There's a junk science fad on right now. I expect to be modded down severely for pointing out that the global warming idea is not supported by evidence, and it is a wonderful example of assumptions driving the data-collection.
Global warming denial is SOOO 1990's. I thought you people had moved on to quibbling over whether or not humans are causing it.
I just don't get it. What is the deal with people never changing their minds, or letting in new information? Most people aren't stupid...I'm sure the average person in Iceland isn't any smarter than the average american
Actually, I've heard that even the taxi drivers in Iceland can quote Shakespeare. I wouldn't be so sure that Americans are not less intelligent (or educated) than in other places.
Simply because evolution doesn't work that way. Just because a mutation occurs and creates a branch in the evolutionary tree, doesn't necessarily mean that the ancestor must die. A balance can be achieved among the mutated branch and the original species.
It should be noted that the ancestor the is thought to be common between humans and modern apes (chimpanzees in particular) *is* extinct. But as you say, it is concievable that the ancestor could still exist along with its decendents.
I mean, yes, I believe that all biological creatures on earth have evolved, and still evolve. But, I'm not sure I buy the starting place. I have a hard time thinking humans came from apes. I think we started as a primative human, but, not that far. But, who knows...I guess I often think of something I heard someone say: "If humans evolved from apes...why are there still apes?"
Well, I hope you included that smiley at the end of your post because you know that last quote is just silly.
Question is, what essential difference is there between what you might consider a "primitive human" and an ape? You do know that apes can use tools, right? You know that they have social structures and even (primitive) language, right? And all major physiological structures correspond to humans. So what do you think it is that serves as a evolutionary barrier to between apes such as the chimpamzee and humans?
Sadly the way this is fixed is brutally obvious: Make death fatal. But that will never happen in world of warcraft, their design precludes it.
Indeed. Such a scheme has worked for some of the most replayable games out there: rougue-like games. I played Nethack on and off for nearly a decade and recently (as in a few years ago) switched to ToME (Angband variant). Although these games don't work well when made multiplayer. And I don't think most people can handle the death of a character. Too much stress. When you spend weeks or months working on a character, it can be pretty traumatic to have it die... especially if the death was "unfair." Too much like real life.;-P
Also, Windows 3.1 was actually based on the core code on which IBM and Microsoft had collaborated. After they terminated the joint project, IBM continued development on the core code and turned it into OS/2. Meanwhile Microsoft gutted the parts (e.g., preemptive multitasking) that, in its opinion, the consumer would not value and morphed the result into Windows 3.1.
Hmm, I'm pretty sure that the technology from the combined MS/IBM effort went into Windows NT, not Windows 3.1. (unless, of course, you meant Windows NT 3.1).
It always puzzled me as to why Microsoft would think that consumers would want to use a crappy version of Windows and not NT. Was it just because NT didn't have good (or any) DOS support?
Did the AmigaOS provide much in the way of APIs and hardware abstraction? I was under the impression that most amiga programs worked directly with the hardware much of the time. That could have been a significant hurdle for general software development.
Ok so I'm not sure I get this. So say you want to support features in G3, G4 and G5 chips - you have a binary that is 3 times the size with a little forking code that identifies CPU type at the beginning? That seems a bit bloated.
Maybe it isn't worth it for G3 vs. G4 when the primary difference is Altivec vs. no Altivec. You would probalby do as you would in the PC world and just detect the presense of Altivec. But there are some advantages to be gained by compiling the whole binary for a specific CPU. GCC is full of CPU specific optimizations. When it comes to G4 vs. G5, you must have different bianries if you want to take advantage of the G5 because it is 64bit. If you want a fat binary that ALSO works on intel, then you would have a fat binary that is 3 times larger. And i imagine if you wanted support for x86-64 you'd have a fat binary that is 4 times larger. But so what? Hard drive space is cheap. And it is only the executable. There is MUCH more to modern software than just the program executable itself. That stuff isn't duplicted by using fat binaries.
But if the "bloat" really bothers you, any end user can use the 'lipo' tool to strip out the unwanted binaries. Fat binaries make the whole process of using different CPUs a lot more transparent to users. 64bit users on the PC could only dream of such a smooth transition as Apple has provided. Hell, even the move to intel has been relatively painless and that is a whole different architecture completely.
According to Wikipedia Apple does indeed use fat binaries when producing code optmized for variants of the a CPU type (G4 vs. G5). The same mechanism is used to support both ppc and x86. "Universal Binary" is just a specific use of the fat binary system.
This is not how it is done on Windows to support SSE2 vs. MMX. On Windows one simply writes different functions within a single codebase. One function uses MMX and another using SSE2, for example.
They don't need a mechanism. You could use a 32 bit stub that launches either the 32 bit or 64 bit program depending on which your system was. Apple just makes it easier.
I imagine you'd have issues with DLLs. Anyway, I wasn't saying it is IMPOSSIBLE to make a "universal" program under windows. Just that there is no mechanism for it. A lot of time just making something easy or obvious makes all the difference in the world. It is the difference between something becoming standard operating procedure (as on the Mac) and having only a few people figuring it out and doing it (as on Windows).
Also, due to Apples applicaiton structure (e.g. Universal Binaries), it is a lot easier to distribute one program that will run optimally on many different processors. Some apps have come with both G4 and G5 optimized binaries for some time. As far as I know, Microsoft has no such mechanism for distributing binaries like that.
You are using Mail.app and Spotlight (I do too) so you don't think gmail is so amazing.
Whew! I thought I was missing something about gmail too.
I actually got tired of using the slowish AJAX gmail client and just started POPing my gmail items into Mail.app. Now, if only gmail supported IMAP. One thing that gmail (and many, but not all, webmail clients) can't do is aggregate many accounts.
I'm sure that the programmers think they've very clever by choosing a name that means something in some obscure language- or they just thing the name sounds cool-
Or they are just being whimsical can don't really care if the the name is business friendly. Most are just programming for fun. I wholeheartedly support the practice of "strange" naming. If Linux, or more generally, FOSS ever loses its whimsical and 'fun' nature, it'll be dead. Programmers get enough marketting pressure and other business related distractions at work. They don't need to come home to their hobby project and get the same pressure to sell themselves to users.
but that simple lack of meaningful names is detrimental.
Detrimental to whom, exactly? If I am developing something for fum, what do I care if a few people who merely object to naming don't use my software?
If I start up a GNOME session and want to use network meeting functionality, how is there any possible way that I could guess that "Ekiga" is the application I'm looking for?
The same way you know that your friend "Joe" is the guy with the dark hair, funny nose, and a good sense of humor. Have you ever asked your friend Joe to change his name to something a little more descriptive?
Like I said, I think they are probably targetting a specific kind of filter. Perhaps in this case it would be an organizational Bayesian dictionary. Having run a Baysian system for a group (2,000 users), I can tell you that it is VERY resource intensive to maintain individual dictionaries. I'd say it is MORE resource intensive than maintaining the Email boxes themselves. It is very tempting to use group dictionaries.
I wonder what view into the various statistics that Jupiter Research employed to make this claim. Perhaps spam filters have improved, and the spam that people actually see in their inbox has fallen. Google's spam filter seems to work better than others, but I don't think Google could account for a 17% drop overall, and I don't see much evidence of major improvements in spam filtering technology overall.
No major improvement in spam filtering technology is required. All that is requires is for more people to implement the systems that are already available. There are still a lot of organizations which do not employ any significant spam filtering. As more implement filtering systems, the amount recieved by users will drop.
In my experience, even a relatively untrained system can catch like 70% of all spam. Imagine if everyone implemented such a system.
It seems to me that, and please prove me wrong, that whatever technique legitimate researchers come up with to stop spam, is quickly outsmarted by independent teams of illegal spammers. Do the spammers have an easier job, or are they just smarter?
No, it is just that spammers are, by definition, ahead of the game. I mean, we are reacting to the problem.
That said, I think we are doing pretty good against spam. At least as far as keeping it out of users' mailboxes go. Only a few years ago, mailboxes were littered with the crap. Any decent filter can stop a large majority of spam and I have not see any evidence that these new tactics are particularly effective.
I think that is the point. They want to either poison those words so you get more false positives or they want to push other REAL spam related words out of the "this is spam" dictionaries. Maybe both. If these messages had some common theme, they would all get blocked and would have no net effect. They need you to click "this is spam" to poison your filters.
Question is, does it work? I don't know. Seems to be highly dependent on the nature of your spam filter. Maybe they are only targeting a specific, popular filtering system.
To me it seems like an act of deparation. I think filters are finally catching up with spammers. It is getting more and more difficult to get spam through a half way decent filter and there are a lot of decent filters out there.
I seem to remember years ago people talking about how banner ads and pay-per-click just don't work. What happened? Is Adsense really that effective? People I talk to hardly even notice the text ads, much less click on them. No, this isn't one of those "I never click on an ad" rants. I'm really curious here. What has really changed besides a little targetting?
My gut feeling is that Google is scamming the world. They took a model that was broken, applied some superficial "fixes" to it and got everyone to believe that banner/text ads are "in" again. Meanwhile, they hide all of their logs in the name of privacy so nobody can really tell who is clicking on what. I would not be at all surprised if 'net advertising has become like email is today... 80% fraud and junk. I trust the consultants over the companies (Google) who have an interest in protecting they're primary source of income. But that is just my gut feeling. The facts could be completely different.:-)
Same here, but what I DON'T prefer is segmented memory and trying to figure out when to use the small library, the medium, or the large (or whatever it was called). Not that it is Borland's fault. DOS programming sucks ass.
It might be interesting if Apple licensed someone's virtualization tech and used it to create a sort of downloadable "demo" version of OS X that Windows users could play around with, though. Can virtualized operating systems take advantage of GPU acceleration? Seems like that would be necessary for such an application, as OS X is somewhat less impressive for demo purposes without its GPU-accellerated eye candy.
A friend of mine got OS X running inside VMWare on a PC. He said it was not very usable. Accessing the GPU from virtualization requires special OS video drivers that are virtualization aware. I wonder if VMWare supplies such drivers.
Slashdot Mirror
Global warming denial is SOOO 1990's. I thought you people had moved on to quibbling over whether or not humans are causing it.
-matthew
Actually, I've heard that even the taxi drivers in Iceland can quote Shakespeare. I wouldn't be so sure that Americans are not less intelligent (or educated) than in other places.
-matthew
It should be noted that the ancestor the is thought to be common between humans and modern apes (chimpanzees in particular) *is* extinct. But as you say, it is concievable that the ancestor could still exist along with its decendents.
-matthew
Well, I hope you included that smiley at the end of your post because you know that last quote is just silly.
Question is, what essential difference is there between what you might consider a "primitive human" and an ape? You do know that apes can use tools, right? You know that they have social structures and even (primitive) language, right? And all major physiological structures correspond to humans. So what do you think it is that serves as a evolutionary barrier to between apes such as the chimpamzee and humans?
-matthew
I hardly think that writing managed code that runs on both Windows and XBOX will give one a chance to explore "how the system really works."
Yeah, I'm sure gamers will be climbing over each other to run Joe Bob's Tetris.NET on their shiny new XBox-360.
-matthew
Indeed. Such a scheme has worked for some of the most replayable games out there: rougue-like games. I played Nethack on and off for nearly a decade and recently (as in a few years ago) switched to ToME (Angband variant). Although these games don't work well when made multiplayer. And I don't think most people can handle the death of a character. Too much stress. When you spend weeks or months working on a character, it can be pretty traumatic to have it die... especially if the death was "unfair." Too much like real life.
-matthew
Hardly a GUI. All it did was multitask DOS programs. ALthough there was something called DESQview/X. Never used it, personally.
Hmm, I'm pretty sure that the technology from the combined MS/IBM effort went into Windows NT, not Windows 3.1. (unless, of course, you meant Windows NT 3.1).
It always puzzled me as to why Microsoft would think that consumers would want to use a crappy version of Windows and not NT. Was it just because NT didn't have good (or any) DOS support?
-matthew
Did the AmigaOS provide much in the way of APIs and hardware abstraction? I was under the impression that most amiga programs worked directly with the hardware much of the time. That could have been a significant hurdle for general software development.
-matthew
Maybe it isn't worth it for G3 vs. G4 when the primary difference is Altivec vs. no Altivec. You would probalby do as you would in the PC world and just detect the presense of Altivec. But there are some advantages to be gained by compiling the whole binary for a specific CPU. GCC is full of CPU specific optimizations. When it comes to G4 vs. G5, you must have different bianries if you want to take advantage of the G5 because it is 64bit. If you want a fat binary that ALSO works on intel, then you would have a fat binary that is 3 times larger. And i imagine if you wanted support for x86-64 you'd have a fat binary that is 4 times larger. But so what? Hard drive space is cheap. And it is only the executable. There is MUCH more to modern software than just the program executable itself. That stuff isn't duplicted by using fat binaries.
But if the "bloat" really bothers you, any end user can use the 'lipo' tool to strip out the unwanted binaries. Fat binaries make the whole process of using different CPUs a lot more transparent to users. 64bit users on the PC could only dream of such a smooth transition as Apple has provided. Hell, even the move to intel has been relatively painless and that is a whole different architecture completely.
-matthew
No! Please let Symantec die! They screw up every half-way decent piece of software they get their grubby little paws on.
According to Wikipedia Apple does indeed use fat binaries when producing code optmized for variants of the a CPU type (G4 vs. G5). The same mechanism is used to support both ppc and x86. "Universal Binary" is just a specific use of the fat binary system.
This is not how it is done on Windows to support SSE2 vs. MMX. On Windows one simply writes different functions within a single codebase. One function uses MMX and another using SSE2, for example.
-matthew
I imagine you'd have issues with DLLs. Anyway, I wasn't saying it is IMPOSSIBLE to make a "universal" program under windows. Just that there is no mechanism for it. A lot of time just making something easy or obvious makes all the difference in the world. It is the difference between something becoming standard operating procedure (as on the Mac) and having only a few people figuring it out and doing it (as on Windows).
Also, due to Apples applicaiton structure (e.g. Universal Binaries), it is a lot easier to distribute one program that will run optimally on many different processors. Some apps have come with both G4 and G5 optimized binaries for some time. As far as I know, Microsoft has no such mechanism for distributing binaries like that.
-matthew
Whew! I thought I was missing something about gmail too.
I actually got tired of using the slowish AJAX gmail client and just started POPing my gmail items into Mail.app. Now, if only gmail supported IMAP. One thing that gmail (and many, but not all, webmail clients) can't do is aggregate many accounts.
-matthew
Or they are just being whimsical can don't really care if the the name is business friendly. Most are just programming for fun. I wholeheartedly support the practice of "strange" naming. If Linux, or more generally, FOSS ever loses its whimsical and 'fun' nature, it'll be dead. Programmers get enough marketting pressure and other business related distractions at work. They don't need to come home to their hobby project and get the same pressure to sell themselves to users.
Detrimental to whom, exactly? If I am developing something for fum, what do I care if a few people who merely object to naming don't use my software?
The same way you know that your friend "Joe" is the guy with the dark hair, funny nose, and a good sense of humor. Have you ever asked your friend Joe to change his name to something a little more descriptive?
-matthew
What about people who either don't get a lot of mail or get far more spam than ham? They can have a difficult time training an effective dictionary.
-matthew
Like I said, I think they are probably targetting a specific kind of filter. Perhaps in this case it would be an organizational Bayesian dictionary. Having run a Baysian system for a group (2,000 users), I can tell you that it is VERY resource intensive to maintain individual dictionaries. I'd say it is MORE resource intensive than maintaining the Email boxes themselves. It is very tempting to use group dictionaries.
-matthew
No major improvement in spam filtering technology is required. All that is requires is for more people to implement the systems that are already available. There are still a lot of organizations which do not employ any significant spam filtering. As more implement filtering systems, the amount recieved by users will drop.
In my experience, even a relatively untrained system can catch like 70% of all spam. Imagine if everyone implemented such a system.
-matthew
No, it is just that spammers are, by definition, ahead of the game. I mean, we are reacting to the problem.
That said, I think we are doing pretty good against spam. At least as far as keeping it out of users' mailboxes go. Only a few years ago, mailboxes were littered with the crap. Any decent filter can stop a large majority of spam and I have not see any evidence that these new tactics are particularly effective.
-matthew
I think that is the point. They want to either poison those words so you get more false positives or they want to push other REAL spam related words out of the "this is spam" dictionaries. Maybe both. If these messages had some common theme, they would all get blocked and would have no net effect. They need you to click "this is spam" to poison your filters.
Question is, does it work? I don't know. Seems to be highly dependent on the nature of your spam filter. Maybe they are only targeting a specific, popular filtering system.
To me it seems like an act of deparation. I think filters are finally catching up with spammers. It is getting more and more difficult to get spam through a half way decent filter and there are a lot of decent filters out there.
-matthew
I seem to remember years ago people talking about how banner ads and pay-per-click just don't work. What happened? Is Adsense really that effective? People I talk to hardly even notice the text ads, much less click on them. No, this isn't one of those "I never click on an ad" rants. I'm really curious here. What has really changed besides a little targetting?
:-)
My gut feeling is that Google is scamming the world. They took a model that was broken, applied some superficial "fixes" to it and got everyone to believe that banner/text ads are "in" again. Meanwhile, they hide all of their logs in the name of privacy so nobody can really tell who is clicking on what. I would not be at all surprised if 'net advertising has become like email is today... 80% fraud and junk. I trust the consultants over the companies (Google) who have an interest in protecting they're primary source of income. But that is just my gut feeling. The facts could be completely different.
Same here, but what I DON'T prefer is segmented memory and trying to figure out when to use the small library, the medium, or the large (or whatever it was called). Not that it is Borland's fault. DOS programming sucks ass.
A friend of mine got OS X running inside VMWare on a PC. He said it was not very usable. Accessing the GPU from virtualization requires special OS video drivers that are virtualization aware. I wonder if VMWare supplies such drivers.
-matthew
They don't call it an End User License Agreement for nothing.