You're giving E.T. Bell's "Men of Mathematics" (almost undoubtedly your source) entirely too much credit. While a fun read, it is not a serious work of historical scholarship; rather it is targeted at a popular audience. Bell's sources on the subject of Cantor are mainly Cantor's own letters; with regard to Kronecker's alleged attacks on Cantor, much of those those letters are directly contradicted by published transcripts of Kronecker's lectures.
Certainly it is true that some of "the smart people" may have not understood Cantor's theory of transfinite numbers; Kronecker, however, surely did, and Poincare did as well. What they objected to was the use of "number" -- a term which at the time had a radically different precise mathematical meaning than it does today -- to describe such constructions.
I had no difficulty getting my girlfriend's mini convertible home this way (about 15 miles) without any coolant at all. Didn't even run hot./shrug, usual disclaimer about anecdotes.
Any "engineer" who doesn't figure out to just roll down the windows, turn the heater on full blast, and turn the engine off at traffic lights and downhills doesn't deserve a promotion, anyway.
I guess Google has finally decided to ditch their "don't be evil" slogan: there's nothing more evil than trying to overturn the will of the people after a democratic vote. The people have spoken. Google should live with it.
If "the people" (meaning 50% of the minority of the population that actually turned out to vote) passed a ballot initiative that condemned you to death, would you accept their judgement?
I understand that you might see the LDS Church's views as hateful. We can agree to disagree and that is why this country is great.
No, we can't.
Using the ballot proposition system to take explicitly take rights away from a minority group, when those rights pose no risk to anyone else's rights and are allowed to the majority is wrong, unethical, and of questionable legality.
Funding groups that want to do so, on the basis of delusions in which an imaginary man in the sky tells you that said minority group is evil, is hateful at best, but probably more aptly described as psychopathic or paranoid-delusional.
as always, one should use the level of abstraction that one feels comfortable with to deal with the specific "algorithmic needs", no?
Of course. But one shouldn't pretend that this makes the language one has chosen somehow better suited to the task than the language another person would choose, even if their solution looks unintelligible to you. It just means that you don't understand the way the other person did it.
I did get one yesterday. I knew I wanted one, just did not know when... Also, know that I usually take the wait and see approach to tech and not an early adopter. I am very tech savvy
Wait... what?
You take the wait and see approach to tech, so you went out and bought an completely untested product on launch day?
If you ever break super_sort, you will find out as soon as you run your unit tests.
Unless you break it in a way that the tests don't exercise. To be perfectly clear, let's suppose your "super_sort" only sorts two doubles. Even for this extremely limited sorting problem, there are 2^128 possible inputs. If you can test one input per cycle on a 3.2 ghz machine, it will take you only... 3371963459626879434053 years to completely test the input space.
IF YOU'RE LUCKY, you will make really really obvious mistakes that break almost all inputs, and your tests will find them. You won't always be lucky.
I'm not saying that you shouldn't test things. Just don't suffer from the misapprehension that tests passing is proof that the code works.
Your first "sentence" makes no sense. That point aside, you seem to be saying that TDD will test every possible usage of a program, thereby guaranteeing that bugs will be found. This may be possible for very, very limited toy examples, but is impossible for any real world code. You can't even test all possible input pairs to a single-precision floating-point divide in a reasonable time period for a development cycle -- how will you completely test code that does anything interesting?
Of course, it could be that I'm just misinterpreting the (horribly mangled, non-sensical) sentence you wrote.
Probably 10% of the ugliest code I read is mangled *precisely because* someone went out of their way to avoid using a goto when it really was the appropriate control structure. They couldn't even tell you why they did it, just that they heard somewhere to avoid goto. They probably think that it's a performance hazard or something.
You don't. You make friends with some engineers, find out about their former former college roommate who is the best engineer they know and who is between jobs, and either pay him a great salary or give him 50% ownership.
If you want great work, you need to hire someone great to do it, and people who are great aren't (generally) bidding on cheapo contract jobs.
The processor is just a Core2 Duo T5750 which barely clocks at 2.0ghz.
This processor should be completely adequate for the task. If the device isn't performing well, the problem is software, not the hardware. I mean, christ, it's orders of magnitude more capable than the processor in an iPhone, and the keyboard on the iPhone doesn't "show as much latency as someone trying to play WoW on a 28.8kbps dial-up connection".
Re:The main thing missing, is introspection
on
Boost 1.36 Released
·
· Score: 1
As long as I can still implement an ALU in the preprocessor, I'll be happy.
Slashdot Mirror
You're giving E.T. Bell's "Men of Mathematics" (almost undoubtedly your source) entirely too much credit. While a fun read, it is not a serious work of historical scholarship; rather it is targeted at a popular audience. Bell's sources on the subject of Cantor are mainly Cantor's own letters; with regard to Kronecker's alleged attacks on Cantor, much of those those letters are directly contradicted by published transcripts of Kronecker's lectures.
Certainly it is true that some of "the smart people" may have not understood Cantor's theory of transfinite numbers; Kronecker, however, surely did, and Poincare did as well. What they objected to was the use of "number" -- a term which at the time had a radically different precise mathematical meaning than it does today -- to describe such constructions.
16-proc MacPro
O RLY? Where can I get me one of them?
That aside, this is a pretty decent description of what GCD is good for.
Puma was OS X 10.1.
Isn't it?
Not really, no.
Anyone who would wear a turtleneck and glasses that only cost $150 together is *obviously* not a mac user.
(I kid... sorta...)
Processor with markedly higher power draw achieves superior benchmark results. News at 11.
I had no difficulty getting my girlfriend's mini convertible home this way (about 15 miles) without any coolant at all. Didn't even run hot. /shrug, usual disclaimer about anecdotes.
Any "engineer" who doesn't figure out to just roll down the windows, turn the heater on full blast, and turn the engine off at traffic lights and downhills doesn't deserve a promotion, anyway.
I guess Google has finally decided to ditch their "don't be evil" slogan: there's nothing more evil than trying to overturn the will of the people after a democratic vote. The people have spoken. Google should live with it.
If "the people" (meaning 50% of the minority of the population that actually turned out to vote) passed a ballot initiative that condemned you to death, would you accept their judgement?
I understand that you might see the LDS Church's views as hateful. We can agree to disagree and that is why this country is great.
No, we can't.
Using the ballot proposition system to take explicitly take rights away from a minority group, when those rights pose no risk to anyone else's rights and are allowed to the majority is wrong, unethical, and of questionable legality.
Funding groups that want to do so, on the basis of delusions in which an imaginary man in the sky tells you that said minority group is evil, is hateful at best, but probably more aptly described as psychopathic or paranoid-delusional.
as always, one should use the level of abstraction that one feels comfortable with to deal with the specific "algorithmic needs", no?
Of course. But one shouldn't pretend that this makes the language one has chosen somehow better suited to the task than the language another person would choose, even if their solution looks unintelligible to you. It just means that you don't understand the way the other person did it.
I have no idea how i would start to code that in C, python, etc. in a way that's remotely efficient ;)
How about:
#include <clapack.h>
dgesdd( argument list );
This sort of thing is a feature of libraries, not an inherent advantage of one language.
Safari is far from "fully featured". It lacks a ton of features that one would expect from a browser today.
Care to name them? I don't know how much a feature weighs, but I'm pretty sure that it's less than a pound, so I'm expecting at least 2000 of them.
In all seriousness, I don't find myself missing browser features on my iPhone. Do tell me what I'm missing out on.
Is McDonalds the best food on the market?
I did get one yesterday. I knew I wanted one, just did not know when... Also, know that I usually take the wait and see approach to tech and not an early adopter. I am very tech savvy
Wait... what?
You take the wait and see approach to tech, so you went out and bought an completely untested product on launch day?
I'm in a software engineering class listening to how to use metrics on code.
No, you're in a software engineering class posting on Slashdot.
On the contrary. IEEE-1394 is most definitely a standard.
http://en.wikipedia.org/wiki/FireWire
I'm not arguing with the approach to testing you describe, I'm just pointing out that your statement
If you ever break super_sort, you will find out as soon as you run your unit tests.
is false (or wishful thinking, whichever you prefer).
If you ever break super_sort, you will find out as soon as you run your unit tests.
Unless you break it in a way that the tests don't exercise. To be perfectly clear, let's suppose your "super_sort" only sorts two doubles. Even for this extremely limited sorting problem, there are 2^128 possible inputs. If you can test one input per cycle on a 3.2 ghz machine, it will take you only... 3371963459626879434053 years to completely test the input space.
IF YOU'RE LUCKY, you will make really really obvious mistakes that break almost all inputs, and your tests will find them. You won't always be lucky.
I'm not saying that you shouldn't test things. Just don't suffer from the misapprehension that tests passing is proof that the code works.
Your first "sentence" makes no sense. That point aside, you seem to be saying that TDD will test every possible usage of a program, thereby guaranteeing that bugs will be found. This may be possible for very, very limited toy examples, but is impossible for any real world code. You can't even test all possible input pairs to a single-precision floating-point divide in a reasonable time period for a development cycle -- how will you completely test code that does anything interesting?
Of course, it could be that I'm just misinterpreting the (horribly mangled, non-sensical) sentence you wrote.
Probably 10% of the ugliest code I read is mangled *precisely because* someone went out of their way to avoid using a goto when it really was the appropriate control structure. They couldn't even tell you why they did it, just that they heard somewhere to avoid goto. They probably think that it's a performance hazard or something.
How do i find someone good on rent a coder?
You don't. You make friends with some engineers, find out about their former former college roommate who is the best engineer they know and who is between jobs, and either pay him a great salary or give him 50% ownership.
If you want great work, you need to hire someone great to do it, and people who are great aren't (generally) bidding on cheapo contract jobs.
The bosun says before I'm through /
John Kanakanaka Too-ri-ay /
You'll curse your mother just for having you /
John Kanakanaka Too-ri-ay
(Apologies to everyone who cares not a whit for Sea Chanteys)
The processor is just a Core2 Duo T5750 which barely clocks at 2.0ghz.
This processor should be completely adequate for the task. If the device isn't performing well, the problem is software, not the hardware. I mean, christ, it's orders of magnitude more capable than the processor in an iPhone, and the keyboard on the iPhone doesn't "show as much latency as someone trying to play WoW on a 28.8kbps dial-up connection".
As long as I can still implement an ALU in the preprocessor, I'll be happy.