I actually think the EU is one of the smaller economic areas, in comparison to the US or to the Asias; do you remeber Japan, Taiwan, China, and
Korea? They form a formidable engineering and technology quartet, with India rising quickly too. But this is an uninformed 'opinion' post, on my
part.
Either you are comparing apples to oranges : Europe[an Union] is tightly-bound politically and economically, which is definitely not the case in Asia in general;
or you just don't care about the numbers. The European Union is the biggest economic zone in the world (GNP, production, intra-zone/inter-country and inter-zone commerce, etc.). And, yes, bigger than NAFTA.
China and India also have space flight programs, as well as many other countries that have the ability to launch small, unmanned flights.
Indeed. Europe does, too; with quite serious numbers and success rates, BTW. And they aren't particularly plagued by the failures we have come to expect from NASA, recently. How many probes did they lose or mishandle in the last few years?
It is also why NASA can succeed in safe, reliable space flight time, and time again, while
other space programs are struggling.
How exactly would you back up your claims about this? The only other country with "space flight time" (I assume manned flights there; if you're also talking probes or satellites your affirmation is awfully clueless/uninformed) is Russia, which is broke. Mir was scheduled to fall because of lack of funding. Not because of a technical failure.
I'm curious what kind of update/package system OS X uses...
Not exactly the answer you want... anyway. Open Packages aims at unifying the {Free,Net,Open}BSD, BSDi and [Apple] Darwin package systems. So, when (if?) this gets functional, Unixy software will be nicely packaged.
one of the most powerful things about *nix is the ability to chain together multiple commands in a pipeline to perform
some desired task. This is why most *nix commands are very simple and only accomplish a very limited task.
How true!
Alas the current, uh, wave in the Unix/Linux people is more like "do everything with perl and emacs". Forget about pipes and redirections, say hello to the language with 42 different ways of writing if(x) y; (BTW, is there anyone besides Larry Wall who knows all of these?) and a HUGE libc-duplicated-for-perl.
You don't even need a gui to work on 90% of the data you want to process everyday. You are just too lazy to
learn how the real system works.
Yes, you don't even need a gui to work on 90% of the data you want to process everyday. However, other people are not lazy, they have better things to do with their time than tinkering with a compulsory-manpage-and-CLI OS.
I hack a BSD kernel as a job, I hack Drawin and LinuxPPC for pleasure, but I made sure my mom doesn't ever have to see a command line. Her job is not understanding computers, her job is publishing.
I'm tired of people who just can't accept that the complexity of using a computer is not virtue, but lousy design.
I was thinking of the proofs of "while" loops I sweated over back in school. B-)
Did that too.. worthless as a tool but a nice introduction to more interesting stuff; it's a bit like the ubiquitous toy compiler project in CS studies: everyone does that someday and everyone buries the result and uses gcc instead past the deadline.:)
The first is that you can't easily back up and punt; when you realize half-way through that the
formal design which looked good on paper is eating resources or presents a poor user interface in practice, it is very hard to go back and
improve things.
Indeed ! Proving user interfaces is unreasonable in the first place (what exactly are you proving then?)
you have to prove that the code matches the mathematical model
Not. The idea is to translate automatically from model to code using robust, time-proven translators (in the same way that one can more or less trust a non-RedHat gcc-O2 to translate C code correctly, at least on an x86; or in the way you trust the CPU to accurately execute the machine-language instructions you feed it). Interesting properties are proven on the model.
Perhaps you thought of, say, UML when you read "formal methods"? That's where the disagreement comes from. UML is not a proving tool, it's a design helper, roughly as provable as C (that is, in the real world, not provable at all.)
Automatic translation of proven models is used in embedded software for critical stuff, e.g. antilock brakes, in protocol design and implementation, and such stuff.
Turing did not prove that you can't prove any software. Turing did prove that you can't prove all software.
There's a huge difference. For many problems you can build a solution that is provably right. On the other hand, given a "random" program you can't generally prove it; ie. there is no systematic method to "prove" software, but there are methods to turn some provable methods and descriptions into programs that keep the same properties. That's what formal languages are about.
Why on Earth should it be other users who have to pay for what certain abusers are doing with their broadband connections?
I'm not usually that pro-market, but this time it works quite well: let some competition (dis)solve the problem. Here in Paris one can get consumer broadband via France Telecom ADSL: expensive, with a bandwidth cap as the sole traffic limit and a decent backbone serving it; or cable access, less expensive (not cheap yet), with an undersized backbone (though it's good enough for mail and casual browsing), and a monthly upload cap. (and there are third-party DSL providers, cleverly combining the drawbacks of both)
Overall if you want real bandwidth you have to pay more; if you want Joe Average's browsing bandwidth, the cheaper service is okay.
Why should game designers create something so innovative that you keep going back to the same game and playing over and over? Then
there's no turnaround for them.
Easy. Make me come back and buy a better version of that game that I love so much.
Civilization, Civilization II, Alpha Centauri, Alien Crossfire... Civ III when it comes out (Sid Meier's, not Call to Power). Each one was even better than the previous one, each one was better than almost any game I've ever played.
The morale is: I'll buy more if the quality is there.
Konqueror is the best I've seen in this regard. Each site that asks for a cookie Konq prompts you for. I know other browsers have this
option, but in Konq you can specify to allow or deny all future cookies from a specific domain. It is perhaps Konq's best feature yet.
Funny thing... IE 5.0 for Mac (yes, the Microsoft product) has this ability too. It can also prevent gifs from animating, plugins from loading, etc. That makes it a much better browser than IE 5.5 Win and Netscape 4.*, in my opinion.
> so playing a DVD at 1600x1200 is just as fast as 800x600...
I'm afraid you are wrong. Motion compensation and iDCT does nothing to do with output resolution.
Please, read what I wrote before disagreeing. Yes, it is true that motion compensation and iDCT have nothing to do with output resolution. This is why I wrote and scaling/smoothing, which you conveniently snipped from the sentence to trick people into believing you are the smart guy.
Damn, I wish Slashcode had a killfile. Maybe I should learn some Perl and add it myself.
all macs [...] use software dvd decoding. ati has nothing to do with it
In a way, yes: decryption, demuxing, audio decoding, subitiling is done by software. Some high-level MPEG video stuff is done by the software but the bulk of the effort (that is, motion compensation, inverse DCT and scaling/smoothing) is done by the video card (so playing a DVD at 1600x1200 is just as fast as 800x600). This is why this "software decoder" only works with Rage 128/Radeon (and now, GeForce2 MX). Actually it is about the same thing as what you get bundled with said cards in their retail PC version.
BTW, the hardware support in ATI chips is first-rate, quality-wise. The image is better than most hardware decoders (we checked a Hollywood-whatever and an ATI card on the same PC. The ATI's image is much cleaner)
The inventor gets nigh-total control over their good idea for the next X years, in exchange for telling us what it is.
"in exchange for telling us what it is" There, you found the problem yourself, only you didn't see it.
Did RAMBUS say anything? No, they kept the patent around, well hidden. So that 1-no one knows that someone patented this thing, and 2-no one benefits from the research anyway (since the behaviour of Rambus here is, "hide our patent and wait till someone else builds the same thing")
A CubeSat can hold anything [...] to the ashes of a loved one
Guess there are many more romantic millionnaires than non-starving labs. LEO will soon be swamped in a cloud of ashes and tiny engraved cubes blown by other debris.
Rant Isn't there too much stuff loose in orbit already?
Re:Affect hardware sales?
on
OS X on x86?
·
· Score: 1
He [Steve Jobs] never really cared much for the software... what really excited him was to make and ship hardware.
I believe he actually doesn't separate hardware and software in his aims: it's all integrated. Hence the original Mac, where the hardware looked great, the interface looked great, and each was tied to the other and designed from a common concept. Same thing now with the "digital hub" idea: it can only work if you see both software and hardware as two facets of a common idea.
Slashdot Mirror
Either you are comparing apples to oranges : Europe[an Union] is tightly-bound politically and economically, which is definitely not the case in Asia in general;
or you just don't care about the numbers. The European Union is the biggest economic zone in the world (GNP, production, intra-zone/inter-country and inter-zone commerce, etc.). And, yes, bigger than NAFTA.
So French Guyana is in Africa? How fascinatingly clueless...
Bandwidth-wise it's a matter of network size. Reliability is quite good -this I know, I work on that stuff.
Indeed. Europe does, too; with quite serious numbers and success rates, BTW. And they aren't particularly plagued by the failures we have come to expect from NASA, recently. How many probes did they lose or mishandle in the last few years?
How exactly would you back up your claims about this? The only other country with "space flight time" (I assume manned flights there; if you're also talking probes or satellites your affirmation is awfully clueless/uninformed) is Russia, which is broke. Mir was scheduled to fall because of lack of funding. Not because of a technical failure.
Just wanted to set things straight.
Not exactly the answer you want... anyway. Open Packages aims at unifying the {Free,Net,Open}BSD, BSDi and [Apple] Darwin package systems. So, when (if?) this gets functional, Unixy software will be nicely packaged.
Did you write the OS that runs in the box under your desk?
Eh?
How true!
Alas the current, uh, wave in the Unix/Linux people is more like "do everything with perl and emacs". Forget about pipes and redirections, say hello to the language with 42 different ways of writing if(x) y; (BTW, is there anyone besides Larry Wall who knows all of these?) and a HUGE libc-duplicated-for-perl.
Yes, you don't even need a gui to work on 90% of the data you want to process everyday. However, other people are not lazy, they have better things to do with their time than tinkering with a compulsory-manpage-and-CLI OS.
I hack a BSD kernel as a job, I hack Drawin and LinuxPPC for pleasure, but I made sure my mom doesn't ever have to see a command line. Her job is not understanding computers, her job is publishing.
I'm tired of people who just can't accept that the complexity of using a computer is not virtue, but lousy design.
BTW, I don't intend this as a flamebait.
They did. I'm running Carbon apps on 8.6.
Did that too.. worthless as a tool but a nice introduction to more interesting stuff; it's a bit like the ubiquitous toy compiler project in CS studies: everyone does that someday and everyone buries the result and uses gcc instead past the deadline. :)
Indeed ! Proving user interfaces is unreasonable in the first place (what exactly are you proving then?)
Not. The idea is to translate automatically from model to code using robust, time-proven translators (in the same way that one can more or less trust a non-RedHat gcc-O2 to translate C code correctly, at least on an x86; or in the way you trust the CPU to accurately execute the machine-language instructions you feed it). Interesting properties are proven on the model.
Perhaps you thought of, say, UML when you read "formal methods"? That's where the disagreement comes from. UML is not a proving tool, it's a design helper, roughly as provable as C (that is, in the real world, not provable at all.)
Automatic translation of proven models is used in embedded software for critical stuff, e.g. antilock brakes, in protocol design and implementation, and such stuff.
There's a huge difference. For many problems you can build a solution that is provably right. On the other hand, given a "random" program you can't generally prove it; ie. there is no systematic method to "prove" software, but there are methods to turn some provable methods and descriptions into programs that keep the same properties. That's what formal languages are about.
I'm not usually that pro-market, but this time it works quite well: let some competition (dis)solve the problem. Here in Paris one can get consumer broadband via France Telecom ADSL: expensive, with a bandwidth cap as the sole traffic limit and a decent backbone serving it; or cable access, less expensive (not cheap yet), with an undersized backbone (though it's good enough for mail and casual browsing), and a monthly upload cap. (and there are third-party DSL providers, cleverly combining the drawbacks of both)
Overall if you want real bandwidth you have to pay more; if you want Joe Average's browsing bandwidth, the cheaper service is okay.
Easy. Make me come back and buy a better version of that game that I love so much.
Civilization, Civilization II, Alpha Centauri, Alien Crossfire... Civ III when it comes out (Sid Meier's, not Call to Power). Each one was even better than the previous one, each one was better than almost any game I've ever played.
The morale is: I'll buy more if the quality is there.
Konqueror is the best I've seen in this regard. Each site that asks for a cookie Konq prompts you for. I know other browsers have this option, but in Konq you can specify to allow or deny all future cookies from a specific domain. It is perhaps Konq's best feature yet.
Funny thing... IE 5.0 for Mac (yes, the Microsoft product) has this ability too. It can also prevent gifs from animating, plugins from loading, etc. That makes it a much better browser than IE 5.5 Win and Netscape 4.*, in my opinion.
(Please note that this is not a troll!)
I'm afraid you are wrong. Motion compensation and iDCT does nothing to do with output resolution.
Please, read what I wrote before disagreeing. Yes, it is true that motion compensation and iDCT have nothing to do with output resolution. This is why I wrote and scaling/smoothing, which you conveniently snipped from the sentence to trick people into believing you are the smart guy.
Damn, I wish Slashcode had a killfile. Maybe I should learn some Perl and add it myself.
In a way, yes: decryption, demuxing, audio decoding, subitiling is done by software. Some high-level MPEG video stuff is done by the software but the bulk of the effort (that is, motion compensation, inverse DCT and scaling/smoothing) is done by the video card (so playing a DVD at 1600x1200 is just as fast as 800x600). This is why this "software decoder" only works with Rage 128/Radeon (and now, GeForce2 MX). Actually it is about the same thing as what you get bundled with said cards in their retail PC version.
BTW, the hardware support in ATI chips is first-rate, quality-wise. The image is better than most hardware decoders (we checked a Hollywood-whatever and an ATI card on the same PC. The ATI's image is much cleaner)
I'm almost certain NetBSD handles this... at least there is a package called [award for the longest package name, btw:] airportbasestationconfig
"in exchange for telling us what it is" There, you found the problem yourself, only you didn't see it.
Did RAMBUS say anything? No, they kept the patent around, well hidden. So that 1-no one knows that someone patented this thing, and 2-no one benefits from the research anyway (since the behaviour of Rambus here is, "hide our patent and wait till someone else builds the same thing")
Guess there are many more romantic millionnaires than non-starving labs. LEO will soon be swamped in a cloud of ashes and tiny engraved cubes blown by other debris.
Rant Isn't there too much stuff loose in orbit already?
Could you demonstrate, please?
I believe he actually doesn't separate hardware and software in his aims: it's all integrated. Hence the original Mac, where the hardware looked great, the interface looked great, and each was tied to the other and designed from a common concept. Same thing now with the "digital hub" idea: it can only work if you see both software and hardware as two facets of a common idea.
That's the reason I use NEdit and not Emacs. I learned these shortcuts and I don't want to change my mindset.
BTW, I've been thinking for a long time now of adding the most common mod*-arrow shortcuts to tcsh, just never got to actually work on it.