Even taking global warming into account, carbon dioxide is simply changing the rate at which energy is radiated at a given temperature level. This raises the equilibrium temperature of the Earth body, but at the new equilibrium, the energy gained and lost would still be equal.
IIRC, the US copyright law explicitly allow some kind of sampling as fair use. I don't remember the criteria, and there is a thin borderline between fair use and infringement. But impression has it that copyright law is more permissive than the proposed Noncommercial Sampling Plus license.
Someone should conduct a survey on why people keep coming back to Slashdot for mod points, doing meta moderating to get mod points, and post comments that earn mod points. This is beginning to look like a reward system where a success is reflected in the number of mod points you get from your comments.
So I was in a hurry to get this program to work, so I didn't investigate very far until I gave up and use the Solaris cc instead. But here are some details. It was gcc 3.3 with Solaris 2.6. The program was using ucontext fairly throughly (userland context switching). The same program worked under Solaris cc.
As to your comment on segfault, if a program is correct, where can it get its bad pointer from? Misaligned struct members, of course. And this is not an ABI problem? You must be kidding me. And when you see the desired value is at an offset +4 than the location your program is peeking at (observed in gdb, of course), you know there is a problem.
Allow me to refresh your terminology. ABI, Application Binary Interface, does not restrict itself to system calls. It refers to the binary encoding of data structures so both programs using the same data structure definition can understand each other at a binary level.
To think back a bit more, it is probably some compiler #define differences that altered the definition in the ucontext.h header, and not necessarily due to ABI incompatibility either (since they wouldn't be using the same definition). I wouldn't be surprised if this is the case. But fixing that involves hacking the gcc spec file, which is not something I wanted to do in a hurry.
Anyhow, I respect your choice of using gcc and defending it very passionately. However, my choice is ambivalent. I just use the right tool for the right job.
I know the Sun and SGI geeks have complained for years that gcc does not run well on their platforms compared to expensive alternatives by their vendors.
I tried gcc 3.3.2 on Solaris. Most problems I've had so far have to deal with are header/library path problems. The reason, according to $(prefix)/lib/gcc-lib/$(platform)/$(gcc-version)/
include/README, is because "GCC requires ANSI C headers and many vendors supply ANSI-incompatible
headers."
Also, some programs compiled with gcc tends to segfault. I haven't found the exact reason, but those programs tend to be linking with more exotic libraries compiled using the vendor cc, so I think it's an ABI problem.
The NewsForge article is so bipartite structured such that it gives Connolly's view, then Emir Sakic's view. There is no formal introduction nor conclusion. If you start reading and bail out, you'll think the article is entirely pro-Connolly. If you read towards all the way to end, you'll get a much better perspective.
To pull our heads away from the heat, if this story was never/.'ed, Connolly and his Furthermore would probably never be heard (NewsForge said he was "far from being the pioneer in this industry"). Now he's given much more publicity than he could never earn if he had not hyped up the issue.
Actually, according to the developer in the question of breach of contract, there was no contract or copyright agreement signed, and the code contributed to Connolly was itself based on another GPL code (a merely 9 lines of code). So it seems that Connolly paid someone over the course of 8 months just to get 9 lines of code already written by someone else.
One of the key feature of ZFS is that you can create a file system over a pool of storage. Nothing stops you from building a distributed storage pool of 18.3 million desktop drives (they don't have to be locally connected). You could apply the same concept as SETI@HOME and allow end users with excessive storage space to lend them. Didn't someone talk about a peer to peer backup system a while ago?
And com'n, don't be so against hypes. Not all numbers are evil. And the overhead to process some extra bits are miniscule. The space and time required are in logarithmic time to the size of the number set. E.g., 128-bit is some billions billion times the size of 64-bit, but only takes 2 times more to store and process. And this time is already small compared to the actual I/O time, and the space compared to combined storage space.
It sucks when ...
on
Is IP Property?
·
· Score: 0, Redundant
As
featured on slashdot before, IP can be abused by a company to own an employee's thought that wasn't even well formed. When you work for a company, do they own your thoughts? Are your thoughts your soul? Does a company own your soul? And does money buy souls?
First of all, APR is not a virtual machine or bytecode interpreter like that of.NET common language runtime or JVM. APR is a library (collection of functions) written in C, for C programs. It contains a lot of wrappers to the real standard C library functions, because some conventions of standard library still varies from OS to OS.
For example, the path separator is different in Unix ("/"), Windows ("\"), MacOS (":" - pre X, but also Finder in OS X). Another example is loading dynamically linked libraries (DSO in APR speech). Yet another example is threads.
Besides wrappers, APR has its own memory management routines. APR also adds utility functions not found in the standard library, such as hash table.
By the way, it would be helpful if someone can post a comparison between NSPR (netscape portable runtime) and APR.
I hope you can put up some robots.txt file that allows legitimate search engines like google to escape. I'm pretty sure RIAA or MPAA won't observe robots.txt.
A more reasonable change would be SMTP-TLS, employing a policy of using authorized certificates like the secure websites. This protocol is already there, but it's the wide adoption that is the problem.
... is not to block spam, but to identify the source of an e-mail. Spammers can definitely identify themselves if they so choose. I think it is still a welcoming trend.
It simply means that USA is not scaling to the economics. In other words, the economy of scale is sublinear. It is the same as how a cluster of two computers with a 1Ghz processors does not perform as good as a computer with a 2Ghz processor.
Since your intention is to write unbiased material, I'd like to refer you to WikiPedia's Neutral Point of View article. Notice that NPOV doesn't mean you can't say your opinion, but you must say so ("my opinion is that...") and possibly present other people's opinion. In your writing, there is a mixed personal opinion and small number of facts. It might have been better to highlight both of them separately.
To say that CMYK cancels RGB is a misunderstanding of the context. RGB is a color model. CMY(K) is another color model. RGBCMY is yet another color model. The "additiveness" is a way to describe how the coordinates of a color model contribute to the actual frequency spectrum and intensity of light. Accordingly, RGBCMY (or more appropriately, RYGCBM, sorted by spectrum frequency) is an additive model. The advantage is that finer spectrum prescence of the primary colors makes it easier to reproduce the visible spectrum of light by adding the primary colors on top of each other.
Your remark on LCD using CMYk approach is also misleading. It is true that LCD masks out light coming from a white light source, but the resulting light, being in RGB primaries, still "adds up" to reproduce color. If you don't believe me, you can do a little experiment. Just spill some droplets of water on the screen, and watch the enlarged subpixels.
Although theoretically, since human only has three cone cell (color) receptors in the eye, three primary colors should suffice stimulating them. But different person has slightly different response curves. To an extreme, we have Color blind individuals who only has two types of cone cells. Some others are anomalous trichromats, who can see the three primary colors like normal people, but the mixture of the colors are different.
On the other hand, I don't think RGBCMY (or should that be RYGCBM if sorted in increasing frequency order?) is innovative either. It's like going from stereo to surround sound--you just add more speakers to facilitate better sound positioning (but we only have two ears, darling). In the case of RYGCBM, you get better light spectrum positioning.
I don't believe the claim that the software they use can, in general, "see and hear." Software agents can "see" and "hear" only in a very specific sense. For example, using Motion, you can capture only the frames that are "interesting," i.e., with some things moving. Some existing vision technology allow the recognition of large areas of exposure (visible light or infra red), like that caused by an explosion. It may be able to count cars and see if they're moving. I know of a project at my school to recognize faces in partial images (taken by uncalibrated cameras) in a conference room and see how many distinguished individuals are actually present.
As for sound, I accept that there are speech recognition software for many languages available nowadays, but none of them are good enough because: (1) You have to speak in a certain way for good recognition (no biting tongue, clear prenounciation, clear word break, standard accent), and (2) The ambient (on the streets) where the samples are taken is too noisy.
I think the way CNN runs this article is misleading of the current state of technology. It might be the case the their reporter doesn't understand the press release well, or that the reporter just wants to write something cool. Maybe they want to scare the public too. Shame on them.
This is a common phenomom of how people don't care how things are done, just the fact that it is done. Maybe one way (utilizing local resources for computing) works better than another (ssh, X session, VNC), but aside from the performance (with some numbers to measure), how the thing works is out of hand for most people to grasp.
I suspect there are still wonders of the problem, some of them you already mentioned, namely how to abstract the hardware so changing the physical machines don't incur changes to the state of the programs you're running.
Alternatively, in a different line of thinking, one can come up with a programming language (possibly some popular language dialect augmented with a "distributed" semantic) with a runtime system that abstracts location. It could end up being implemented as some machine relocatable bytecode (JIT compiled) with RPC interfaces... again, using existing technology. This way, maybe data accessing code continues to run on the server backend, while interactive code runs on the local end to give a snappy response. Now, this starts to sound like a beefed up Web Browser, doesn't it?
Slashdot Mirror
Even taking global warming into account, carbon dioxide is simply changing the rate at which energy is radiated at a given temperature level. This raises the equilibrium temperature of the Earth body, but at the new equilibrium, the energy gained and lost would still be equal.
IIRC, the US copyright law explicitly allow some kind of sampling as fair use. I don't remember the criteria, and there is a thin borderline between fair use and infringement. But impression has it that copyright law is more permissive than the proposed Noncommercial Sampling Plus license.
Someone should conduct a survey on why people keep coming back to Slashdot for mod points, doing meta moderating to get mod points, and post comments that earn mod points. This is beginning to look like a reward system where a success is reflected in the number of mod points you get from your comments.
From the author's page:
So I was in a hurry to get this program to work, so I didn't investigate very far until I gave up and use the Solaris cc instead. But here are some details. It was gcc 3.3 with Solaris 2.6. The program was using ucontext fairly throughly (userland context switching). The same program worked under Solaris cc.
As to your comment on segfault, if a program is correct, where can it get its bad pointer from? Misaligned struct members, of course. And this is not an ABI problem? You must be kidding me. And when you see the desired value is at an offset +4 than the location your program is peeking at (observed in gdb, of course), you know there is a problem.
Allow me to refresh your terminology. ABI, Application Binary Interface, does not restrict itself to system calls. It refers to the binary encoding of data structures so both programs using the same data structure definition can understand each other at a binary level.
To think back a bit more, it is probably some compiler #define differences that altered the definition in the ucontext.h header, and not necessarily due to ABI incompatibility either (since they wouldn't be using the same definition). I wouldn't be surprised if this is the case. But fixing that involves hacking the gcc spec file, which is not something I wanted to do in a hurry.
Anyhow, I respect your choice of using gcc and defending it very passionately. However, my choice is ambivalent. I just use the right tool for the right job.
I tried gcc 3.3.2 on Solaris. Most problems I've had so far have to deal with are header/library path problems. The reason, according to $(prefix)/lib/gcc-lib/$(platform)/$(gcc-version)/ include/README, is because "GCC requires ANSI C headers and many vendors supply ANSI-incompatible headers."
Also, some programs compiled with gcc tends to segfault. I haven't found the exact reason, but those programs tend to be linking with more exotic libraries compiled using the vendor cc, so I think it's an ABI problem.
... if I had mod points right now.
The NewsForge article is so bipartite structured such that it gives Connolly's view, then Emir Sakic's view. There is no formal introduction nor conclusion. If you start reading and bail out, you'll think the article is entirely pro-Connolly. If you read towards all the way to end, you'll get a much better perspective.
To pull our heads away from the heat, if this story was never /.'ed, Connolly and his Furthermore would probably never be heard (NewsForge said he was "far from being the pioneer in this industry"). Now he's given much more publicity than he could never earn if he had not hyped up the issue.
Actually, according to the developer in the question of breach of contract, there was no contract or copyright agreement signed, and the code contributed to Connolly was itself based on another GPL code (a merely 9 lines of code). So it seems that Connolly paid someone over the course of 8 months just to get 9 lines of code already written by someone else.
One of the key feature of ZFS is that you can create a file system over a pool of storage. Nothing stops you from building a distributed storage pool of 18.3 million desktop drives (they don't have to be locally connected). You could apply the same concept as SETI@HOME and allow end users with excessive storage space to lend them. Didn't someone talk about a peer to peer backup system a while ago?
And com'n, don't be so against hypes. Not all numbers are evil. And the overhead to process some extra bits are miniscule. The space and time required are in logarithmic time to the size of the number set. E.g., 128-bit is some billions billion times the size of 64-bit, but only takes 2 times more to store and process. And this time is already small compared to the actual I/O time, and the space compared to combined storage space.
As featured on slashdot before, IP can be abused by a company to own an employee's thought that wasn't even well formed. When you work for a company, do they own your thoughts? Are your thoughts your soul? Does a company own your soul? And does money buy souls?
First of all, APR is not a virtual machine or bytecode interpreter like that of .NET common language runtime or JVM. APR is a library (collection of functions) written in C, for C programs. It contains a lot of wrappers to the real standard C library functions, because some conventions of standard library still varies from OS to OS.
For example, the path separator is different in Unix ("/"), Windows ("\"), MacOS (":" - pre X, but also Finder in OS X). Another example is loading dynamically linked libraries (DSO in APR speech). Yet another example is threads.
Besides wrappers, APR has its own memory management routines. APR also adds utility functions not found in the standard library, such as hash table.
By the way, it would be helpful if someone can post a comparison between NSPR (netscape portable runtime) and APR.
I hope you can put up some robots.txt file that allows legitimate search engines like google to escape. I'm pretty sure RIAA or MPAA won't observe robots.txt.
A more reasonable change would be SMTP-TLS, employing a policy of using authorized certificates like the secure websites. This protocol is already there, but it's the wide adoption that is the problem.
It simply means that USA is not scaling to the economics. In other words, the economy of scale is sublinear. It is the same as how a cluster of two computers with a 1Ghz processors does not perform as good as a computer with a 2Ghz processor.
Dear Anonymous Coward,
Since your intention is to write unbiased material, I'd like to refer you to WikiPedia's Neutral Point of View article. Notice that NPOV doesn't mean you can't say your opinion, but you must say so ("my opinion is that ...") and possibly present other people's opinion. In your writing, there is a mixed personal opinion and small number of facts. It might have been better to highlight both of them separately.
If you bomb the place,
you too destroy their broadband.
Nothing left to steal.
--- Pikine's Haiku
us1 mirror and see Google cache for more.
No, I think i'll prefer to build one with Lego.
... or does the girl really look like the bride of chucky?
To say that CMYK cancels RGB is a misunderstanding of the context. RGB is a color model. CMY(K) is another color model. RGBCMY is yet another color model. The "additiveness" is a way to describe how the coordinates of a color model contribute to the actual frequency spectrum and intensity of light. Accordingly, RGBCMY (or more appropriately, RYGCBM, sorted by spectrum frequency) is an additive model. The advantage is that finer spectrum prescence of the primary colors makes it easier to reproduce the visible spectrum of light by adding the primary colors on top of each other.
Your remark on LCD using CMYk approach is also misleading. It is true that LCD masks out light coming from a white light source, but the resulting light, being in RGB primaries, still "adds up" to reproduce color. If you don't believe me, you can do a little experiment. Just spill some droplets of water on the screen, and watch the enlarged subpixels.
see also: Cone Cell on wikipedia.org
Although theoretically, since human only has three cone cell (color) receptors in the eye, three primary colors should suffice stimulating them. But different person has slightly different response curves. To an extreme, we have Color blind individuals who only has two types of cone cells. Some others are anomalous trichromats, who can see the three primary colors like normal people, but the mixture of the colors are different.
On the other hand, I don't think RGBCMY (or should that be RYGCBM if sorted in increasing frequency order?) is innovative either. It's like going from stereo to surround sound--you just add more speakers to facilitate better sound positioning (but we only have two ears, darling). In the case of RYGCBM, you get better light spectrum positioning.
I don't believe the claim that the software they use can, in general, "see and hear." Software agents can "see" and "hear" only in a very specific sense. For example, using Motion, you can capture only the frames that are "interesting," i.e., with some things moving. Some existing vision technology allow the recognition of large areas of exposure (visible light or infra red), like that caused by an explosion. It may be able to count cars and see if they're moving. I know of a project at my school to recognize faces in partial images (taken by uncalibrated cameras) in a conference room and see how many distinguished individuals are actually present.
As for sound, I accept that there are speech recognition software for many languages available nowadays, but none of them are good enough because: (1) You have to speak in a certain way for good recognition (no biting tongue, clear prenounciation, clear word break, standard accent), and (2) The ambient (on the streets) where the samples are taken is too noisy.
I think the way CNN runs this article is misleading of the current state of technology. It might be the case the their reporter doesn't understand the press release well, or that the reporter just wants to write something cool. Maybe they want to scare the public too. Shame on them.
I wonder when Apple is going to run out of feline names. Maybe Mac OS X 10.6 will be called, in its full honestly, "Pussy."
I'm a dedicated Mac OS X user. Just check my user-agent: Safari.
This is a common phenomom of how people don't care how things are done, just the fact that it is done. Maybe one way (utilizing local resources for computing) works better than another (ssh, X session, VNC), but aside from the performance (with some numbers to measure), how the thing works is out of hand for most people to grasp.
I suspect there are still wonders of the problem, some of them you already mentioned, namely how to abstract the hardware so changing the physical machines don't incur changes to the state of the programs you're running.
Alternatively, in a different line of thinking, one can come up with a programming language (possibly some popular language dialect augmented with a "distributed" semantic) with a runtime system that abstracts location. It could end up being implemented as some machine relocatable bytecode (JIT compiled) with RPC interfaces ... again, using existing technology. This way, maybe data accessing code continues to run on the server backend, while interactive code runs on the local end to give a snappy response. Now, this starts to sound like a beefed up Web Browser, doesn't it?