I know, no one cares about the reasons, but that battery scene in the movie just completely ruined it for me (and it was too loud, but anyway).
Maybe a more reasonable explanation for the existence of the Matrix and the need for humans to wake up and do something, would be if say, in the not so distant future the Earth really does get too polluted and star travel is infeasible. The existing population decides to build this facility where everyone can all be part of a virtual world where everything is nice, peachy and comfortable, as a way to minimize energy use (everyone fits neatly in a small space that does not require much heating or cooling and eats very little). This sounds like a great idea so most everyone joins up and the simulation starts.
After a few years or generations people forget the ugly real world altogether. That's when machines that were supposed to look after the simulation become uppity and start having ideas of their own, such as for example having plans to halt the simulation and kill everybody, once they have complete control over it, which isn't just yet.
A few people who had refused to join the simulation somehow become aware of this fact. The task is to wake up everybody to fix the problem before the rogue machines kill them.
You would have a largely still neutral Matrix with some Humans trying to find out how to alert everybody of the fact of the existence of the Matrix (they don't have the instruction manual so they must work a homegrown solution) and a number of rogue AI trying to take over the whole lot, starting with neutralizing the independent Humans. Basically you would have two factions with independent super-human powers in this simulation with most everything else looking normal, but this time with some consistency.
The story would be largely untouched but it may be a bit more beliveable than the energy bit.
Don't you think the US has a little bit of a head start? Or India and Pakistan (and Israel!) for that matter. Restarting an arm race around that argument may not be a great idea for world peace. Then again, what do I know?
I don't know about Nevada, but here in Australia we have very definitely a cancer hotspot around the old nuclear testing ground that the British used in the Australian desert. The hotspot is in the Aboriginal population so for the longest time no one cared, but thankfully now people do.
The people most exposed in Nevada would have been the US military and these guys move around a lot.
There are actually rules for those. Under fair use you can copy 10% of a book without infringing copyright for the purpose of research and teaching, for example.
For mathematicians, 9.9999... infinitely repeating is *exactly* equal to 10. They are different representation of the same number. They are both solutions to the equation:
Myself I find the MS-Windows window manager more than a little annoying. Click to focus, auto-raise on is a nighmare. No way to push back a window, no shading. I like multiple desktops too. I know there are hacks to get around some of these but they don't work as well as they do under Linux. BTW I have similar problems with Acqua/Quartz under MacOS/X.
Windows people usually whinge about the lack of proper drag-and-drop and non-text cut-and-paste in X or that they don't like the way cut and paste is done (with the middle button). That may be a valid criticism but these problems don't make the X desktop unusable, you just need to work in a slightly different way.
My point is that if you are used to either X or Windows you'll have trouble adjusting to the other and that this goes both ways. So if you are an X person KDE/Gnome are very acceptable, yes.
The flip side of this comment is that Linux was just as *good* in 1994 as it is now, except for a few minor enhancements such as support for SMP, non-x86 CPUs (including mainframes), 3D hardware, multimedia, clusters (Beowulf) and more. On the appearance thing we went from twm + fixed font to KDE/Gnome which really does not make much of a difference.
Of cource in 1994 we had a striving free software movement which has changed all that much, still producing a usable C/C++ compiler and what basically is an editor, and now we only have a few extra tens of thousands of open-source projects which don't have much of an impact, really. Except maybe for that Office rip-off and a rehash of that old browser, remember Netscape?
Meanwhile on the windows front, we have had 2 *complete* redesign of the GUI. Woah (3.1 -> 95 -> XP). And now we have a real kernel which does not crash all that much. An now Windows works on SMP systems, if you pay more, but only works on Intel CPUs, sorry (used to be that it worked on Alphas and PowerPCs, but no longer).
Technically at the kernel level Windows has been playing catch up, not the other way around. DOS, win3.x and 9x were all pieces of c*@p which only survived because of the installed based and Microsoft monopoly practices. Now win2k and XP are technically OK, but please don't rewrite history.
What you are complaining about is that Linux never had a DOS moment. And this is bad why?
Because specifying the start of the sequence in the suite of digits of Pi would be a number at least as large as the installer itself (and in all likelyhood much much larger), therefore nothing is gained.
Please judge a developer primarily with the quality of his/her code.
That good old Ar'pi has one or many chips on his shoulder does not worry me in the least. On the other hand not having mplayer and mencoder anymore would make Linux a less pleasant experience.
People don't choose to use Windows and Word due to the inter-personal qualities of William H. Gates Junior.
> The other kind is where someone writes software > they want lots of other people to use and they > develop it using the open source model. > If you are making the second kind of software > you need to make a gui.
Indeed I'm often reminded of the stunning GUIs associated with gcc, samba, apache and the Linux kernel.
As you know, Monsanto, besides marketing its genetically-altered plants, also produces RoundUp, which is the #1 used pesticide.
Monsanto has every incentive to increase dependency on pesticides, not the opposite. Its genetically-modified plants resist Roundup, so you can put more of the stuff on your crops, so more pests are killed more surely, but not what you want to grow.
It's beautiful: Monsanto sells you both the seeds and the pesticide, and soon you can only use this combination and you are locked in with them under their term.
The fear is also that the roundup-resistant gene will spread to other plants (it can happen) and make them somewhat roundup-resistant. Then you have to use even more of the stuff and now you have a vicious circle you can't get out of.
Basically this lady wanted a Linux distribution that installed cleanly, completely and simply in a moderately complex environment and she found none that fitted the bill.
It has been said elsewhere that Linux will not be ready for the desktop until something like that happens for the majority of people even vaguely interested in Linux.
My opinion is that this goal is not only elusive and very hard to get to (even Microsoft is not able to do it all that well), it's also where we do not want to be. Personally I don't really care how hard the installation is as long as I can get there somehow, and that afterwards the system actually works. The opposite (easy installation but flakey system) holds no interest for me.
Also I have the somewhat elitist view that if people don't want to be bothered with installation and interface problems they are welcome to pay the Microsoft (or Apple) tax. Sure Linux will never take over Microsoft as long as this is the case but is this desirable?
Finally even if Linux were uniformly better than Microsoft in every aspect (not so far off) even useability, most PC users would still not use it, not by choice but by ignorance, lack of pre-installation deals, poor OEM support, lack of misleading and overwhelming marketing support, etc.
In other words what is limiting Linux on the desktop today are not the technical aspects but all the rest, and no amount of programming will change that.
A RH.0 release really did mean that they'd changed something big, like a kernel major version, a new glibc, a new compiler system. Historically the.0 releases have been bug-ridden, especially the 7.0
Reviews are pretty much useless. You need to work with a distribution that works for you. You might find an important problem with a new release right away, or later down the track. If you have a distribution that works for you now, why upgrade at all?
The `export' keyword was supposed to help with this, but it really doesn't.
If you think about it, the compiler has to deal with Foo.cpp in a different way as it does with main.cpp. In some sense Foo.cpp cannot be compiled just once. It needs to be reprocessed by the compiler as many times as Foo gets instanciated with different types.
It is probably possible to do this (treat Foo.cpp differently) but it is a major endeavour, and probably not all that useful.
It's useful to think of template as advanced macros. As such all of their definition belongs to.h files.
Note that Java does not deal with templates at all. It does all of its genericity with inheritance and dynamic binding. This works well but carries a run-time cost. Like you mention, you can do the same thing in C++, and you will incur the same cost as with Java.
Templates solve a different problem, the one when the run-time efficiency is paramount.
Hang on, you write that you record under Linux with mencoder?
What about the sound problem, has it been fixed (it used to be that mencoder would record the picture directly from the capture card, but not the sound. At all. Is that still true?).
I have had lots of problems with nvrec and sync. Sync is fine in the NUV format, but once transcoded (with mencoder) it's horrible.
My solution of choice has been `vcr' even though it hasn't been updated in more than a year. No need to transcode since it records directly in DivX format (the hacked MS version, 3.x). Sync is perfect even after 1h.
Slashdot Mirror
I know, no one cares about the reasons, but that battery scene in the movie just completely ruined it for me (and it was too loud, but anyway).
Maybe a more reasonable explanation for the existence of the Matrix and the need for humans to wake up and do something, would be if say, in the not so distant future the Earth really does get too polluted and star travel is infeasible. The existing population decides to build this facility where everyone can all be part of a virtual world where everything is nice, peachy and comfortable, as a way to minimize energy use (everyone fits neatly in a small space that does not require much heating or cooling and eats very little). This sounds like a great idea so most everyone joins up and the simulation starts.
After a few years or generations people forget the ugly real world altogether. That's when machines that were supposed to look after the simulation become uppity and start having ideas of their own, such as for example having plans to halt the simulation and kill everybody, once they have complete control over it, which isn't just yet.
A few people who had refused to join the simulation somehow become aware of this fact. The task is to wake up everybody to fix the problem before the rogue machines kill them.
You would have a largely still neutral Matrix with some Humans trying to find out how to alert everybody of the fact of the existence of the Matrix (they don't have the instruction manual so they must work a homegrown solution) and a number of rogue AI trying to take over the whole lot, starting with neutralizing the independent Humans. Basically you would have two factions with independent super-human powers in this simulation with most everything else looking normal, but this time with some consistency.
The story would be largely untouched but it may be a bit more beliveable than the energy bit.
In floating point calculations, yes. In integer calculations it is not quite as good.
Question for you: do you think there is a daycare center at Guantanamo Bay?
Correct, and you can practice this in the comfort of your own home in various PC flight simulators, Falcon 4 for example (MS-windows required).
Don't you think the US has a little bit of a head start? Or India and Pakistan (and Israel!) for that matter. Restarting an arm race around that argument may not be a great idea for world peace. Then again, what do I know?
I don't know about Nevada, but here in Australia we have very definitely a cancer hotspot around the old nuclear testing ground that the British used in the Australian desert. The hotspot is in the Aboriginal population so for the longest time no one cared, but thankfully now people do.
The people most exposed in Nevada would have been the US military and these guys move around a lot.
There are actually rules for those. Under fair use you can copy 10% of a book without infringing copyright for the purpose of research and teaching, for example.
For mathematicians, 9.9999... infinitely repeating is *exactly* equal to 10. They are different representation of the same number. They are both solutions to the equation:
10x - x = 90
Radius vs. Diameter, maybe?
There will be a tech bubble again, for sure, but not centered on IT. Maybe what comes after biotech.
Myself I find the MS-Windows window manager more than a little annoying. Click to focus, auto-raise on is a nighmare. No way to push back a window, no shading. I like multiple desktops too. I know there are hacks to get around some of these but they don't work as well as they do under Linux. BTW I have similar problems with Acqua/Quartz under MacOS/X.
Windows people usually whinge about the lack of proper drag-and-drop and non-text cut-and-paste in X or that they don't like the way cut and paste is done (with the middle button). That may be a valid criticism but these problems don't make the X desktop unusable, you just need to work in a slightly different way.
My point is that if you are used to either X or Windows you'll have trouble adjusting to the other and that this goes both ways. So if you are an X person KDE/Gnome are very acceptable, yes.
The cheap Sun workstations are just PCs with Sparcs instead of Pentium, running Solaris. IDE disks, PCI bus, USB, you name it.
No difference, not even so much in price. You can run Linux on these boxes too if you want.
You are quite right of course,
The flip side of this comment is that Linux was just as *good* in 1994 as it is now, except for a few minor enhancements such as support for SMP, non-x86 CPUs (including mainframes), 3D hardware, multimedia, clusters (Beowulf) and more. On the appearance thing we went from twm + fixed font to KDE/Gnome which really does not make much of a difference.
Of cource in 1994 we had a striving free software movement which has changed all that much, still producing a usable C/C++ compiler and what basically is an editor, and now we only have a few extra tens of thousands of open-source projects which don't have much of an impact, really. Except maybe for that Office rip-off and a rehash of that old browser, remember Netscape?
Meanwhile on the windows front, we have had 2 *complete* redesign of the GUI. Woah (3.1 -> 95 -> XP). And now we have a real kernel which does not crash all that much. An now Windows works on SMP systems, if you pay more, but only works on Intel CPUs, sorry (used to be that it worked on Alphas and PowerPCs, but no longer).
Technically at the kernel level Windows has been playing catch up, not the other way around. DOS, win3.x and 9x were all pieces of c*@p which only survived because of the installed based and Microsoft monopoly practices. Now win2k and XP are technically OK, but please don't rewrite history.
What you are complaining about is that Linux never had a DOS moment. And this is bad why?
This 4% mortality rate has nothing to do whether it spread or not. The total number of cases do.
Because specifying the start of the sequence in the suite of digits of Pi would be a number at least as large as the installer itself (and in all likelyhood much much larger), therefore nothing is gained.
Actually Mplayer has excellent documentation. In several languages too, complete, and pretty accurate.
Everything of interest that I've seen in the mailing list has more or less made it in the documentation.
Please judge a developer primarily with the quality of his/her code.
That good old Ar'pi has one or many chips on his shoulder does not worry me in the least. On the other hand not having mplayer and mencoder anymore would make Linux a less pleasant experience.
People don't choose to use Windows and Word due to the inter-personal qualities of William H. Gates Junior.
Hi,
You write:
> The other kind is where someone writes software
> they want lots of other people to use and they
> develop it using the open source model.
> If you are making the second kind of software
> you need to make a gui.
Indeed I'm often reminded of the stunning GUIs associated with gcc, samba, apache and the Linux kernel.
Sorry, I think your facts are not quite correct.
As you know, Monsanto, besides marketing its genetically-altered plants, also produces RoundUp, which is the #1 used pesticide.
Monsanto has every incentive to increase dependency on pesticides, not the opposite. Its genetically-modified plants resist Roundup, so you can put more of the stuff on your crops, so more pests are killed more surely, but not what you want to grow.
It's beautiful: Monsanto sells you both the seeds and the pesticide, and soon you can only use this combination and you are locked in with them under their term.
The fear is also that the roundup-resistant gene will spread to other plants (it can happen) and make them somewhat roundup-resistant. Then you have to use even more of the stuff and now you have a vicious circle you can't get out of.
Great, isn't it?
Basically this lady wanted a Linux distribution that installed cleanly, completely and simply in a moderately complex environment and she found none that fitted the bill.
It has been said elsewhere that Linux will not be ready for the desktop until something like that happens for the majority of people even vaguely interested in Linux.
My opinion is that this goal is not only elusive and very hard to get to (even Microsoft is not able to do it all that well), it's also where we do not want to be. Personally I don't really care how hard the installation is as long as I can get there somehow, and that afterwards the system actually works. The opposite (easy installation but flakey system) holds no interest for me.
Also I have the somewhat elitist view that if people don't want to be bothered with installation and interface problems they are welcome to pay the Microsoft (or Apple) tax. Sure Linux will never take over Microsoft as long as this is the case but is this desirable?
Finally even if Linux were uniformly better than Microsoft in every aspect (not so far off) even useability, most PC users would still not use it, not by choice but by ignorance, lack of pre-installation deals, poor OEM support, lack of misleading and overwhelming marketing support, etc.
In other words what is limiting Linux on the desktop today are not the technical aspects but all the rest, and no amount of programming will change that.
You chose the wrong example. Actually sendmail.org is pretty good at providing security patches for older versions of sendmail.
See this
A RH .0 release really did mean that they'd changed something big, like a kernel major version, a new glibc, a new compiler system. Historically the .0 releases have been bug-ridden, especially the 7.0
Reviews are pretty much useless. You need to work with a distribution that works for you. You might find an important problem with a new release right away, or later down the track. If you have a distribution that works for you now, why upgrade at all?
The `export' keyword was supposed to help with this, but it really doesn't.
.h files.
If you think about it, the compiler has to deal with Foo.cpp in a different way as it does with main.cpp. In some sense Foo.cpp cannot be compiled just once. It needs to be reprocessed by the compiler as many times as Foo gets instanciated with different types.
It is probably possible to do this (treat Foo.cpp differently) but it is a major endeavour, and probably not all that useful.
It's useful to think of template as advanced macros. As such all of their definition belongs to
Note that Java does not deal with templates at all. It does all of its genericity with inheritance and dynamic binding. This works well but carries a run-time cost. Like you mention, you can do the same thing in C++, and you will incur the same cost as with Java.
Templates solve a different problem, the one when the run-time efficiency is paramount.
Thanks!
I had no idea the sound problem had been fixed. I'll give it a try and hopefully get rid of `vcr'.
Fantastic!
Hang on, you write that you record under Linux with mencoder?
What about the sound problem, has it been fixed (it used to be that mencoder would record the picture directly from the capture card, but not the sound. At all. Is that still true?).
I have had lots of problems with nvrec and sync. Sync is fine in the NUV format, but once transcoded (with mencoder) it's horrible.
My solution of choice has been `vcr' even though it hasn't been updated in more than a year. No need to transcode since it records directly in DivX format (the hacked MS version, 3.x). Sync is perfect even after 1h.