Plus, did you know that it takes much harder work to become a good musician than to become a good software engineer?
Maybe to become a good enough musician or software engineer, but I'd strongly disagree with your point as written - it's just less obvious to people outside of the field when a software engineer isn't very good.
But, although they have included hypothesis testing as a component, they have not included testability as a *requirement*. This is a fundamental difference between their definition of science and the generally accepted version of same.
The fact remains--you can't tell kids its true when there is no proof that it is true.
One thing that people seem to be missing in this debate is a fundamental misunderstanding of the nature of "scientific truth". There's no such thing.
Science works by proposing a hypothesis and testing it for falsehood. *Not* by testing it for truth, because that is impossible outside of a tautology.
Therefore the "output" of science is not "truth" but "the model that matches objective reality - and can be tested against objective reality - better than any other we have at present". Calling the results of this process "scientific facts" shows a lack of understanding of the actual epistemological structure of science, and calling them "theories" is technically correct but mistakes (or deliberately misrepresents) the input (hypothesis) for the output (hypothesis that has survived a period of testing).
The real people who don't want the truth to be known are those who refuse to let the children know that evolution remains a theory and not a proven fact.
The problem is, there's no such thing as an *absolutely* proven fact about objective reality - Is your monitor displaying these words or are you hallucinating them? It's a question of "degree of confidence" and we have philosophical tools for determining how confident we can be about scientific "truths" but not religious ones. That is not to say that religion cannot be true, but the degree of truth of its assertions cannot be tested in the same way that the assertions of science can be.
The other point in favour of science is that is does, when practiced correctly, admit fallability, incompleteness, and the provisional nature of its truth. Unfortunately, with the way the evolution-vs-creation debate is being carried out in the US, the pro-evolution camp is omitting this from their presentation. They may be doing this to simplify their argument to suit their audience, or they may have compromised their scientific integrity by actually believing that provisional truth is absolute truth.
That being said, evolution is as much a theory as creationism,
Here's the point where I must start to respectfully disagree with you. It's not a binary distinction. Evolution explains most of the evidence we have so far, and creates few additional unanswered questions. Creationism explains most of the evidence we have so far, but creates a large number of unanswered questions (relating to the nature of the Creator). Therefore they are not the same order of theory.
That's not a statement about truth, but a statement about utility and completeness.
My opinion is that children should be taught a lot more critical thinking skills and about the nature of knowledge before people start waving words like "theory" and "fact" in their faces.
I took a 50/50 degree: basically we did the BEng EE course and the BEng CS course (4 years to do both - a Bachelor's in Britain is normally 3 years) and came out with an undergrad MEng. After living through the crash out here in CA I am very glad that I did. There is a big shortage of people who can think across the hardware/software boundary, can see things from a "systems" perspective, and can roll up their sleeves and do whatever's required when the work needs to be done: be it writing C, designing gate-level logic, or working out the architecture at a high level and partitioning it to meet requirements.
It's worth putting in the extra work. Learn both. You'll be more valuable in the job market and have more scope for finding fun, well-paid things to do. I've worked in areas from low-level logic design up to AI software and wouldn't have had the chance if I'd limited myself to only half of the world.
When I went to the US the flicker from flourescent tubes drove me insane (in the UK they flicker at 50Hz, what is it in the states?).
60Hz in the US, so for single tube installations you should see less flicker. However, in the UK, the Health and Safety regulations for offices require that multi-tube installations have the tubes fed from different phases of the supply. So a typical office setup with three tubes, one on each phase, gives you almost no noticable flicker.
Once they receive this letter from you, they are no longer allowed to contact you, and the alleged "debt" will not show up on your credit report, either.
The last time a collection agency sent me a notice (wrongly, btw) my reply was along the lines of:
"Your client is in error and has provided you with incorrect information. If you would like my assistance in resolving the issue between yourselves and your client, my consulting rates are $250/hr with a minimum of 3 days per request answered.
The last GPU driver I worked on had three complete compilers embedded in it, each for a different piece of hardware. One for the pixel shader, one for the vertex shader, and one to construct optimised texture transfer loops for all possible combinations of motherboard chipset, bus, memory timing, source pixel format, destination pixel format and pixel layout in memory. The transfer compiler alone was about 200k lines of source (including all the hand-assembled MMX code fragments it used to build the transfer loops out of).
A: Security updates will migrate into the testing distribution via unstable. They are usually uploaded with their priority set to high, which will reduce the quarantine time to two days. After this period, the packages will migrate into testing automatically, given that they are built for all architectures and their dependencies are fulfilled in testing.
Two days isn't exactly bad going. About 3 pico-Microsofts, I'd say;-)
I suspect blacklists are about the only useful weapon here, but even they are only so successful.
I've started blacklisting at the firewall instead of the mail server. Partly because it's a lot less load on the system, and partly because I'd rather those machines weren't on "my" internet in any way.
I wonder if some kind of collaborative filtering at this level would help reduce the problem. If people start seeing that their zombied machine can't access anywhere on the intarweb they might start to try and figure out why no-one wants to talk to them anymore. Blocking at the mail server is far less visible because they never knew that the mails were being sent in the first place.
Yes, I know that this has problems with dynamic IPs, but I don't have a better solution right now. Hopefully the complaints propagated up to the ISPs will be propagated down to the users of compromised machines with the assistance of a big cluestick and/or account cancellation.
Well, if it's Turing complete, then we're not talking about regular expressions any more, by definition;-)
More seriously, how would you write a palindrome detector?
You could, in theory, use heavy ECC to tolerate a substantial defect rate.
There are DRAM technologies in production right now that do ECC on die to improve effective yield. The 90-98% yield you quote is probably effective yield, i.e. yield after error correction (either via ECC or on-die patching) is included.
Even a Cadbury Dark bar is better chocolate than Milka.
IIRC, most of the "Cadbury" chocolate available in the US is actually made by Hershey under license. If you want real Cadbury's you have to do some hunting around in delis. And, yes, it does taste completely different to the rebadged Hershey crap.
What makes you think MS is _required_ to give you any updates whatsoever for any of their products?
If I paid for something and it's broken, then I expect the manufacturer to repair it free of charge. If, however, we're talking about additional features then I would agree with you.
Right, because the Germans invented DES which started the rush of crypto algorithms while IBM (an American company) was still using polyalphabetic substitution cipers. No, wait, it's the other way around, stupid ass. You're not only wrong, but stupid.
One thing Capitalism does very well is foster innovation, both in invention and improvement of other inventions. We didn't invent the rocket, but we made it better. We invented the atomic bomb. We made serious cryptography.
Slashdot Mirror
Plus, did you know that it takes much harder work to become a good musician than to become a good software engineer?
Maybe to become a good enough musician or software engineer, but I'd strongly disagree with your point as written - it's just less obvious to people outside of the field when a software engineer isn't very good.
I have a policy of always buying someone who finds a bug in my code a beer and saying "thankyou".
Only if you are capable of remembering the basic rules of sentence capitalisation yourself.
No, "they", "their", and so on are plural. If you want to be gender-neutral, you should use "he" and "his."
The OED disagrees with you and cites historical precedent. However, it does note that your opinion exists.
'79 MGB, 300 stroked Range Rover V8, posi, 2000lbs road weight :)
(replying to self, sorry)
But, although they have included hypothesis testing as a component, they have not included testability as a *requirement*. This is a fundamental difference between their definition of science and the generally accepted version of same.
My bad, I was working with the material you had quoted rather than the original article. Sorry.
It's not just the acceptance of inductive reasoning but the removal of the requirement that a hypothesis be testable.
The fact remains--you can't tell kids its true when there is no proof that it is true.
One thing that people seem to be missing in this debate is a fundamental misunderstanding of the nature of "scientific truth". There's no such thing.
Science works by proposing a hypothesis and testing it for falsehood. *Not* by testing it for truth, because that is impossible outside of a tautology.
Therefore the "output" of science is not "truth" but "the model that matches objective reality - and can be tested against objective reality - better than any other we have at present". Calling the results of this process "scientific facts" shows a lack of understanding of the actual epistemological structure of science, and calling them "theories" is technically correct but mistakes (or deliberately misrepresents) the input (hypothesis) for the output (hypothesis that has survived a period of testing).
The real people who don't want the truth to be known are those who refuse to let the children know that evolution remains a theory and not a proven fact.
The problem is, there's no such thing as an *absolutely* proven fact about objective reality - Is your monitor displaying these words or are you hallucinating them? It's a question of "degree of confidence" and we have philosophical tools for determining how confident we can be about scientific "truths" but not religious ones. That is not to say that religion cannot be true, but the degree of truth of its assertions cannot be tested in the same way that the assertions of science can be.
The other point in favour of science is that is does, when practiced correctly, admit fallability, incompleteness, and the provisional nature of its truth. Unfortunately, with the way the evolution-vs-creation debate is being carried out in the US, the pro-evolution camp is omitting this from their presentation. They may be doing this to simplify their argument to suit their audience, or they may have compromised their scientific integrity by actually believing that provisional truth is absolute truth.
That being said, evolution is as much a theory as creationism,
Here's the point where I must start to respectfully disagree with you. It's not a binary distinction. Evolution explains most of the evidence we have so far, and creates few additional unanswered questions. Creationism explains most of the evidence we have so far, but creates a large number of unanswered questions (relating to the nature of the Creator). Therefore they are not the same order of theory.
That's not a statement about truth, but a statement about utility and completeness.
My opinion is that children should be taught a lot more critical thinking skills and about the nature of knowledge before people start waving words like "theory" and "fact" in their faces.
And yes, I am a practicing agnostic.
That too is legal. So tell me, where exactly are Microsoft corrupt?
"Corrupt" and "illegal" are not synonyms.
I took a 50/50 degree: basically we did the BEng EE course and the BEng CS course (4 years to do both - a Bachelor's in Britain is normally 3 years) and came out with an undergrad MEng. After living through the crash out here in CA I am very glad that I did. There is a big shortage of people who can think across the hardware/software boundary, can see things from a "systems" perspective, and can roll up their sleeves and do whatever's required when the work needs to be done: be it writing C, designing gate-level logic, or working out the architecture at a high level and partitioning it to meet requirements.
It's worth putting in the extra work. Learn both. You'll be more valuable in the job market and have more scope for finding fun, well-paid things to do. I've worked in areas from low-level logic design up to AI software and wouldn't have had the chance if I'd limited myself to only half of the world.
That's only true if the complexity you add is serial.
Sort of. Fanout comes into play very quickly when you start parallelising stuff. Try making a 2-level implementation of a 32 bit adder, for example.
Films are practically never as good as the books they follow (one or two exceptions like 2001
In that particular case, the book followed the film.
When I went to the US the flicker from flourescent tubes drove me insane (in the UK they flicker at 50Hz, what is it in the states?).
60Hz in the US, so for single tube installations you should see less flicker. However, in the UK, the Health and Safety regulations for offices require that multi-tube installations have the tubes fed from different phases of the supply. So a typical office setup with three tubes, one on each phase, gives you almost no noticable flicker.
Once they receive this letter from you, they are no longer allowed to contact you, and the alleged "debt" will not show up on your credit report, either.
The last time a collection agency sent me a notice (wrongly, btw) my reply was along the lines of:
"Your client is in error and has provided you with incorrect information. If you would like my assistance in resolving the issue between yourselves and your client, my consulting rates are $250/hr with a minimum of 3 days per request answered.
Not surprisingly, I never heard from them again.
Not really. Device drivers are not that complex.
The last GPU driver I worked on had three complete compilers embedded in it, each for a different piece of hardware. One for the pixel shader, one for the vertex shader, and one to construct optimised texture transfer loops for all possible combinations of motherboard chipset, bus, memory timing, source pixel format, destination pixel format and pixel layout in memory. The transfer compiler alone was about 200k lines of source (including all the hand-assembled MMX code fragments it used to build the transfer loops out of).
If it was just about what the market wants, there wouldn't be DRM.
Markets require sellers as well as buyers.
You can't copy an encrypted DVD because DVD-R media (of all forms) has the block where the encryption keys go zeroed out.
Let me just finish swapping out the laser on this burner for something with a little more oomph and we'll see how long they stay zeroed out ...
From the Debian security FAQ:
Q: How does testing get security updates?
A: Security updates will migrate into the testing distribution via unstable. They are usually uploaded with their priority set to high, which will reduce the quarantine time to two days. After this period, the packages will migrate into testing automatically, given that they are built for all architectures and their dependencies are fulfilled in testing.
Two days isn't exactly bad going. About 3 pico-Microsofts, I'd say ;-)
I suspect blacklists are about the only useful weapon here, but even they are only so successful.
I've started blacklisting at the firewall instead of the mail server. Partly because it's a lot less load on the system, and partly because I'd rather those machines weren't on "my" internet in any way.
I wonder if some kind of collaborative filtering at this level would help reduce the problem. If people start seeing that their zombied machine can't access anywhere on the intarweb they might start to try and figure out why no-one wants to talk to them anymore. Blocking at the mail server is far less visible because they never knew that the mails were being sent in the first place.
Yes, I know that this has problems with dynamic IPs, but I don't have a better solution right now. Hopefully the complaints propagated up to the ISPs will be propagated down to the users of compromised machines with the assistance of a big cluestick and/or account cancellation.
Well, if it's Turing complete, then we're not talking about regular expressions any more, by definition ;-)
More seriously, how would you write a palindrome detector?
You could, in theory, use heavy ECC to tolerate a substantial defect rate.
There are DRAM technologies in production right now that do ECC on die to improve effective yield. The 90-98% yield you quote is probably effective yield, i.e. yield after error correction (either via ECC or on-die patching) is included.
Even a Cadbury Dark bar is better chocolate than Milka.
IIRC, most of the "Cadbury" chocolate available in the US is actually made by Hershey under license. If you want real Cadbury's you have to do some hunting around in delis. And, yes, it does taste completely different to the rebadged Hershey crap.
What makes you think MS is _required_ to give you any updates whatsoever for any of their products?
If I paid for something and it's broken, then I expect the manufacturer to repair it free of charge. If, however, we're talking about additional features then I would agree with you.
Right, because the Germans invented DES which started the rush of crypto algorithms while IBM (an American company) was still using polyalphabetic substitution cipers. No, wait, it's the other way around, stupid ass. You're not only wrong, but stupid.
One thing Capitalism does very well is foster innovation, both in invention and improvement of other inventions. We didn't invent the rocket, but we made it better. We invented the atomic bomb. We made serious cryptography.
Partly true.
http://www.schneier.com/crypto-gram-9805.htmlWe invented the automobile and the cotton gin.
The automobile? As apparently everyone but you knows, you're dead wrong there:
http://www.uh.edu/engines/epi1596.htmIt's amusing to watch people go out of their way to try to find fault with the USA. History won't even bother recording you guys.
The historical record is only of relevance to those who actually bother using it to check their facts before posting.