLed text editor (e.g. GNU Emacs) and compile it with a GPLed C compiler (e.g. gcc), and you're not required to open-source it at all. Most open-source licenses are considerably less restrictive than the GPL and you can incorporate them into proprietary software quite happily. A couple of examples: - Xiphophorus use a BSD-style license for libogg, libvorbis and libvorbisfile (the Ogg Vorbis libraries). Unreal Tournament 2003 uses Ogg Vorbis for its in-game music, and sure enough, its System folder contains ogg.dll, vorbis.dll and vorbisfile.dll. Hmm, I wonder what those could be? Epic Games are entirely free to do that; it would also be legal for them to compile the same code into UT2003.exe, if they so wished. (They don't, because the command-line dedicated server, a separate executable, also needs to access them for some obscure technical reason.) - Many, many commercial products incorporate zlib. If an author decides to open-source their code, they get to choose which is more important to them - ensuring that users' freedom is preserved in the license of modified/derived versions (the GPL is a good choice if that's important), or giving proprietary software authors the freedom to incorporate your code in their work (the zlib license is pretty good if that's important). The LGPL is actually a pretty good compromise, although its originators (the Free Software Foundation) discourage its use, because they value the first of the freedoms I mentioned over the second.
dcgui, a.k.a. dc_gui (GTK+) and dctc (text-mode client, not designed for actual users - it's designed to provide a backend for things like dcgui).
:-)
For server aps there's dchub.
Debian users: "apt-get install dcgui dchub" should do it
If I remember correctly, the original Unreal Tournament release (or possibly it was just the demo) was fairly glitchy on everything except Glide; it was obvious what sort of video cards the staff at Epic had :-)
:-)
The later patches upgraded it to good Glide and D3D, and reputed-to-be-slightly-wobbly OpenGL (which wasn't officially supported); once Loki Games made progress with the Linux port, Epic released an improved GL driver heavily based on Loki's work! (obviously, non-3dfx cards on Linux don't have the option of D3D, so Loki had a rather large incentive to make a decent OpenGL driver.)
If UT2003's anything like UT, the graphics drivers will work as a swappable plugin system. UT has software, "pure" OpenGL, Glide, D3D and S3 Metal on Windows; software, Glide, "pure" OpenGL and SDL OpenGL on Linux; and presumably much the same options as Linux on MacOS, although possibly without the SDL GL driver.
Incidentally, SDL is the default choice of renderer for new Linux UT installs, and all you open-source game hackers have Loki Games (they of the Linux UT port) to thank for SDL - they developed it specifically to make Windows to Linux ports easier.
Back on topic, this is great news for the state of Linux gaming, and Linux mass-marketness in general; and purely from my own point of view, I was wondering whether to bother with UT2003 before, and I probably will now. ++EpicSales
-- Psychic_313, Unreal Tournament mod programmer
I'm currently running Gaim (a Linux/X/Gtk client originally developed for AIM compatibility) which uses a plugin system to connect to several competing IM servers (ICQ, AIM, MSN, Napster etc.) - as far as I can see, Trillian is its Windows equivalent (I don't know which came first).
The Gaim developers have stopped working on their ICQ plugin, because the same protocol ("Oscar") and server (login.oscar.aol.com port 5190) will work for both services, and their AIM plugin has expanded to have full ICQ functionality - you just fill in an ICQ number and password rather than an AIM screenname and password.
AIM and ICQ still don't seem to interoperate - I'm not sure whether this is a Gaim-ism or an AOL problem, but sending a message from my ICQ account to my AIM account (or vice versa) fails.
SFS (the Self-Certifying Filesystem, http://www.fs.net) is like NFS, but encrypted and more secure. I use it to mount /home from a server on another computer, and it seems to work (home directories in a SFS share are a little unreliable, though, so I'm looking for an alternative). I'll probably stick with it for my "files" (music, archived downloads, installers, etc.) directory though.
If Lucas wanted to make more films, it'd be difficult to fit anything less than 15-20 years after SW4:ANH without completely breaking continuity with the books.
Having said that, some of the books could make great films; Heir to the Empire/Dark Force Rising/Last Command, anyone? (They've already made pretty good graphic novels, if the one I read is anything to go by). They introduce some great characters (former Imperial assassin Mara Jade, and of course the inimitable Grand Admiral Thrawn), and even portray Imperial leaders in a more ambiguous light than in the films (particularly Thrawn and his second-in-command Pellaeon). Outside that trilogy, I, Jedi also rules.