Because it is argument from design, and is an important historical approach to proving God's existence that is already taught in introductory Philosophy.
it would be proving... except that it uses defective logic. You might want to try Spinoza instead for more historical relevance.
I think the most level-headed answer to that question that I encountered is 'a psychic[*] utility' - acting as a balance against... many things, many of which raise from fear. Bear in mind, this is about the *belief* - not the existence, name, etc. of God. From this point of view, the utility (or efficacity) is the same that a belief in Science has on other people.
Now, if you actually meant 'what utility an axiom about the existence of God has?", that would be a different kettle of fish. What utility is there in any spiritual undertaking? usually, it means to help dealing with some inner realities, just as natural sciences try to deal with physical realities. 'God' is just a face put to an answer - and a changing one at that, too. I find it interesting to notice similarities between the way spiritual/religious beliefs and scientific theories evolved - they all seem to be products of some continuous attempts to refine approximative images for some things that can never quite be fully described.
[*] I'd have said psychological, but this term was abused into too many twisted meanings
How can you install spyware if you're just playing an avi file?
I hate to break it to you, but it's been done already. It's true that the process is not completely automated yet, unless you happen to have a vulnerable IE. Some social engineering might also prove useful. Anyway, it's lovely to see DRM at work, don't you think?
For what it's worth, I'll have to take something back from the previous post. I DO fault Miyazaki for adding a pointless war plot to Howl's Moving Castle and thus making a mess out of it. That was stupid, especially since in doing so he threw away big chunks of the original storyline that actually helped it make sense. Whatever he tried to do, it ended up as possibly his worst movie so far. Aside from animation, there's very little left to it. Bad, really bad.
Do you even realize why your response is wrong? Does Dependency Hell ring a bell?
Ok, this smells a lot like a troll, but I'll work under the assumption that you were honest and just over-reacted.
I never said 'go to the other extreme' - there's a balance point in between, where some things are safe to be assumed as stable and present on the system, while others aren't necessarily so. As with all balance points, it's hard to stick to it, with the amount of development going on. There's always some new thing that developers haven't settled for a standard way to deal with - and it's a natural evolutive process. You're bound to have some package to point to invoking 'dependency hell', even on stable platforms.
I still think you're framing things wrong. It's mostly an issue of convenience that you seem to be attacking, otherwise as a non-root user you can do a lot of things with well-packaged vanilla rpms (like maintaining an user-level package database, doing $HOME/whatever/ rpm installs and so on). What's missing is a tool to handle that nicely for the famous 'average user' whoever that is, but this is touching a different issue altogether - you still need to educate the user about the way the system works. Simple click-and-run is nice for a few packages, but does not scale well so it's not a one-size-fits-all solution. So while I agree that there's enormous space for evolution, I believe a little caution is in order, as to not throw the baby with the water.
As to your notes, they also seem to be work 'in progress' - and touch this from a different angle, so this post would have been rather off-topic there. I could throw in my $0.02 about some of the topics you raised, but with the current market I can't guarantee my comments would actually add so much value.
If a dependency isn't offered by the OS, it needs to be included in the install.
Do you even realize why this is wrong? DLL hell rings a bell? multiple, even incompatible versions for THE SAME FREAKING DLL?
ok, backtracking... you need a specific library. Say the official distro does not provide it for reasons of patent/copyright (you said VLC, right?) You have 2 choices: track it down yourself and compile/install (hopefully integrated into the package management system) or track down a packager for your distro. Surely you're not the only one trying to get VLC on SuSe.
And how is this different from a random unsupported package in Windows? Why, *I* want k3b in Windows and I want it yesterday. Where's my click-click-click-done procedure??? But I'm straying, as this was not about Windows. The point, that you seem to keep trying to avoid, is that unsupported software is unsupported for some reasons. You should be asking yourself the 'why?' question before bashing. Then, if you found the answer and don't like it, feel free to complain.
That's just stupid. A home user wants to click, click, done.
no, this is stupid actually. A home user has the right to expect click-click-click-done for something he/she purchases. You never purchased VLC. In fact, they are under no obligation to support the said procedure for your distro of choice - why, there are too many distros out there for any vendor to support them all. On the other hand, if SuSe would have employed a maintainer for an official VLC package, then you'd have had a click-click-click-done procedure for it as you do for everything that the distro supports. Or you can try paying someone at videolan to package it for you.
As a result, the application developer can then say, "This program requires Ubuntu v5.2 and up" as opposed to, "This program requires a Linux distribution with Kernel 2.4, KDE 3.4, GLib 2.2, GCC 2.95, etc, etc, etc."
I think you either never did any F/OSS programming or are deliberately trying to confuse the issues. Here are some pointers:
a developer more often than not not have enough resources to test with significant linux distro version; if there are contributors for various distros, they provide packages exactly like that: for distro XYZ version number A.B
if the distro packages the program/lib, then you get exactly that - a binary packade targeted to a specific distro version
if you go into the 'bleeding edge' region, you should not always expect full support for a 'stable' distro.
and so on.
There are points to criticize about package management on various Linux distros, buy you need to bring up valid ones if you want to be taken seriously. And no, personally I don't believe the re are benefits in bundling everything with the app, or static linking. How would you like to have to update a whole bunch of programs just because there was a fix for the xvid version they all statically link against? But this is an ancient argument already.
btw, for SuSe unsupported stuff, you might want to try packman. They happen to have vlc here and you should be able to find all the dependencies there as well. All with "for SuSe version x.y" just as you seem to like it.
10g on a dual 2.0 G5 was about 30% faster than on a dual Xeon 2.4 on Windows
with no other details (at least RAM and RAID settings). Besides, dual Xeon 2.4 is OLD tech from any angle (clock speed, core, chipset, you name it) - it's also really silly to compare it to a 2.0GHz G5; why not a 2x 2.0GHz Xeon if they're into showcasing old server hardware?. And 30% is embarrassingly small a difference to be talking about in this case.
Heh, it's hard not to flame you when you go out of your way to invite it. I'll try to abstain, so please excuse me if I stray.
Nothing was about 'genius', the topic was plotline. You don't see any. Other people do. Simple as that. Since the two positions are either/or, one has to be wrong. All you do here is say you're right - and I'm supposed to believe that. Nobody asked for a 'scene-by-scene deconstruction', just some of those major plotline holes - even one would have been an improvement. Instead, you keep repeating the same non-arguments that one can't even answer because they have no content. Just because someone doesn't (or can't be bothered to) see something it does not necessarily mean there's nothing to be seen. On the other hand, you don't need anyone's blessing to decide how much like a movie - de gustibus non disputandum.
I would only agree with the statement that Miyazaki is not some super-genius visionary. He has better films and... less than better - like any artist. As to which is which, from what you listed back in this thread I would think we only agree on Porco Rosso, although even for that one it would probably be for the wrong reasons. I can't imagine how you'd argue Laputa to be above Spirited Away (which, btw, is no 'holy cow', I just think it's about the most complex film Miyazaki produced so far; but heck, it could have been a fluke for all it matters) but I won't be holding my breath for details. Eh, for whatever good it does, my random opinion on Laputa is that it tries too hard on the 'message' part and kind of flops the rest of the story. But for kids that won't matter much. Besides, having a bone to pick with war is not something I'd fault Ghibli too much for. Even though I'll admit to prefer Grave of the Fireflies in that department.
Anyway, this has gone too far for a silly argument. My initial reply was based on the assumption of dialogue; if I was wrong and it was just a flame, as your replies suggest, then I'm sorry for wasting the time for both of us.
Oh, I'm sorry, I never meant to challenge your holier-than-thou rampage. I merely believed in thinking while watching movies instead of expecting everything spelled out for the lowest common idiot writing for salon.com. I see my mistake now.
And yeah, you've probably seen way more anime than I did. I only continue to watch something that I think worth watching. Contrary to a common opinion, I believe accumulating a big pile of garbage makes zero qualitative difference.
btw, unless you would care to write a factual answer for a change and point out some those major story problems with SA (as that's what I was referring to) please assume that I won't be pursuing this thread any longer. I could find plenty of unsubstantiated bashing on the web if I cared to look for it.
Sorry, but you chose the wrong bone to pick with Spirited Away. That one in particular has one of the most interesting plotLINEs in all of Miyazaki's productions. To put it simply, it's a transformation rite and as such the steps make a lot of sense. The nice touch is in the little details that make sense as well[*]. Perhaps your problem has less to do with the movie and more with the fact that the cultural difference works against your understanding of it? No, you don't have to be Japanese to get it. Just don't look at it through the eyes of a strict Christian education w.r.t. mythos. And, as usual, beauty is in the eye of the beholder - even moreso with art.
[*] that, of course, being utterly relative. However, it's telling when many little things go 'click' without anyone having to explain to you the 'official' interpretation - which might very well be different from yours. That's probably a valid definition of art, something that is able to create experiences of a certain quality inside those that are exposed to it, regardless of what the author intended to express.
We understand that you are, or will soon be, the recipient of a payment totalizing one hundred billion kajillion fafillion dollars. We have taken the liberty to help you with calculating the amount of tax due for the said sum, which amounts to one hundred billion kajillion fafillion dollars minus one. Please send a check or a truck with the appropriate amount of due tax to the usual PA address.
Kindly, The IRS
P.S. Remember to also file in your state tax return.
No OpenMP support in gcc though - you'll have to go the route of expensive compilers to get that (or Intel's one, assuming they don't keep on 'tweaking' it against AMD) This might change by the time AMD actually ships cpus with 4 cores.
It will be fun to see whether multicore Itanium will get out of the door in any quantities worth talking of in time before quad-core Opterons start making inroads.
Sorry, looking at the picture (and assuming that at least the info is correct), it can't be a dual-core as the clock is too high. It would be reasonable though to assume it's hyperthreaded, for all the good that might do (or not). I should have checked first; this way, even if what I argued for, that you can get the system to see/use 4 cpus with just one p4, is still correct, it's also irrelevant here:-)
Using a special kit for the demo would however make sense. Remember, there's a lot more at stake here than just some developers complaining their rented kit is slow (which in itself is not even a big problem - a dev kit needs not be fast in general, just at some specific tasks, like compiling code) This has to show Intel Macs in the best possible light and erase from memory Apple's previous claims of G5 versus P4 performance. It's about sales, profit, stock... too many ripples to afford a shakey demo at this time. Whether they actually did that or not is a rhetorical question right now. What will matter is 1. people's perception of the transition (given that it's still a long way off and right now it's a faith-based thing, no meat yet) and 2. when it comes, the actual transition (and Apple has 1+ year to make sure it's pulled out fine, irrespective of what was in this week's hardware black box)
All in all, it's going to be an interesting year. If only to see how Apple handles phasing out the kit it's still selling without making the buyers feel like second-class citizens after 2007. I'm especially curious to see the spin for this fall's 'back to school' campaign.
The P4 is not multi-processor capable, you need a Xeon for that.
Did you open up the box? If not, for all you know it COULD have been a dual-core hyper-threaded P4EE, which makes for 4 virtual processors. It's not like Steve couldn't have afforded the top-of-the-line Intel p4, you know.
So if anyone has reliable info on what was in the box, please speak up - model numbers and all.
So you want to push the MVC model to the extreme and have a32bit UI process, a background 64bit process for heavy lifting and pass messages between them? Because if the UI libs are 32bit there's no way you can make the full app 64bit and just talk to a 32bit graphics server[*]. Anyway, if that's what yopu have in mind it's the most ass-backwards solution that Apple could choose. Slow as hell for one, full message processing rewrite for another... and Microsoft would laugh all the way to the bank, as Macs would be as slow as a turtle for image/video/audio processing (a.k.a. 'creative use') Also, I doubt many vendors would go for it - better make the app full 32bit and take speed penalty instead of the IPC one. It would be infinitely better for Apple to clean up the UI code for 64bit and simply 'extend' the ABI later - it's not like they don't have time to do it in 1-2 years if they start with 32bit P-M processors and introduce 64bit Xeons/P4s later. But of course that is Yet Another ABI Change.
[*] Hint: why do you think in Tiger only console apps can be 64bit?
On point #4, there is no need to convert the GUI to 64 bit.
This is so wrong I'm not sure where to start. If you want to do 64bit, you need 64bit libs, including UI ones. 32bit UI libs means 32bit graphics only. Now, on PPC it might not make a lot of difference, but with x86 going 64bit actually does - more registers, if not anything else. I seriously doubt you'll see Apple/x86-32 running Photoshop faster than WinXP/x86-64, now that would be an embarassing benchmark. Also with video processing stations, being able to stuff 16GB of memory in a dual x86-64 box and actually use most of it (leaving aside the OS, peripheral mappings and so on) is a boon and it's already here as opposed to using 2G (or at most 3G with tweaks) on x86-32.
That said, Intel seems to be moving to 64bit across the desktop line, so unless they keep old Pentium4s especially for Apple, those OSX desktop machines will be 64bit capable. And Apple will have to make use of that sooner rather than later. So yeah, I would agree there's another switch in the pipe, even if nowhere close to how radical this one is.
If anything, maybe Macs getting better gaming hardware will encourage moving away from DirectX... which will benefit more than just OSX gamers. Provided that MS doesn't play the XBox card too well, that is.
The point is that communism has it built in - the state HAS to have all the power to ensure the 'common' part works. Even Marx said it has to start with an authoritarian regime to 'educate people into sharing'... well, guess what, those leading the authoritarian regime were never 'educated into it' themselves.
People who never were close enough to a Communist country when it started would tend to forget that it begun as an oppressive regime already, often with blood baths (the Soviet gulags were by no means an exception - those counter-revolutionaries had to be 'educated' everywhere) It's easy to just look at the theoretical ideas and say 'hmm... this sounds ok." And what government who had absolute rights on its citizens will ever give them up willingly just because it publicly states "we're heading towards the enlightened all-people-are-equal Communism"?
Though I do think it could make a dent in Linux on the desktop
You have to remember that right now Linux on desktop appears to have the biggest growth potential in China, India, Brasil and other not-so-rich countries where government is beginning to look at open source. Apple won't be able to touch that, not if even Microsoft is having trouble discounting its OS enough to prevent the change.
Other than that, I see no reason they can't compete better with Dell and HP on their own turf.
Volume. Apple does not have it yet and the x86 vendors will notgive it a break. Also, keep in mind the speed of hardware evolution in the x86 world. Apple is not well-adapted to that either. Switching over changes will cost money and that means more expensive, at least for a while.
Finally, there's the 'halo effect' from the server room - if the servers run some flavor of Linux (and as far as I can tell XServe just lost a lot of its appeal, with no clear Intel 64bit roadmap or Opteron support) support will be cheaper if the desktops run the same OS (provided they get moved from Windows) so there's not a lot of reason to move Win -> OSX.
the developer doc ("universal binary guidlines") talks up and down about IA-32 and not a stitch about EM64T/x86-64/amd64. In fact, the ABI section explicitly mentions 32-bit details only (registers, limits on returning values in registers).
Makes one wonder, what kind of game is Apple playing? It does not make much sense to withhold information from developers and say later "actually... it's going to be 64bit too" - but they could be doing that on the scenario that migration starts with notebook-class CPUs (P-M is 32bit only) and towards the end PowerMacs get the 64bit dual-cores or something along these lines.
Goes without saying that a 32bit-only x86 PowerMac would flop when you can buy a 64bit machine from any other vendor and have the Windows version of your image/video processing toolchain run faster/better.
Either way, it will suck for the short term and remains to be seen what the long term will bring.
Also, nothing in the document mentions 64-bit computing. And G4- and G5-specific applications do not run under Rosetta.
Indeed, the docs specifically mention only IA-32. If that means anything, Apple will be at least a few years late to the x86-64 party. Especially if the transition will start with Pentium-M's inside minis and *Books and PowerMacs with 64-bit CPUs will be lagging behind towards the 'end of transition period.' Given the register starvation of the IA-32 ISA, it's going to make it a hard case to sell 32bit OSX for heavy number crunching when you can get the same software in the 64bit flavor for the 'other' OS and even on the 'other' company's CPUs, if they are more efficient. Video processing crowds might just be staring at the sun for some 2 years - or go with 64bit XP altogether.
This is without mentioning the breakneck pace of hardware evolution in the x86 world, as opposed to what Apple used to do in its PPC universe up until now. Or XServes. Speaking of which, why would anyone buy an XServe now? with server-side Linux booming and without the PPC reason the whole exercise seems moot.
This move looks like it will be hurting Apple from so many directions it's hard to count. It would be sad to have them go away altogether, here's hoping that MS will not take too much advantage of that opening, although given that you can drive several trucks through it they might do it involuntarily.
Oh, by the way, welcome Mac devs to the whole new world of security quirks for x86. Hope you will enjoy your stay here.
My guess is Apple is betting on new adopters. Old Macheads will probably be loathe to upgrade from a G4 PowerBook/iBook to Intel unless it will be pretty compelling (something like Yonah/Napa and the OS no longer running smooth enough on the old hw) and there's always the problem with one's old software which doesn't have fat PPC/x86 binaries.
As monkeyboy said, "\(DEVELOPERS!\)\+" and I wonder how many developers will simply not bother to create x86 binaries for the time being or not do it out of spite[*]. Besides, without an x86 Mac testing will be annoying... and with a big enough backlash in this area Apple could take quite a hit.
Of course, these are just random thoughts, but too many Mac users seem to feel like Apple went over to the dark/Borg/Evil Overlord side - and if enough of the 'halo effect' was due to the aforementioned Mac users, adoption of x86 Macs might just falter enough to send Apple spinning.
I guess time will tell, but this seems to be the biggest chance Jobs took - ever.
[*] à la old iPod owners hiding the player in order not to be lumped together with the newcomers - make binaries only for the 'true macs'
Slashdot Mirror
Because it is argument from design, and is an important historical approach to proving God's existence that is already taught in introductory Philosophy.
... except that it uses defective logic. You might want to try Spinoza instead for more historical relevance.
it would be proving
What utility does a belief in God have?
... many things, many of which raise from fear. Bear in mind, this is about the *belief* - not the existence, name, etc. of God. From this point of view, the utility (or efficacity) is the same that a belief in Science has on other people.
I think the most level-headed answer to that question that I encountered is 'a psychic[*] utility' - acting as a balance against
Now, if you actually meant 'what utility an axiom about the existence of God has?", that would be a different kettle of fish. What utility is there in any spiritual undertaking? usually, it means to help dealing with some inner realities, just as natural sciences try to deal with physical realities. 'God' is just a face put to an answer - and a changing one at that, too. I find it interesting to notice similarities between the way spiritual/religious beliefs and scientific theories evolved - they all seem to be products of some continuous attempts to refine approximative images for some things that can never quite be fully described.
[*] I'd have said psychological, but this term was abused into too many twisted meanings
How can you install spyware if you're just playing an avi file?
I hate to break it to you, but it's been done already. It's true that the process is not completely automated yet, unless you happen to have a vulnerable IE. Some social engineering might also prove useful. Anyway, it's lovely to see DRM at work, don't you think?
For what it's worth, I'll have to take something back from the previous post. I DO fault Miyazaki for adding a pointless war plot to Howl's Moving Castle and thus making a mess out of it. That was stupid, especially since in doing so he threw away big chunks of the original storyline that actually helped it make sense. Whatever he tried to do, it ended up as possibly his worst movie so far. Aside from animation, there's very little left to it. Bad, really bad.
Do you even realize why your response is wrong? Does Dependency Hell ring a bell?
Ok, this smells a lot like a troll, but I'll work under the assumption that you were honest and just over-reacted.
I never said 'go to the other extreme' - there's a balance point in between, where some things are safe to be assumed as stable and present on the system, while others aren't necessarily so. As with all balance points, it's hard to stick to it, with the amount of development going on. There's always some new thing that developers haven't settled for a standard way to deal with - and it's a natural evolutive process. You're bound to have some package to point to invoking 'dependency hell', even on stable platforms.
I still think you're framing things wrong. It's mostly an issue of convenience that you seem to be attacking, otherwise as a non-root user you can do a lot of things with well-packaged vanilla rpms (like maintaining an user-level package database, doing $HOME/whatever/ rpm installs and so on). What's missing is a tool to handle that nicely for the famous 'average user' whoever that is, but this is touching a different issue altogether - you still need to educate the user about the way the system works. Simple click-and-run is nice for a few packages, but does not scale well so it's not a one-size-fits-all solution. So while I agree that there's enormous space for evolution, I believe a little caution is in order, as to not throw the baby with the water.
As to your notes, they also seem to be work 'in progress' - and touch this from a different angle, so this post would have been rather off-topic there. I could throw in my $0.02 about some of the topics you raised, but with the current market I can't guarantee my comments would actually add so much value.
Do you even realize why this is wrong? DLL hell rings a bell? multiple, even incompatible versions for THE SAME FREAKING DLL?
ok, backtracking
And how is this different from a random unsupported package in Windows? Why, *I* want k3b in Windows and I want it yesterday. Where's my click-click-click-done procedure??? But I'm straying, as this was not about Windows. The point, that you seem to keep trying to avoid, is that unsupported software is unsupported for some reasons. You should be asking yourself the 'why?' question before bashing. Then, if you found the answer and don't like it, feel free to complain.
That's just stupid. A home user wants to click, click, done.
no, this is stupid actually. A home user has the right to expect click-click-click-done for something he/she purchases. You never purchased VLC. In fact, they are under no obligation to support the said procedure for your distro of choice - why, there are too many distros out there for any vendor to support them all. On the other hand, if SuSe would have employed a maintainer for an official VLC package, then you'd have had a click-click-click-done procedure for it as you do for everything that the distro supports. Or you can try paying someone at videolan to package it for you.
As a result, the application developer can then say, "This program requires Ubuntu v5.2 and up" as opposed to, "This program requires a Linux distribution with Kernel 2.4, KDE 3.4, GLib 2.2, GCC 2.95, etc, etc, etc."
I think you either never did any F/OSS programming or are deliberately trying to confuse the issues. Here are some pointers:
and so on.
There are points to criticize about package management on various Linux distros, buy you need to bring up valid ones if you want to be taken seriously. And no, personally I don't believe the re are benefits in bundling everything with the app, or static linking. How would you like to have to update a whole bunch of programs just because there was a fix for the xvid version they all statically link against? But this is an ancient argument already.
btw, for SuSe unsupported stuff, you might want to try packman. They happen to have vlc here and you should be able to find all the dependencies there as well. All with "for SuSe version x.y" just as you seem to like it.
with no other details (at least RAM and RAID settings). Besides, dual Xeon 2.4 is OLD tech from any angle (clock speed, core, chipset, you name it) - it's also really silly to compare it to a 2.0GHz G5; why not a 2x 2.0GHz Xeon if they're into showcasing old server hardware?. And 30% is embarrassingly small a difference to be talking about in this case.
Heh, it's hard not to flame you when you go out of your way to invite it. I'll try to abstain, so please excuse me if I stray.
... less than better - like any artist. As to which is which, from what you listed back in this thread I would think we only agree on Porco Rosso, although even for that one it would probably be for the wrong reasons. I can't imagine how you'd argue Laputa to be above Spirited Away (which, btw, is no 'holy cow', I just think it's about the most complex film Miyazaki produced so far; but heck, it could have been a fluke for all it matters) but I won't be holding my breath for details. Eh, for whatever good it does, my random opinion on Laputa is that it tries too hard on the 'message' part and kind of flops the rest of the story. But for kids that won't matter much. Besides, having a bone to pick with war is not something I'd fault Ghibli too much for. Even though I'll admit to prefer Grave of the Fireflies in that department.
Nothing was about 'genius', the topic was plotline. You don't see any. Other people do. Simple as that. Since the two positions are either/or, one has to be wrong. All you do here is say you're right - and I'm supposed to believe that. Nobody asked for a 'scene-by-scene deconstruction', just some of those major plotline holes - even one would have been an improvement. Instead, you keep repeating the same non-arguments that one can't even answer because they have no content. Just because someone doesn't (or can't be bothered to) see something it does not necessarily mean there's nothing to be seen. On the other hand, you don't need anyone's blessing to decide how much like a movie - de gustibus non disputandum.
I would only agree with the statement that Miyazaki is not some super-genius visionary. He has better films and
Anyway, this has gone too far for a silly argument. My initial reply was based on the assumption of dialogue; if I was wrong and it was just a flame, as your replies suggest, then I'm sorry for wasting the time for both of us.
Oh, I'm sorry, I never meant to challenge your holier-than-thou rampage. I merely believed in thinking while watching movies instead of expecting everything spelled out for the lowest common idiot writing for salon.com. I see my mistake now.
And yeah, you've probably seen way more anime than I did. I only continue to watch something that I think worth watching. Contrary to a common opinion, I believe accumulating a big pile of garbage makes zero qualitative difference.
btw, unless you would care to write a factual answer for a change and point out some those major story problems with SA (as that's what I was referring to) please assume that I won't be pursuing this thread any longer. I could find plenty of unsubstantiated bashing on the web if I cared to look for it.
Sorry, but you chose the wrong bone to pick with Spirited Away. That one in particular has one of the most interesting plotLINEs in all of Miyazaki's productions. To put it simply, it's a transformation rite and as such the steps make a lot of sense. The nice touch is in the little details that make sense as well[*]. Perhaps your problem has less to do with the movie and more with the fact that the cultural difference works against your understanding of it? No, you don't have to be Japanese to get it. Just don't look at it through the eyes of a strict Christian education w.r.t. mythos. And, as usual, beauty is in the eye of the beholder - even moreso with art.
[*] that, of course, being utterly relative. However, it's telling when many little things go 'click' without anyone having to explain to you the 'official' interpretation - which might very well be different from yours. That's probably a valid definition of art, something that is able to create experiences of a certain quality inside those that are exposed to it, regardless of what the author intended to express.
... the US was reported to have gotten closer to the South pole by roughly the same amount.
Hello Dr. Evil,
We understand that you are, or will soon be, the recipient of a payment totalizing one hundred billion kajillion fafillion dollars. We have taken the liberty to help you with calculating the amount of tax due for the said sum, which amounts to one hundred billion kajillion fafillion dollars minus one. Please send a check or a truck with the appropriate amount of due tax to the usual PA address.
Kindly,
The IRS
P.S. Remember to also file in your state tax return.
No OpenMP support in gcc though - you'll have to go the route of expensive compilers to get that (or Intel's one, assuming they don't keep on 'tweaking' it against AMD) This might change by the time AMD actually ships cpus with 4 cores.
It will be fun to see whether multicore Itanium will get out of the door in any quantities worth talking of in time before quad-core Opterons start making inroads.
Sorry, looking at the picture (and assuming that at least the info is correct), it can't be a dual-core as the clock is too high. It would be reasonable though to assume it's hyperthreaded, for all the good that might do (or not). I should have checked first; this way, even if what I argued for, that you can get the system to see/use 4 cpus with just one p4, is still correct, it's also irrelevant here :-)
... too many ripples to afford a shakey demo at this time. Whether they actually did that or not is a rhetorical question right now. What will matter is 1. people's perception of the transition (given that it's still a long way off and right now it's a faith-based thing, no meat yet) and 2. when it comes, the actual transition (and Apple has 1+ year to make sure it's pulled out fine, irrespective of what was in this week's hardware black box)
Using a special kit for the demo would however make sense. Remember, there's a lot more at stake here than just some developers complaining their rented kit is slow (which in itself is not even a big problem - a dev kit needs not be fast in general, just at some specific tasks, like compiling code) This has to show Intel Macs in the best possible light and erase from memory Apple's previous claims of G5 versus P4 performance. It's about sales, profit, stock
All in all, it's going to be an interesting year. If only to see how Apple handles phasing out the kit it's still selling without making the buyers feel like second-class citizens after 2007. I'm especially curious to see the spin for this fall's 'back to school' campaign.
The P4 is not multi-processor capable, you need a Xeon for that.
Did you open up the box? If not, for all you know it COULD have been a dual-core hyper-threaded P4EE, which makes for 4 virtual processors. It's not like Steve couldn't have afforded the top-of-the-line Intel p4, you know.
So if anyone has reliable info on what was in the box, please speak up - model numbers and all.
So you want to push the MVC model to the extreme and have a32bit UI process, a background 64bit process for heavy lifting and pass messages between them? Because if the UI libs are 32bit there's no way you can make the full app 64bit and just talk to a 32bit graphics server[*]. Anyway, if that's what yopu have in mind it's the most ass-backwards solution that Apple could choose. Slow as hell for one, full message processing rewrite for another ... and Microsoft would laugh all the way to the bank, as Macs would be as slow as a turtle for image/video/audio processing (a.k.a. 'creative use') Also, I doubt many vendors would go for it - better make the app full 32bit and take speed penalty instead of the IPC one. It would be infinitely better for Apple to clean up the UI code for 64bit and simply 'extend' the ABI later - it's not like they don't have time to do it in 1-2 years if they start with 32bit P-M processors and introduce 64bit Xeons/P4s later. But of course that is Yet Another ABI Change.
[*] Hint: why do you think in Tiger only console apps can be 64bit?
yeah, downloading a pirated copy and burning it on a cd is a waste of bandwith and a perfectly good blank. Looks like negative value to me.
On point #4, there is no need to convert the GUI to 64 bit.
This is so wrong I'm not sure where to start. If you want to do 64bit, you need 64bit libs, including UI ones. 32bit UI libs means 32bit graphics only. Now, on PPC it might not make a lot of difference, but with x86 going 64bit actually does - more registers, if not anything else. I seriously doubt you'll see Apple/x86-32 running Photoshop faster than WinXP/x86-64, now that would be an embarassing benchmark. Also with video processing stations, being able to stuff 16GB of memory in a dual x86-64 box and actually use most of it (leaving aside the OS, peripheral mappings and so on) is a boon and it's already here as opposed to using 2G (or at most 3G with tweaks) on x86-32.
That said, Intel seems to be moving to 64bit across the desktop line, so unless they keep old Pentium4s especially for Apple, those OSX desktop machines will be 64bit capable. And Apple will have to make use of that sooner rather than later. So yeah, I would agree there's another switch in the pipe, even if nowhere close to how radical this one is.
If anything, maybe Macs getting better gaming hardware will encourage moving away from DirectX ... which will benefit more than just OSX gamers. Provided that MS doesn't play the XBox card too well, that is.
The point is that communism has it built in - the state HAS to have all the power to ensure the 'common' part works. Even Marx said it has to start with an authoritarian regime to 'educate people into sharing' ... well, guess what, those leading the authoritarian regime were never 'educated into it' themselves.
... this sounds ok." And what government who had absolute rights on its citizens will ever give them up willingly just because it publicly states "we're heading towards the enlightened all-people-are-equal Communism"?
People who never were close enough to a Communist country when it started would tend to forget that it begun as an oppressive regime already, often with blood baths (the Soviet gulags were by no means an exception - those counter-revolutionaries had to be 'educated' everywhere) It's easy to just look at the theoretical ideas and say 'hmm
Though I do think it could make a dent in Linux on the desktop
You have to remember that right now Linux on desktop appears to have the biggest growth potential in China, India, Brasil and other not-so-rich countries where government is beginning to look at open source. Apple won't be able to touch that, not if even Microsoft is having trouble discounting its OS enough to prevent the change.
Other than that, I see no reason they can't compete better with Dell and HP on their own turf.
Volume. Apple does not have it yet and the x86 vendors will notgive it a break. Also, keep in mind the speed of hardware evolution in the x86 world. Apple is not well-adapted to that either. Switching over changes will cost money and that means more expensive, at least for a while.
Finally, there's the 'halo effect' from the server room - if the servers run some flavor of Linux (and as far as I can tell XServe just lost a lot of its appeal, with no clear Intel 64bit roadmap or Opteron support) support will be cheaper if the desktops run the same OS (provided they get moved from Windows) so there's not a lot of reason to move Win -> OSX.
the developer doc ("universal binary guidlines") talks up and down about IA-32 and not a stitch about EM64T/x86-64/amd64. In fact, the ABI section explicitly mentions 32-bit details only (registers, limits on returning values in registers).
... it's going to be 64bit too" - but they could be doing that on the scenario that migration starts with notebook-class CPUs (P-M is 32bit only) and towards the end PowerMacs get the 64bit dual-cores or something along these lines.
Makes one wonder, what kind of game is Apple playing? It does not make much sense to withhold information from developers and say later "actually
Goes without saying that a 32bit-only x86 PowerMac would flop when you can buy a 64bit machine from any other vendor and have the Windows version of your image/video processing toolchain run faster/better.
Either way, it will suck for the short term and remains to be seen what the long term will bring.
Also, nothing in the document mentions 64-bit computing. And G4- and G5-specific applications do not run under Rosetta.
Indeed, the docs specifically mention only IA-32. If that means anything, Apple will be at least a few years late to the x86-64 party. Especially if the transition will start with Pentium-M's inside minis and *Books and PowerMacs with 64-bit CPUs will be lagging behind towards the 'end of transition period.' Given the register starvation of the IA-32 ISA, it's going to make it a hard case to sell 32bit OSX for heavy number crunching when you can get the same software in the 64bit flavor for the 'other' OS and even on the 'other' company's CPUs, if they are more efficient. Video processing crowds might just be staring at the sun for some 2 years - or go with 64bit XP altogether.
This is without mentioning the breakneck pace of hardware evolution in the x86 world, as opposed to what Apple used to do in its PPC universe up until now. Or XServes. Speaking of which, why would anyone buy an XServe now? with server-side Linux booming and without the PPC reason the whole exercise seems moot.
This move looks like it will be hurting Apple from so many directions it's hard to count. It would be sad to have them go away altogether, here's hoping that MS will not take too much advantage of that opening, although given that you can drive several trucks through it they might do it involuntarily.
Oh, by the way, welcome Mac devs to the whole new world of security quirks for x86. Hope you will enjoy your stay here.
And therein lies the problem. From Apple's docs:
Now, how comfortable are you with items 2 and 4? My bet is that Photoshop owners are not the only ones that see a forced software re-buy.
My guess is Apple is betting on new adopters. Old Macheads will probably be loathe to upgrade from a G4 PowerBook/iBook to Intel unless it will be pretty compelling (something like Yonah/Napa and the OS no longer running smooth enough on the old hw) and there's always the problem with one's old software which doesn't have fat PPC/x86 binaries.
... and with a big enough backlash in this area Apple could take quite a hit.
As monkeyboy said, "\(DEVELOPERS!\)\+" and I wonder how many developers will simply not bother to create x86 binaries for the time being or not do it out of spite[*]. Besides, without an x86 Mac testing will be annoying
Of course, these are just random thoughts, but too many Mac users seem to feel like Apple went over to the dark/Borg/Evil Overlord side - and if enough of the 'halo effect' was due to the aforementioned Mac users, adoption of x86 Macs might just falter enough to send Apple spinning.
I guess time will tell, but this seems to be the biggest chance Jobs took - ever.
[*] à la old iPod owners hiding the player in order not to be lumped together with the newcomers - make binaries only for the 'true macs'