The worm wouldn't be the cause of the deaths. It would be a symptom of the real cause - incompetence. Nobody should be installing Windows on these types of systems. Period. Even Microsoft will tell you that.
Hmm, I'm not sure of that. They claim their stuff is suited for mission critical systems. Though, if they ever sell a system, and that sale leads to human deaths because of flaws in it, even their eula isn't going to save them. That's manslaughter.
the "science proves the theory of evolution" statement is not accurate. I would be happy to answer any questions or continue this list.
How do you explain the remarkable genetic match between humans and chimpanzees, up to the point where diseases can make the leap between them and us? Was God just lazy and copy/pasting code?
See, the thing with evolution is that we don't have a complete map from the earliest life to any current species. Anti-evolutionists like to say "ha! no complete and fully detailed evidence means your theory is wrong!", but absence of evidence is not evidence in itself, and there is still a lot of evidence in support of evolution (or rather natural selection), though not necessarily related to humans.
Now, we can quibbly about the exact processes of evolution, but to deny that species evolve involves putting a very large blind spot in your science. Speciation is clear, and all around us. You can even make it happen experimentally in labs if you go through enough generations.
People working in an office (or even an overwhelming majority of home users) dont get paid to fuck with the source code, nor would most of them even want to. Only programmers care about that shit, and at least 99% of computer users are not programmers.
Yeah, and people not working in capitol hill don't get paid to fuck with laws, nor would most of them even want to. Only lawmakers care about that shit, and at least 99% of citizens are not lawmakers.
Just because you're not involved in creating something doesn't mean you shouldn't have input, especially if you're going to be the user of said product (which most computer users are).
Hmm, I wonder why you got modded up, this sounds very trollish to me. Lots of long complicated sentences to make a very basic point:
"In order for Java to become for Free Software and Gnu/Linux what VB became for Microsoft, Java has to be Freed and put out under the GPL."
Note that I used your exact words to not get accused of making a strawman argument here.
Your point is one that is deeply flawed. VB isn't GPL, yet it is ubiquitous. Why? Because it doesn't require you to know about programming. It doesn't require you to actually learn the trade. You can drag-and-drop and copy/paste your way to something that works. Sure, it's a really rotten environment to do any real development in, but for prototypes, or for people who don't want to learn a real language, it is excellent. Java, by it's compulsory OO nature could never be that. It's not "easy" enough.
Re:What the law says and what's done in practice .
on
Canadian Privacy Act
·
· Score: 2, Informative
here in Germany we have very tough laws with regard to your personal information and how it must be handled by businesses and the government
This is actually a EU directive. Or actually, two different ones. One dealing with regular privacy (enforced since 1998), and one with online privacy (enforced since last year). Seemingly when you read the text of the directive, it has a lot of teeth, but in practice they make exceptions every time someone asks. Like when the US insisted on having every bit of available information on EU citizens flying into the US (including the kind of meal they took, and how they paid for their ticket). The EU after some haggling made an exception that allows some, but not all, of the passenger information to pass to the US.
At least, a privacy law, even if it's not being enforced, is still better than no privacy law.
This is sort of off-topic, but relevant to the parent.
I'm looking for Geoworks to throw onto some 486's I want to bring back to life -- the last version I remember had a web-browser and everything!
If the 486's have at least 16 megs of ram, and are at least 33 mhz machines, you can put a stripped down debian on them with a custom compiled linux kernel, opera, abiword, and some really lightweight wm like pwm. It does work at a decent speed, but you have the advantage of using software updated for this millenium. If opera is too bloated as a webbrowser (only on low end 486's, since opera is ridiculously light) you could try dillo, which doesn't support all the standards, but is usable for webbrowsing.
I ran debian like that on a 486 DX4/100 with 20 megs of ram and an 800 meg hd for a long time. Worked great. Usable as a primary desktop. As long as you don't want to multitask ofcourse, but then on a 486 multitasking is pretty much out of the question anyway. That was with the regular debian Xfree, which I suppose could be seriously stripped down, but I stopped stripping down once I got the desktop functionality I wanted working at a decent speed.
A 386 can be made quite useful with linux if you stick to shell apps. There are a ton of great shell tools. Links or w3m for webbrowsing, latex for making pretty documents (you don't need to see the graphical output of something you do in latex until you print), any of the norton commander clones for a shell. Useful, even with only 8 megs of ram.
It's already there. It's just not in button form --
--max_upload_rate
maximum kB/s to upload at, 0 means no limit (defaults to 0) Setting that to 1 kB/s should be slow enough even for a modem user...
To those saying this speeds things down, it doesn't necessarily. On my cable modem connection I have an upload cap of 16K / sec, bittorrent quickly swamps that to such a degree that downloading drops to less than 20 K / sec. By setting the upload cap in bittorrent to 10K / sec the tcp/ip acks are still going through fast, and my dl speeds are 300K+.
How else does the Code Red author decide, "Hey! I found this buffer overflow routine in the unicode support for URLs in the IIS Indexing Server"?
Didn't code red take advantage from an already published security flaw? Most worms exploit known flaws for which patches are available.
Anyway, reverse engineering an entire operating system into something like source is quite difficult. If you don't have debug symbols you can't deduce from the name what each variable stores and what each function does. And assembly doesn't even have a concept of variables and functions (well, that last bit can be argued, but let's generalise). It's all byte manipulations and conditional jumps. Not easy to see what's happening, especially when you're talking about literally tens of megabytes of assembly. Blackbox testing (throwing random data into the input and seeing what the output does) is much more useful to find exploitable problems.
If CrossOver Office starts to bug me, then I can go to WineX and it will work about the same.
The focus of crossover office and winex is entirely different. Crossover office targets desktop apps, while winex targets games. As a result crossover office runs apps like office and photoshop much better, but winex supports directx 9 and runs the majority of popular games.
Ofcourse, improving the win32 api's to run either desktop apps or games will improve the ability to run the other category as well, which is why winex is reasonably good at running desktop apps and crossover office / regular wine is reasonably good at running games.
if a newbie can insert his windows app cd, run the installer under wine and have a application link inserted in a windows directory in the gnome menu that is already associated with wine and that app
Crossover office adds apps to the menu and desktop on my system (debian + kde). That's the diff between crossover and wine, you pay for the polish.
Re:Perhaps this is an improvement?
on
WineConf 2004 Wrapup
·
· Score: 2, Informative
man "How do I check my filesystem?"... not really
That's because you're using the wrong command:
joeri@angelina:~$ apropos check filesystem ... fsck (8) - check and repair a Linux file system fsck.ext2 (8) - check a Linux second extended file system fsck.ext3 (8) - check a Linux second extended file system fsck.minix (8) - a file system consistency checker for Linux ...
man is for when you already know what you're looking for. apropos is for when you want a man page of something, but you don't know what it should be.
Also you can repeaditly stream a text version of War and Peace or some other lengthy book, with a counter on the recieving end showing how many times you have downloaded it.
Why does everyone assume that causality can't be violated (in "small" ways)? I mean, there's no such thing as causality at the quantum level, is there?
If you receive an answer before you ask a question, will you still ask the question? If you don't, no answer will be returned to the unasked question, so you will have an answer that was never sent. This is impossible. Therefore, causality must be respected.
Now, I'm not saying it is impossible to send information back in time, just that it leads to all kinds of strange and disturbing problems if you assume that it is possible and that we have free will. One or the other must go, and I prefer for us to have free will.
Colonel Sandurz: Prepare ship for light speed. Dark Helmet: No, no, no, light speed is too slow. Colonel Sandurz: Light speed, too slow? Dark Helmet: Yes, we're gonna have to go right to ludicrous speed.
If you really want improved resolution, you're better off with cleartype-like tech on an lcd (subpixel hinting, anti-aliasing on a subpixel level, only lcd's can do this). It triples horizontal resolution (the one we're most sensitive to). As a result, my 75 dpi laptop screen displays ebooks almost as if they were printed when using the ms ebook reader.
Although it may be more "standard compliant", it is not as forgiving as IE in terms of bad HTML. I still get many sites that don't work in Mozilla - and because I know how HTML works and know the whole history behind W3C compatability standards I'll launch IE and look at the site with that. my mother would probably think the website was screwed. The sad fact of the matter is that there are a myriad of WYSIWYG HTML authoring tools that produde non-compliant HTML and to use the argument that they should fix their problems and Mozilla is god because it adheres to standards is horribly narrow-minded.
The gecko engine is a best effort with respect to approaching IE. It already does a lot of things which aren't in the standards per se. It has two different rendering modes which aren't standards compliant for pages that are buggy. The problem is that making a browser that acts just like IE is a HUGE waste of development resources. IE is a moving target. You'd always be playing catch up. And for what? If your engine is EXACTLY like the IE engine, why not embed that? And it really has to be EXACTLY like the IE engine before all sites will work, because a lot of sites depend on the bugs in the IE engine in order for them to show up correctly.
Even big sites like the internation herald tribune depend on IE bugs to render correctly. (Load it up in mozilla and due to mozilla actually interpreting the html and css correctly you'll get overlap between the image at the top and the text next to it.)
Also, we're just now getting the web dev tools vendors to output standards. Dreamweaver mx now produces good clean standards-compliant code. Frontpage 2003 has much improved standards support. The various blogging tools play MUCH nicer with respect to standards. We're finally seeing the tide change wrt getting people to use standards, and now you propose to throw that away and give the web to microsoft. Why?!?
There are several sites which makes IE screenshots for you. There's also the option of using codeweavers' crossover products, which run IE on linux (which is how I test websites for IE in linux).
I forgot to mention the reason why better standards support matters. It matters because a lot of cool webdesign techniques exist that create pages that show the content correctly in basic browser (like IE), but do all kinds of cool stuff in advanced browsers (like firefox). There exist no equivalent techniques for IE's proprietary stuff. So if you develop for standards you can do all the cool stuff while offering basic functionality for the other camp, but if you develop for IE, you're locking out the entire market.
As a result, standards-based sites often look prettier in firefox.
What has firefox got that's not coming in the next service pack for IE?
Better standards support. IE would have to break backwards compatibility to fix this. They won't.
Better security track record. No known unfixed security flaws. This is a good reason to go from outlook (express) to thunderbird too, by the way. Don't be a worm victim. Switch.
Smaller download. Firefox 0.8 is less than 7 megs and getting smaller. I very much doubt IE's next service pack will fit in that category. If your choice is between firefox or IE's next service pack, and you're on modem, it makes sense to go for firefox. I have a modem-using friend who uses firebird 0.6 as a browser because she doesn't want to suffer the huge download to bring her basic installation up to the current service pack.
Better extensibility. IE doesn't have a framework for easily and quickly developing extensions with just a zip tool and notepad. It won't have it till longhorn.
Better web development functionality. The javascript console, the dom inspector, the various bookmarklets, the less permissive engine (which points out errors in your code much more easily than IE), and the in general better standards support work to make gecko-based browsers much better choices for web developers. Though I admit that is a niche market.
Currently in most cases firefox is also faster (except on first load of the app, because it's not preloaded). But I suppose an IE service pack might fix this. I doubt it though.
You're overestimating how many people know of the google bar. Even some techie friends of mine were unaware of it until I mentioned it. Besides, if you're going to install third party browsing software, you might as well install a whole different browser, because IE's inability to block popups is most definitely not its biggest flaw.
I think you're giving the konqueror guys too much credit. Konqueror didn't become a usable day-to-day browser (for me) until after mozilla 0.7, which was the first usable day-to-day version (again, for me).
i know it crashes more often than not for me, and after a bit i have to restart it because it gets so slow it's unuseable
I'm always curious what exact configuration the people who say this are running. I've run both the suite and the separate browsing app (firebird/firefox) for literally years, and ever since about mozilla 1.2 it has been fast and stable for me, on at least 10 different machines.
One well known caveat is that if you're having stability problems you should start with a fresh profile. I've had this problem exactly once. Saw crashes, created a new profile, copied over bookmarks.html, and the crashes disappeared.
Creative Criticism: The DHTML or whatever is used to give the advanced editing features of Exchange 2000 web mail, msn hotmail, yahoo mail, and the geocities web site editor don't work in Firebird;
The IE-only way of doing this doesn't work. The standards-based way however does work. Most blogging tools support rich text in gecko browsers. That MS uses its own proprietary stuff instead of the standards is hardly surprising, and I suspect yahoo and geocities are just suffering from inertia (because admittedly, mozilla hasn't had this capability for more than a year).
Slashdot Mirror
The worm wouldn't be the cause of the deaths. It would be a symptom of the real cause - incompetence. Nobody should be installing Windows on these types of systems. Period. Even Microsoft will tell you that.
Hmm, I'm not sure of that. They claim their stuff is suited for mission critical systems. Though, if they ever sell a system, and that sale leads to human deaths because of flaws in it, even their eula isn't going to save them. That's manslaughter.
the "science proves the theory of evolution" statement is not accurate. I would be happy to answer any questions or continue this list.
How do you explain the remarkable genetic match between humans and chimpanzees, up to the point where diseases can make the leap between them and us? Was God just lazy and copy/pasting code?
See, the thing with evolution is that we don't have a complete map from the earliest life to any current species. Anti-evolutionists like to say "ha! no complete and fully detailed evidence means your theory is wrong!", but absence of evidence is not evidence in itself, and there is still a lot of evidence in support of evolution (or rather natural selection), though not necessarily related to humans.
Now, we can quibbly about the exact processes of evolution, but to deny that species evolve involves putting a very large blind spot in your science. Speciation is clear, and all around us. You can even make it happen experimentally in labs if you go through enough generations.
People working in an office (or even an overwhelming majority of home users) dont get paid to fuck with the source code, nor would most of them even want to. Only programmers care about that shit, and at least 99% of computer users are not programmers.
Yeah, and people not working in capitol hill don't get paid to fuck with laws, nor would most of them even want to. Only lawmakers care about that shit, and at least 99% of citizens are not lawmakers.
Just because you're not involved in creating something doesn't mean you shouldn't have input, especially if you're going to be the user of said product (which most computer users are).
What can be the VB for Free Software?
Hmm, I wonder why you got modded up, this sounds very trollish to me. Lots of long complicated sentences to make a very basic point:
"In order for Java to become for Free Software and Gnu/Linux what VB became for Microsoft, Java has to be Freed and put out under the GPL."
Note that I used your exact words to not get accused of making a strawman argument here.
Your point is one that is deeply flawed. VB isn't GPL, yet it is ubiquitous. Why? Because it doesn't require you to know about programming. It doesn't require you to actually learn the trade. You can drag-and-drop and copy/paste your way to something that works. Sure, it's a really rotten environment to do any real development in, but for prototypes, or for people who don't want to learn a real language, it is excellent. Java, by it's compulsory OO nature could never be that. It's not "easy" enough.
here in Germany we have very tough laws with regard to your personal information and how it must be handled by businesses and the government
This is actually a EU directive. Or actually, two different ones. One dealing with regular privacy (enforced since 1998), and one with online privacy (enforced since last year). Seemingly when you read the text of the directive, it has a lot of teeth, but in practice they make exceptions every time someone asks. Like when the US insisted on having every bit of available information on EU citizens flying into the US (including the kind of meal they took, and how they paid for their ticket). The EU after some haggling made an exception that allows some, but not all, of the passenger information to pass to the US.
At least, a privacy law, even if it's not being enforced, is still better than no privacy law.
This is sort of off-topic, but relevant to the parent.
I'm looking for Geoworks to throw onto some 486's I want to bring back to life -- the last version I remember had a web-browser and everything!
If the 486's have at least 16 megs of ram, and are at least 33 mhz machines, you can put a stripped down debian on them with a custom compiled linux kernel, opera, abiword, and some really lightweight wm like pwm. It does work at a decent speed, but you have the advantage of using software updated for this millenium. If opera is too bloated as a webbrowser (only on low end 486's, since opera is ridiculously light) you could try dillo, which doesn't support all the standards, but is usable for webbrowsing.
I ran debian like that on a 486 DX4/100 with 20 megs of ram and an 800 meg hd for a long time. Worked great. Usable as a primary desktop. As long as you don't want to multitask ofcourse, but then on a 486 multitasking is pretty much out of the question anyway. That was with the regular debian Xfree, which I suppose could be seriously stripped down, but I stopped stripping down once I got the desktop functionality I wanted working at a decent speed.
A 386 can be made quite useful with linux if you stick to shell apps. There are a ton of great shell tools. Links or w3m for webbrowsing, latex for making pretty documents (you don't need to see the graphical output of something you do in latex until you print), any of the norton commander clones for a shell. Useful, even with only 8 megs of ram.
Believe me, when you're actually reading it, 3.1 mb is not so small. In fact, it's staggeringly vast.
It's already there. It's just not in button form -- ...
--max_upload_rate
maximum kB/s to upload at, 0 means no limit (defaults to 0)
Setting that to 1 kB/s should be slow enough even for a modem user
To those saying this speeds things down, it doesn't necessarily. On my cable modem connection I have an upload cap of 16K / sec, bittorrent quickly swamps that to such a degree that downloading drops to less than 20 K / sec. By setting the upload cap in bittorrent to 10K / sec the tcp/ip acks are still going through fast, and my dl speeds are 300K+.
How else does the Code Red author decide, "Hey! I found this buffer overflow routine in the unicode support for URLs in the IIS Indexing Server"?
Didn't code red take advantage from an already published security flaw? Most worms exploit known flaws for which patches are available.
Anyway, reverse engineering an entire operating system into something like source is quite difficult. If you don't have debug symbols you can't deduce from the name what each variable stores and what each function does. And assembly doesn't even have a concept of variables and functions (well, that last bit can be argued, but let's generalise). It's all byte manipulations and conditional jumps. Not easy to see what's happening, especially when you're talking about literally tens of megabytes of assembly. Blackbox testing (throwing random data into the input and seeing what the output does) is much more useful to find exploitable problems.
If CrossOver Office starts to bug me, then I can go to WineX and it will work about the same.
The focus of crossover office and winex is entirely different. Crossover office targets desktop apps, while winex targets games. As a result crossover office runs apps like office and photoshop much better, but winex supports directx 9 and runs the majority of popular games.
Ofcourse, improving the win32 api's to run either desktop apps or games will improve the ability to run the other category as well, which is why winex is reasonably good at running desktop apps and crossover office / regular wine is reasonably good at running games.
if a newbie can insert his windows app cd, run the installer under wine and have a application link inserted in a windows directory in the gnome menu that is already associated with wine and that app
Crossover office adds apps to the menu and desktop on my system (debian + kde). That's the diff between crossover and wine, you pay for the polish.
That's because you're using the wrong command:man is for when you already know what you're looking for. apropos is for when you want a man page of something, but you don't know what it should be.
Also you can repeaditly stream a text version of War and Peace or some other lengthy book, with a counter on the recieving end showing how many times you have downloaded it.
Download your text copy from project gutenberg. If you haven't read it, please do, it's a great book.
Holy early-nineties webdesign, Batman!
Early 90's? Frames didn't get introduced until netscape 2.0 in 1996.
I feel so old for remembering the web from before frames even existed, and I'm not even 25 yet.
Why does everyone assume that causality can't be violated (in "small" ways)? I mean, there's no such thing as causality at the quantum level, is there?
If you receive an answer before you ask a question, will you still ask the question? If you don't, no answer will be returned to the unasked question, so you will have an answer that was never sent. This is impossible. Therefore, causality must be respected.
Now, I'm not saying it is impossible to send information back in time, just that it leads to all kinds of strange and disturbing problems if you assume that it is possible and that we have free will. One or the other must go, and I prefer for us to have free will.
So now the only barrier is the speed of light?
Colonel Sandurz: Prepare ship for light speed.
Dark Helmet: No, no, no, light speed is too slow.
Colonel Sandurz: Light speed, too slow?
Dark Helmet: Yes, we're gonna have to go right to ludicrous speed.
If you really want improved resolution, you're better off with cleartype-like tech on an lcd (subpixel hinting, anti-aliasing on a subpixel level, only lcd's can do this). It triples horizontal resolution (the one we're most sensitive to). As a result, my 75 dpi laptop screen displays ebooks almost as if they were printed when using the ms ebook reader.
Although it may be more "standard compliant", it is not as forgiving as IE in terms of bad HTML. I still get many sites that don't work in Mozilla - and because I know how HTML works and know the whole history behind W3C compatability standards I'll launch IE and look at the site with that. my mother would probably think the website was screwed. The sad fact of the matter is that there are a myriad of WYSIWYG HTML authoring tools that produde non-compliant HTML and to use the argument that they should fix their problems and Mozilla is god because it adheres to standards is horribly narrow-minded.
The gecko engine is a best effort with respect to approaching IE. It already does a lot of things which aren't in the standards per se. It has two different rendering modes which aren't standards compliant for pages that are buggy. The problem is that making a browser that acts just like IE is a HUGE waste of development resources. IE is a moving target. You'd always be playing catch up. And for what? If your engine is EXACTLY like the IE engine, why not embed that? And it really has to be EXACTLY like the IE engine before all sites will work, because a lot of sites depend on the bugs in the IE engine in order for them to show up correctly.
Even big sites like the internation herald tribune depend on IE bugs to render correctly. (Load it up in mozilla and due to mozilla actually interpreting the html and css correctly you'll get overlap between the image at the top and the text next to it.)
Also, we're just now getting the web dev tools vendors to output standards. Dreamweaver mx now produces good clean standards-compliant code. Frontpage 2003 has much improved standards support. The various blogging tools play MUCH nicer with respect to standards. We're finally seeing the tide change wrt getting people to use standards, and now you propose to throw that away and give the web to microsoft. Why?!?
There are several sites which makes IE screenshots for you. There's also the option of using codeweavers' crossover products, which run IE on linux (which is how I test websites for IE in linux).
I forgot to mention the reason why better standards support matters. It matters because a lot of cool webdesign techniques exist that create pages that show the content correctly in basic browser (like IE), but do all kinds of cool stuff in advanced browsers (like firefox). There exist no equivalent techniques for IE's proprietary stuff. So if you develop for standards you can do all the cool stuff while offering basic functionality for the other camp, but if you develop for IE, you're locking out the entire market.
As a result, standards-based sites often look prettier in firefox.
What has firefox got that's not coming in the next service pack for IE?
Better standards support. IE would have to break backwards compatibility to fix this. They won't.
Better security track record. No known unfixed security flaws. This is a good reason to go from outlook (express) to thunderbird too, by the way. Don't be a worm victim. Switch.
Smaller download. Firefox 0.8 is less than 7 megs and getting smaller. I very much doubt IE's next service pack will fit in that category. If your choice is between firefox or IE's next service pack, and you're on modem, it makes sense to go for firefox. I have a modem-using friend who uses firebird 0.6 as a browser because she doesn't want to suffer the huge download to bring her basic installation up to the current service pack.
Better extensibility. IE doesn't have a framework for easily and quickly developing extensions with just a zip tool and notepad. It won't have it till longhorn.
Better web development functionality. The javascript console, the dom inspector, the various bookmarklets, the less permissive engine (which points out errors in your code much more easily than IE), and the in general better standards support work to make gecko-based browsers much better choices for web developers. Though I admit that is a niche market.
Currently in most cases firefox is also faster (except on first load of the app, because it's not preloaded). But I suppose an IE service pack might fix this. I doubt it though.
You're overestimating how many people know of the google bar. Even some techie friends of mine were unaware of it until I mentioned it. Besides, if you're going to install third party browsing software, you might as well install a whole different browser, because IE's inability to block popups is most definitely not its biggest flaw.
I think you're giving the konqueror guys too much credit. Konqueror didn't become a usable day-to-day browser (for me) until after mozilla 0.7, which was the first usable day-to-day version (again, for me).
i know it crashes more often than not for me, and after a bit i have to restart it because it gets so slow it's unuseable
I'm always curious what exact configuration the people who say this are running. I've run both the suite and the separate browsing app (firebird/firefox) for literally years, and ever since about mozilla 1.2 it has been fast and stable for me, on at least 10 different machines.
One well known caveat is that if you're having stability problems you should start with a fresh profile. I've had this problem exactly once. Saw crashes, created a new profile, copied over bookmarks.html, and the crashes disappeared.
Creative Criticism: The DHTML or whatever is used to give the advanced editing features of Exchange 2000 web mail, msn hotmail, yahoo mail, and the geocities web site editor don't work in Firebird;
The IE-only way of doing this doesn't work. The standards-based way however does work. Most blogging tools support rich text in gecko browsers. That MS uses its own proprietary stuff instead of the standards is hardly surprising, and I suspect yahoo and geocities are just suffering from inertia (because admittedly, mozilla hasn't had this capability for more than a year).