Remember the time when the most acclaimed minds in the world thought that the world was flat?
No, actually. Throughout recorded history man (the educated ones, at least) has known the world is roughly spherical.
Columbus didn't have to convince Ferdinand and Isabelle that the world was round; they knew that as well as we do. They just also knew as well as we do how big it was (Thales's measurement of the circumference of the earth was not surpassed in accuracy until the 18th century). And they didn't know if there was a continent between the Iberian peninsula and China (and neither did Columbus unless he heard it from an Icelander when he was there).
Very easily, actually. In fact, most of the "great" proofs in mathematics are proof of a negative. I offer one as an example:
Hypothesis: There is no greatest prime number
Proof: Suppose there were a greatest prime number; let's call it P.
Then, by using a sieve method we could create a list of all prime numbers.
Postulate that this has been done, and call the resulting list p1,p2,p3...pn,P
Consider the product of this list, that is, p1 * p2 * p3... * pn * P; call that product Q.
Consder the number Q+1. Q+1 is relatively prime to every number in the list p1,p2,p3...pn,P because Q+1 differs from a multiple of each number in that list by only 1.
But p1,p2,p3...pn,P is by hypothesis a complete list of all prime numbers.
Therefore Q+1 is relatively prime to all prime numbers less than it, and is therefore itself prime.
But Q+1 is greater than P, because Q = P * some integer and Q+1 > Q
Therefore the results are in contradiction to the hypothesis, and by reductio the hypothesis is false.
There, we just proved a negative. Where do people get the ridiculous idea that a negative can't be proven?
Umm, yes if you install the Linux compatibility package for Open, Free, or NetBSD you'll have a/proc filesystem. It doesn't come that way naturally, though.
Once your application will get "serious", you will have to occasionally dig down to the bare metal and debug at a lower level, optimize for performance, etc. And at that point you will need "real" knowledge. My advice: make sure you learn SQL
I hate to defend Windows (as much as I despise it), but people often don't realize what's available to them on it: >> cat (isn't there a DOS equiv to cat...?)
type
>> cmdln grep
findstr
>> w, cal, bc, lynx, ssh
All ported to Windows and as equally supported as they are on UNIX.
Re:Distribution on Windows
on
Why Use GTK+?
·
· Score: 4, Informative
I don't know about vbrun but the MFC DLL is less than a meg in size.
And MFC.DLL is the equivalent of maybe half of glib and the stubs for the rest of the libs. Count all your non-Office OCX files; those are delivering the rest of the equivalent capability
Re:GTK is alright...but no raves
on
Why Use GTK+?
·
· Score: 3, Informative
Then again, actually GTK has what you want anyways, despite my earlier response rant:
Re:GTK is alright...but no raves
on
Why Use GTK+?
·
· Score: 1
Literally, I will be unhappy with *every* GUI toolkit until I find one where you can express an entire dialog box in a SINGLE LINE
Umm... you're supposed to roll that yourself. Write either a wrapper function or a macro for the common stuff you do with your toolkit. That, or move on from C to a more applicative language like lisp or ocaml.
I mean, it's a five-minute thing to write the D wrapper you want; do it and move on with your life.
I can't see any real need to use a macro here, even if you need to use this expression in a function throughout the code -- simply use higher order functions for that.
Yeah, my explanation really missed the point (sorry). I didn't mean to write the quadratic equation as a macro (that would be silly). I meant I usually write macros that write my equations for me. If you write any decent-sized application in lisp you're usually writing to an embedded language you made with macros, anyways.
Why the hell would you use a *macro* to write a simple math expression?
You would use it to generate the appropriate function at compile (or failing that, run) time and pass it to the arguments. I mean, if you somehow knew you were only ever going to use that one function you would pre-define it and probably inline it, but in a real situation...
Wow, I mean, I know LISP heads love using macros excessively (and unnecessarily)
I'm trying to think of how any macro that doesn't break the program is "unneccessary" and I can't. Macros make decisions at compile time that then don't have to be made at run time. They're brilliant.
But if you're going to deviate from what people know already, use RPN
Yeah, I love Forth and Joy. The only problem with RPN is you then have fixed arity of functions (well, Forth allows currying but only by passing execution tokens, so not *really*). Standard Polish notation works great because you have defined precedence and there's no need to have fixed arity.
Except for the 9th level, where you're forced to use ML.
Which, oddly enough, is exactly what the Visual Studio team is developing next, in the form of F#.
In fairness, real ML and its cousin ocaml are absolute dreams to develop in. Great libraries, great compilers (the ocaml native compiler regularly produces faster executables than a C compiler), and no development environments to speak of (though emacs has a great inferior caml mode).
OK, you parenthesis-phobes, tell me what's so much better about:
f(g(x->y));
as opposed to
(f (g (get 'y x)))
?
I absolutely love developing in Lisp and I don't understand why the notation bugs so many people. 3 function calls in C = 3 sets of parentheses. 3 function calls in Lisp = 3 sets of parentheses. The only difference is that operators in Lisp are functions, but on the other hand blocks after control structures don't require braces, since you alread have the parentheses.
Just a pet peeve of mine
Besides, there's no better IDE out there than SLIME on Emacs.
She is the respondant against whom the petitioner is seeking relief. If she had actually stolen something from these people, the state would prosecute her; as it is, she may or may not be found liable for damages done to the petitioner. Ironically enough, she would probably be in better shape if this *were* a criminal matter: criminal convictions require demonstrable harm, intent, and proof beyond a reasonable doubt. Civil actions merely require that the harm of a false victory for her outweighs the harm of a possible false victory for RIAA by 1% (to use the 51% / 49% description someone mentioned).
And incidentally, despite the fact that the media keeps saying the tortious action was "downloading", the actual tortious action by statute was *uploading*.
The phrase "a jury of your peers" has nothing to do with the American legal system, really. It was an English Commonlaw right that guaranteed a nobleman couldn't be tried by peasants. The only holdover I know of that it still has is that an officer facing a court martial can choose for his jury to only be fellow commissioned officers and not enlisted personnel or warrant officers.
All 50 states, all 3 territories, and the District of Columbia require biometrics for their Drivers Licenses and non-driver ID's. Pictures are a type of biometrics.
they aren't required by any leap of faith
No, but they require $50 or so depending on your state, not to mention a stable address and not being a debtor to the State. Poll taxes are unconstitutional for a reason.
Don't use RAID to begin with. It's needlessly expensive for a home server and introduces an unneccesary point of failure. Maybe I have bad luck, but I have experienced many more RAID controller failures than I have hard drive failures. I even once got redundant RAID controllers, and the controller-controller failed. Let's say you were willing to spend $500 on this RAID solution: rather than do that, spend $250 on improving your backups and pocket the difference.
Well, that's Tarkovsky for you. He's IMO the greatest filmmaker yet; he somehow managed to slip 4 or 5 absolute masterpieces past the Soviet censors... how he did that is beyond me. As good as Solaris is, his best is far and away Andrey Rublyov: dark, brooding, painful to watch, and absolutely stunning.
I'd attribute as much as 75% of them to user error and/or ignorance.
Well, yeah. Every single problem I listed except for the problems opening different versions was from user error. 95% of my time is billable to user error. That's my point.
People don't know how to use MS Office. They don't know how to use OOo. The support costs actually caused by bugs in the applications themselves are so miniscule compared to the support costs caused by user ignorance; I don't see why people say OOo would have a higher support cost since the users are stupid whichever suite you use.
$65k, depending on where you're located, could be much cheaper then the amount of money you'll have to spend on supporting Open Office.
Where is this magical world people are from in which MS Office works out of the box and doesn't require support? I "tech guy" for about 20 small organizations and as of this last invoice 65% of my time is supporting people on MS Office (90% if you count Outlook) because it freezes / craps out / corrupts their files / won't open older versions / won't open newer versions / does weird things where bullets aren't all the same size / messes up multi-column calculations half the time but not the other half of the time / etc.
Do you really work with MS Office installations that don't require support?
Slashdot Mirror
No, actually. Throughout recorded history man (the educated ones, at least) has known the world is roughly spherical.
Columbus didn't have to convince Ferdinand and Isabelle that the world was round; they knew that as well as we do. They just also knew as well as we do how big it was (Thales's measurement of the circumference of the earth was not surpassed in accuracy until the 18th century). And they didn't know if there was a continent between the Iberian peninsula and China (and neither did Columbus unless he heard it from an Icelander when he was there).
Very easily, actually. In fact, most of the "great" proofs in mathematics are proof of a negative. I offer one as an example:
Hypothesis: There is no greatest prime number
Proof: Suppose there were a greatest prime number; let's call it P. ... * pn * P; call that product Q.
Then, by using a sieve method we could create a list of all prime numbers.
Postulate that this has been done, and call the resulting list p1,p2,p3...pn,P
Consider the product of this list, that is, p1 * p2 * p3
Consder the number Q+1. Q+1 is relatively prime to every number in the list p1,p2,p3...pn,P because Q+1 differs from a multiple of each number in that list by only 1.
But p1,p2,p3...pn,P is by hypothesis a complete list of all prime numbers.
Therefore Q+1 is relatively prime to all prime numbers less than it, and is therefore itself prime.
But Q+1 is greater than P, because Q = P * some integer and Q+1 > Q
Therefore the results are in contradiction to the hypothesis, and by reductio the hypothesis is false.
There, we just proved a negative. Where do people get the ridiculous idea that a negative can't be proven?
Umm, yes if you install the Linux compatibility package for Open, Free, or NetBSD you'll have a /proc filesystem. It doesn't come that way naturally, though.
Install Open, Free, or NetBSD sometime and look for the /proc filesystem. It's not there.
So in your world, "SQL" means "bare metal"?
I hate to defend Windows (as much as I despise it), but people often don't realize what's available to them on it:
>> cat (isn't there a DOS equiv to cat...?)
type
>> cmdln grep
findstr
>> w, cal, bc, lynx, ssh
All ported to Windows and as equally supported as they are on UNIX.
And MFC.DLL is the equivalent of maybe half of glib and the stubs for the rest of the libs. Count all your non-Office OCX files; those are delivering the rest of the equivalent capability
Then again, actually GTK has what you want anyways, despite my earlier response rant:
Umm... you're supposed to roll that yourself. Write either a wrapper function or a macro for the common stuff you do with your toolkit. That, or move on from C to a more applicative language like lisp or ocaml.
I mean, it's a five-minute thing to write the D wrapper you want; do it and move on with your life.
Yeah, my explanation really missed the point (sorry). I didn't mean to write the quadratic equation as a macro (that would be silly). I meant I usually write macros that write my equations for me. If you write any decent-sized application in lisp you're usually writing to an embedded language you made with macros, anyways.
You would use it to generate the appropriate function at compile (or failing that, run) time and pass it to the arguments. I mean, if you somehow knew you were only ever going to use that one function you would pre-define it and probably inline it, but in a real situation...
I'm trying to think of how any macro that doesn't break the program is "unneccessary" and I can't. Macros make decisions at compile time that then don't have to be made at run time. They're brilliant.
Yeah, I love Forth and Joy. The only problem with RPN is you then have fixed arity of functions (well, Forth allows currying but only by passing execution tokens, so not *really*). Standard Polish notation works great because you have defined precedence and there's no need to have fixed arity.
OK,
*shrug* I guess it's more parenthetic (but also more legible), but you'd probably use a macro to write that anyways.And doesnt x^2 mean x bitwised XOR'd with 2?
Which, oddly enough, is exactly what the Visual Studio team is developing next, in the form of F#.
In fairness, real ML and its cousin ocaml are absolute dreams to develop in. Great libraries, great compilers (the ocaml native compiler regularly produces faster executables than a C compiler), and no development environments to speak of (though emacs has a great inferior caml mode).
sigh
OK, you parenthesis-phobes, tell me what's so much better about:
as opposed to?I absolutely love developing in Lisp and I don't understand why the notation bugs so many people. 3 function calls in C = 3 sets of parentheses. 3 function calls in Lisp = 3 sets of parentheses. The only difference is that operators in Lisp are functions, but on the other hand blocks after control structures don't require braces, since you alread have the parentheses.
Just a pet peeve of mine
Besides, there's no better IDE out there than SLIME on Emacs.
You're lucky you got paid.
I've learned to walk off of job sites quietly the day after checks are cut. Otherwise these parasites will suck you out of every dime they can.
No, she's not.
She is the respondant against whom the petitioner is seeking relief. If she had actually stolen something from these people, the state would prosecute her; as it is, she may or may not be found liable for damages done to the petitioner. Ironically enough, she would probably be in better shape if this *were* a criminal matter: criminal convictions require demonstrable harm, intent, and proof beyond a reasonable doubt. Civil actions merely require that the harm of a false victory for her outweighs the harm of a possible false victory for RIAA by 1% (to use the 51% / 49% description someone mentioned).
And incidentally, despite the fact that the media keeps saying the tortious action was "downloading", the actual tortious action by statute was *uploading*.
The phrase "a jury of your peers" has nothing to do with the American legal system, really. It was an English Commonlaw right that guaranteed a nobleman couldn't be tried by peasants. The only holdover I know of that it still has is that an officer facing a court martial can choose for his jury to only be fellow commissioned officers and not enlisted personnel or warrant officers.
All 50 states, all 3 territories, and the District of Columbia require biometrics for their Drivers Licenses and non-driver ID's. Pictures are a type of biometrics.
No, but they require $50 or so depending on your state, not to mention a stable address and not being a debtor to the State. Poll taxes are unconstitutional for a reason.
Don't use RAID to begin with. It's needlessly expensive for a home server and introduces an unneccesary point of failure. Maybe I have bad luck, but I have experienced many more RAID controller failures than I have hard drive failures. I even once got redundant RAID controllers, and the controller-controller failed. Let's say you were willing to spend $500 on this RAID solution: rather than do that, spend $250 on improving your backups and pocket the difference.
Can they put it on a big frigging shark?
nasema kiswahili vidogo tu, lakini na kamusi ninajifunza. karibuni kamusi.
(still not sure whether -funza is the medial or causative...)
Well, that's Tarkovsky for you. He's IMO the greatest filmmaker yet; he somehow managed to slip 4 or 5 absolute masterpieces past the Soviet censors... how he did that is beyond me. As good as Solaris is, his best is far and away Andrey Rublyov : dark, brooding, painful to watch, and absolutely stunning.
Well, yeah. Every single problem I listed except for the problems opening different versions was from user error. 95% of my time is billable to user error. That's my point.
People don't know how to use MS Office. They don't know how to use OOo. The support costs actually caused by bugs in the applications themselves are so miniscule compared to the support costs caused by user ignorance; I don't see why people say OOo would have a higher support cost since the users are stupid whichever suite you use.
Where is this magical world people are from in which MS Office works out of the box and doesn't require support? I "tech guy" for about 20 small organizations and as of this last invoice 65% of my time is supporting people on MS Office (90% if you count Outlook) because it freezes / craps out / corrupts their files / won't open older versions / won't open newer versions / does weird things where bullets aren't all the same size / messes up multi-column calculations half the time but not the other half of the time / etc.
Do you really work with MS Office installations that don't require support?