The parent post makes a good point. It's a semantics issue. In your example, it can be (and is being) successfully argued that we don't (and can't) know any law of nature. All we know are laws of interaction between nature and the human mind, since all we can get is nature + mind, not nature all by itself.
There is no "Ogg codec". Ogg is the encapsulation format (think AVI or MPEG), Vorbis is the audio codec (think MP3 or WMA). Saying "Ogg codec" is pretty much the same as saying "AVI codec".
1. Er, if you read carefully again (or, to be more precise, probablyt for the *first* time), I specifically used the past tense in relation to the P4. My main system currently is a dual Opteron. 2. Of course, you're entitled to your opinion.
Nod, point taken - however CPU activity resulting from compiling is very differently patterned than database activity. I still maintain that there is no way a dual P3/500MHz would beat a single P4/1.5GHz. For reference, I own a dual P3/500 (Katmai) system and used to own a 2 GHz P4 system - 512M RAM in both cases. Never benchmarked the Gentoo bootstrapping on either one, but the P4 definitely felt faster.
Even there. It's pretty hard to write multithreaded programs right. Many implementations use mutexes and spinlocks, and those are hit heavily on SMP system due to cache coherency issues.
I can't really believe that, unless you ran your hard drive(s) in PIO mode and kept on poking them while running the bootstrap. A dual SMP system has around 85-90% of a processor of the same generation but with double the speed. In your example, two P3/500 should about have matched an 1 GHz P3, and that's *about*. More realistically, they'd have run on par with a 900 MHz P3. There is no *way* that a dual P3/500 system will match an 1 GHz P4, not to speak of your claims of beating an 1.5 GHz P4.
It's a classic marketing device - the "ha, ha, you can't have it!" technique... Builds up a huge expectation momentum, provided you have the resources, and Google certainly does.
Then neither Java nor CMUCL have compilers... Don't know about CMUCL, but Java has a lot of compilers (both source -> binary, and bytecode -> binary ones).
Hmmm... Yup, vira would probably be about right:) I wonder when it actually entered the language - I'll be ethymology-googling after work today now that I'm curious:)
If you know English, you should know that virii is the plural for virus (cactus - cactii, fungus - fungi, etc.) "Virus" entered English indirectly from Latin.
You'd also have to get the/etc/make.profile symlink point to the new 2004.1 directory under/usr/portage/profiles to get full equivalence to the new "release".
Re:A first step, but Unicode support is incomplete
on
JOE Hits 3.0
·
· Score: 1
Uh, it doesn't claim to have full Unicode support. What it claims to have is UTF-8 encoding support. Check your sources. Release announcement on SourceForge
Slashdot Mirror
The parent post makes a good point. It's a semantics issue. In your example, it can be (and is being) successfully argued that we don't (and can't) know any law of nature. All we know are laws of interaction between nature and the human mind, since all we can get is nature + mind, not nature all by itself.
Still, outlining the least informative / most cheaply sensationalistic criticisms doesn't exactly speak highly of the journalistic standards employed.
There is no "Ogg codec". Ogg is the encapsulation format (think AVI or MPEG), Vorbis is the audio codec (think MP3 or WMA). Saying "Ogg codec" is pretty much the same as saying "AVI codec".
1. Er, if you read carefully again (or, to be more precise, probablyt for the *first* time), I specifically used the past tense in relation to the P4. My main system currently is a dual Opteron.
2. Of course, you're entitled to your opinion.
Nod, point taken - however CPU activity resulting from compiling is very differently patterned than database activity. I still maintain that there is no way a dual P3/500MHz would beat a single P4/1.5GHz. For reference, I own a dual P3/500 (Katmai) system and used to own a 2 GHz P4 system - 512M RAM in both cases. Never benchmarked the Gentoo bootstrapping on either one, but the P4 definitely felt faster.
Even there. It's pretty hard to write multithreaded programs right. Many implementations use mutexes and spinlocks, and those are hit heavily on SMP system due to cache coherency issues.
I can't really believe that, unless you ran your hard drive(s) in PIO mode and kept on poking them while running the bootstrap. A dual SMP system has around 85-90% of a processor of the same generation but with double the speed. In your example, two P3/500 should about have matched an 1 GHz P3, and that's *about*. More realistically, they'd have run on par with a 900 MHz P3. There is no *way* that a dual P3/500 system will match an 1 GHz P4, not to speak of your claims of beating an 1.5 GHz P4.
Touche :)
It's a classic marketing device - the "ha, ha, you can't have it!" technique... Builds up a huge expectation momentum, provided you have the resources, and Google certainly does.
You're a bit too late. Apple got a patent on that one :)
Then neither Java nor CMUCL have compilers...
Don't know about CMUCL, but Java has a lot of compilers (both source -> binary, and bytecode -> binary ones).
It's not that difficult to get a second descriptor table entry mapping the same memory area but marked as data, not code. Voila, write.
Nope, didn't originate from any computer science scene, be it from the dark or the light side of it :) Don't blindly trust Wikipedia.
Hmmm... Yup, vira would probably be about right :) I wonder when it actually entered the language - I'll be ethymology-googling after work today now that I'm curious :)
If you know English, you should know that virii is the plural for virus (cactus - cactii, fungus - fungi, etc.) "Virus" entered English indirectly from Latin.
You'd also have to get the /etc/make.profile symlink point to the new 2004.1 directory under /usr/portage/profiles to get full equivalence to the new "release".
Microsoft's patent on the ASF file format
Windows Media Licensing Terms, and specifically the the licensing costs for Windows Media Audio and Video 9 codecs for non-Windows desktops and hardware devices
Apple's Quicktime software and hardware licensing terms
These terms may or may not apply, depending on the local laws. But in the United States at least it's certainly not legal to use Windows DLLs that way. Now I know people will start claiming they don't care, but purposedly breaking laws isn't going to help the Linux community.
Uh, it doesn't claim to have full Unicode support. What it claims to have is UTF-8 encoding support. Check your sources.
Release announcement on SourceForge
Novell has acquired SUSE, and Novell already contributes to the OSDL fund.