This is exactly why I believe that the typical scenario (internet <-> firewall <-> intranet) is pretty much useless for protecting against worms and such.
Much better idea would be to connect everything to firewall directly, basicly replacing switches with something that can do packet filtering.
Unfortunately, this.. well.. costs money.
Another possibility is what MS considers doing, that is, running at least some kind of private, software based firewall on every workstation and server by default, unless there's a better firewall that the server is directly connected to.
Instead of thinking "connection to Internet should go through a firewall" people to should think "connection to a network should go through a firewall".
Some kind of central management for all these workstation-firewalls would be preferrable though.
I don't get this whining about MS installing software to your computer without you knowing.
I mean, if MS wanted to install something without you knowing, MS could just as well ship that on the original installation CD.
If you don't know what's already on your computer, why should you care what MS adds there..
Now, I understand that people fear that MS changes something, that will cause something else to fail.
I think this is the most important difference between the Open-Source (and UNIX) world and the commercial (and Windows) world; OSS gets written to specifications, Windows stuff to implementation.
In OSS, if the underlying implementation is wrong, then somebody will fix it, and in the mean time people will find a work-around that doesn't touch the faulty part. In Windows software, it's not uncommon to rely on some undocumented feature, or even an error (I don't like the word "bug") in the implementation.
I think that OSS is better here for 2 reasons: OSS can be fixed by who-ever finds the error and in UNIX world there are more different implementations, more diversity, so one can't expect portability among different UNIX'es, or even versions, or even different CPU architectures, if one relies on errors.
Actually.. if you wanted to have a secure system, you might want to check EROS instead. The site also explains what is wrong with current systems and who it could be fixed.
I agree though, that going with a microkernels are better way to construct stuff that even might work..
This is exactly what I hoped when I see this.
It will have to happen at some point, by some large company. I don't know if it's MS that will do it, but someone will.
It seems to me that there is quite some action going on in the legal grounds. We have this SCO case, and now we have this. There are proposals for new legislation. Then there's things like Linux gaining a lot of publicity lately, and (ordinary) people getting frustrated with MS, spam, viruses..
I don't know if a qualify as a "younger" gamer (I'm only 21), but personally, I like Nintendo's games more than most other games, because they are more rich on content.
One gets bored to violence. One gets bored to "darkness" and "evil". Once you've spent 10 years of your life with the games for the "older" you get tired of all that. And who cares about sports?
I'm not saying that I don't want to play a violent game though. I don't really care, if the game is otherwise good you'll ignore it anyway, and if it's not, then it's not worth playing in the first place. Good thing about games for the "younger generations" is that you don't have to care who's watching you play. You can play with your little sisters just fine.
I admit that playing Nintendo's games isn't as sexy and cool and all that, as playing some "real" games, but I don't play games to collect fame, I play them to entertain myself. =)
Gentoo's not really faster on my hardware. That's identical hardware. Some things like mplayer might ARE faster than binary packages, but most of things are just the same. Won't notice the difference.
Now, that's not the reason I run Gentoo. The single reason I use it is Portage. It's just as easy to use as apt-get in Debian with 2 advantages.
First, there's USE-flags. You get to choose what you want. I don't want MySQL or KDE support in any packages, so I don't have them. If something supports PostgreSQL, that's compiled in instead.
Second, it's somewhat more up-to-date with newer versions that Debian. For servers I'd still use Debian, but on my desktop I like to stay current. Not bleeding edge, but current. Also there's less trouble with Debians policies about Licenses.
The package management is the best I've seen so far, so that's why I run Gentoo. My previous distro was LFS, but Gentoo gives me all the customization I want, with package management.
"a computer system for management of resources" could mean pretty much anything, including EVERY operating system (including DOS).. so there's some prior art.
"ordering and downloading resources from computerized repositories" basicly could mean each and every database ever built which involves some kind of client/server concept (so we can get the downloading too).. some prior art there too, since a "SELECT"-statement is basicly an order on what to download.
Now, the third part in common English reads "session management" (and possibly some kind of transactions, in which case refer to the previous paragraph) which I can't really say if they could have a valid claim over, except that I'd really imagine that there is prior art prior Netscape introducing cookies for this purpose, since the point of cookies is to make it easier.
Really, the only thing that makes this particular point a little harder to prove wrong is the introduction of the Web which means that the timeframe is much smaller. I'd still imagine that such prior art does exist.
Hopefully the "other claims" are little more realistic. I personally never understood the idea of patents. I could understand a patent for one, maybe even two years, but that's about it.
Immersion talks about force-feedback on their website, but I think either they or I have understood something wrong.
To this date I've lived under the impression that force-feedback means "force, to give you feedback", which is to say, when driving car you feel the road, so to say, because the steering wheel (or whatever it's called) turns when the wheels turn. The same thing with aeroplanes. It's a great thing in gaming, since it's much easier to fly a plane if it resists with some force your attempts to control it, since that provides you with... well.. feedback.
This thing Immersion calls force-feedback. It seems to me more like a vibrator of some kind. I'd imagine female gamers might like it though..
Personally I think most of the uses they advertize it as good for, can be better implemented with old-style shaking display. Btw, why is that so rare these days?
There's no such thing as bug in a program.
Bug sounds like a small insect that walks around causing problems. I'm not the first to believe that this is wrong, and that this use of word BUG is one of the reasons software has so much errors.
I mean, if you accept that every such 'bug' is in reality an ERROR in the software code, an ERROR made by programmer, you should realize that writing software is not only hard, is HARDER.
Reduce complexity until you can understand it,
add about million assertions (and have then crash the software) and PROVE your solutions to be correct and in the end.. you shall have your VALID, ERROR-free program.
I though the whole cookie problem was hot some 7 years ago.. now, Sweden being the first country to actually do something about it now that everybody already figured out cookies are good thing and every decent browser having features to control them, what can we expect on other fronts ?
I'd expect popup adds to be outruled in 10 years from now.. in the first country that happens to do it..
I'd expect most swedish sites to do the same as to this date, except add a notice "We will place a session cookie on your computer when you login. If you don't want this, please, do not login." next to their login box.. big deal..
"I just developed this program. I don't actually use it. So you can't sue me. Sue the users. Who are they? You have to find out yourself, since this is Free Software and I don't keep track of who's downloading it."
Personally, I think it's better to give source when someone buys the software, than not give it at all.
I know several cases where a product (usually a library) was chosen because full source was included in the deal.
Actually, even GPL allows that kind of distribution. I can sell you GPL'ed code (and source) and not give you the source if you don't buy it, since I'm not distributing the product before you buy it:)
If one limits the redistribution, then it's no longer Free Software. I'd like to know what use is having the source if you can't compile it. There's now way to know if something was added or removed or changed before the compile. I assume they allow recompile of Windows CE though.
Also, MS is missing the most important part in OSS world. True, there's always someone that says what goes in and what doesn't, but you always have the freedom of forking the project! That's part of why it works. If the original project likes what the fork produced, they can later be merged again.
I don't consider the price to be the main issue. After all one develops a project one is hoping to use. If there are arbitary limitations to that development, I consider it much more important problem.
Uum... ElectricFence is a library that implements malloc, that check that you don't go over any bounds in dynamically allocated variables. Does a few other checks too (like double free). Link it with your project (or use LD_PRELOAD) and that's it.
Works for C, but probably can be made to work with C++ too by writing an operator new that uses malloc() internally. Not sure if glibc actually does that.
History repeats itself. Look at Algol (and it's different versions) and you find something familiar. Then look at Java, and again you see the same problem. If you want, take a look at Common LISP standard, and you find the even if the language is good, one should never let any committee standardize it.
Actually, to make the new F language worthy of being successor of C++, it's probably better idea to overload ^ to be the exponentiation operator, so as to confuce everybody.
Ok, this seems to compile on me too. I admit I was lazy and didn't test what I posted on the compiler, and since I couldn't paste the original offending code (closed source) I had to simplify the case.
Too bad I already rewrote the code with another algorithm so I don't have anything to refer to anymore. Probably one of these bugs like gcc2.95.3 C++ compiler sometimes having trouble with two nested for-loops where there's no block for the outer, but just the inner for statement. Putting the for-loop into a block fixes the case.. but this is starting to get out of topic so I shut up.
Ps. the java version was 1.4.0_01 (windows) so it's old anyway, and the bug might as well be fixed.
Slashdot Mirror
Just do you google searches through a european proxy or something..
Much better idea would be to connect everything to firewall directly, basicly replacing switches with something that can do packet filtering.
Unfortunately, this.. well.. costs money.
Another possibility is what MS considers doing, that is, running at least some kind of private, software based firewall on every workstation and server by default, unless there's a better firewall that the server is directly connected to.
Instead of thinking "connection to Internet should go through a firewall" people to should think "connection to a network should go through a firewall".
Some kind of central management for all these workstation-firewalls would be preferrable though.
I mean, if MS wanted to install something without you knowing, MS could just as well ship that on the original installation CD. If you don't know what's already on your computer, why should you care what MS adds there..
Now, I understand that people fear that MS changes something, that will cause something else to fail. I think this is the most important difference between the Open-Source (and UNIX) world and the commercial (and Windows) world; OSS gets written to specifications, Windows stuff to implementation.
In OSS, if the underlying implementation is wrong, then somebody will fix it, and in the mean time people will find a work-around that doesn't touch the faulty part. In Windows software, it's not uncommon to rely on some undocumented feature, or even an error (I don't like the word "bug") in the implementation.
I think that OSS is better here for 2 reasons: OSS can be fixed by who-ever finds the error and in UNIX world there are more different implementations, more diversity, so one can't expect portability among different UNIX'es, or even versions, or even different CPU architectures, if one relies on errors.
I know this is a bit out-of-topic but, about this X-box as a Linux firewall thing: does the X-Box have 2 ethernet cards?
I agree though, that going with a microkernels are better way to construct stuff that even might work..
Now that I think of it, for some strange reason, I remember several vulnerabilities with wu-ftpd...
It seems to me that there is quite some action going on in the legal grounds. We have this SCO case, and now we have this. There are proposals for new legislation. Then there's things like Linux gaining a lot of publicity lately, and (ordinary) people getting frustrated with MS, spam, viruses..
We might well live on the edge of new revolution.
One gets bored to violence. One gets bored to "darkness" and "evil". Once you've spent 10 years of your life with the games for the "older" you get tired of all that. And who cares about sports?
I'm not saying that I don't want to play a violent game though. I don't really care, if the game is otherwise good you'll ignore it anyway, and if it's not, then it's not worth playing in the first place. Good thing about games for the "younger generations" is that you don't have to care who's watching you play. You can play with your little sisters just fine.
I admit that playing Nintendo's games isn't as sexy and cool and all that, as playing some "real" games, but I don't play games to collect fame, I play them to entertain myself. =)
Now, that's not the reason I run Gentoo. The single reason I use it is Portage. It's just as easy to use as apt-get in Debian with 2 advantages.
First, there's USE-flags. You get to choose what you want. I don't want MySQL or KDE support in any packages, so I don't have them. If something supports PostgreSQL, that's compiled in instead.
Second, it's somewhat more up-to-date with newer versions that Debian. For servers I'd still use Debian, but on my desktop I like to stay current. Not bleeding edge, but current. Also there's less trouble with Debians policies about Licenses.
The package management is the best I've seen so far, so that's why I run Gentoo. My previous distro was LFS, but Gentoo gives me all the customization I want, with package management.
"ordering and downloading resources from computerized repositories" basicly could mean each and every database ever built which involves some kind of client/server concept (so we can get the downloading too).. some prior art there too, since a "SELECT"-statement is basicly an order on what to download.
Now, the third part in common English reads "session management" (and possibly some kind of transactions, in which case refer to the previous paragraph) which I can't really say if they could have a valid claim over, except that I'd really imagine that there is prior art prior Netscape introducing cookies for this purpose, since the point of cookies is to make it easier.
Really, the only thing that makes this particular point a little harder to prove wrong is the introduction of the Web which means that the timeframe is much smaller. I'd still imagine that such prior art does exist.
Hopefully the "other claims" are little more realistic. I personally never understood the idea of patents. I could understand a patent for one, maybe even two years, but that's about it.
To this date I've lived under the impression that force-feedback means "force, to give you feedback", which is to say, when driving car you feel the road, so to say, because the steering wheel (or whatever it's called) turns when the wheels turn. The same thing with aeroplanes. It's a great thing in gaming, since it's much easier to fly a plane if it resists with some force your attempts to control it, since that provides you with... well.. feedback.
This thing Immersion calls force-feedback. It seems to me more like a vibrator of some kind. I'd imagine female gamers might like it though..
Personally I think most of the uses they advertize it as good for, can be better implemented with old-style shaking display. Btw, why is that so rare these days?
Windows 2000 and XP do this. No reason why Linux couldn't.
I mean, if you accept that every such 'bug' is in reality an ERROR in the software code, an ERROR made by programmer, you should realize that writing software is not only hard, is HARDER.
Reduce complexity until you can understand it, add about million assertions (and have then crash the software) and PROVE your solutions to be correct and in the end.. you shall have your VALID, ERROR-free program.
I'd expect popup adds to be outruled in 10 years from now.. in the first country that happens to do it..
I'd expect most swedish sites to do the same as to this date, except add a notice "We will place a session cookie on your computer when you login. If you don't want this, please, do not login." next to their login box.. big deal..
"I just developed this program. I don't actually use it. So you can't sue me. Sue the users. Who are they? You have to find out yourself, since this is Free Software and I don't keep track of who's downloading it."
I don't think it's too much to put a grub bootdisk (or CD or something) in after installing windows, and running root and setup commands manually.
And btw, you want those boot disks anyway :)
Another thing that you can also get all the Thinkpad drivers from their website.
Actually, even GPL allows that kind of distribution. I can sell you GPL'ed code (and source) and not give you the source if you don't buy it, since I'm not distributing the product before you buy it :)
If one limits the redistribution, then it's no longer Free Software. I'd like to know what use is having the source if you can't compile it. There's now way to know if something was added or removed or changed before the compile. I assume they allow recompile of Windows CE though.
Also, MS is missing the most important part in OSS world. True, there's always someone that says what goes in and what doesn't, but you always have the freedom of forking the project! That's part of why it works. If the original project likes what the fork produced, they can later be merged again.
I don't consider the price to be the main issue. After all one develops a project one is hoping to use. If there are arbitary limitations to that development, I consider it much more important problem.
Tools -> Filters.. -> Add
Works for C, but probably can be made to work with C++ too by writing an operator new that uses malloc() internally. Not sure if glibc actually does that.
History repeats itself. Look at Algol (and it's different versions) and you find something familiar. Then look at Java, and again you see the same problem. If you want, take a look at Common LISP standard, and you find the even if the language is good, one should never let any committee standardize it.
Actually, to make the new F language worthy of being successor of C++, it's probably better idea to overload ^ to be the exponentiation operator, so as to confuce everybody.
actually, I think the behaviour of this construct is undefined in current languages.
Too bad I already rewrote the code with another algorithm so I don't have anything to refer to anymore. Probably one of these bugs like gcc2.95.3 C++ compiler sometimes having trouble with two nested for-loops where there's no block for the outer, but just the inner for statement. Putting the for-loop into a block fixes the case.. but this is starting to get out of topic so I shut up.
Ps. the java version was 1.4.0_01 (windows) so it's old anyway, and the bug might as well be fixed.
ok, slashcode ruined the formatting.. (yeah yeah, problem exists between chair and keyboard).. sorry