My guess is that what you're seeing is all those people who said during the boom: I know I'll make a career out of making web pages by hand. So they learned to create web pages, by hand, in html. Now they're out of work, and calling themselves programmers, because, you know, they work with computers after all.
Ok, please talk to us after you've crunched. How long have you been there? Have you shipped a title yet? You probably won't be able to get a better job if you don't ship a title. I think everyone understands that there are good times in the game business. It's just that they don't make up for the bad in the medium run.
Algorithm actually has the same issue. Note that you said: N(n) = n mod 10, n>=0
What's n? What's mod? What's greater than? All of that requires the backing of a language. And lest you call for the language of mathematics, that's the language with the single largest definition in existence.
In my custom language, I can express the same algorithm as:
Q
Pick any other sequence, and I'll express it as a single character too. It's a very poor definition of the problem.
An excellent point. On reflecting on this comment, I realized it would be relatively trivial to design a language for which the shortest program expressing any given sequence was a single character.
Indeed, and the open question is whether quantum mechanics is the bottom, and there is no more we can know that would allow us to predict quantum mechanical outcomes, or if at some deeper level, it all becomes fixed again.
The problem everyone has with that definition is that it's unmeasurable, because we don't know how to generate the shortest program for a given sequence.
Indeed, there is very little to possibly zero actual randomness in the universe. Many scientists think there may be no actual randomness (just stuff for which we don't know how to examine the internal state yet).
I think you misunderstood me, and that misunderstanding is probably my fault: in all statically typed languages (the languages for which hungarian is promoted) the type of a variable or function is always available somewhere. I wasn't claiming that someone wrote a comment, or whatever ('documented'). Just that the type information is guaranteed to be discoverable in other ways.
In spite of the fact that I already debunked the actual arguments made in that article. If I've already poked holes in the content, what's the point of linking him other than appeal to authority? I'm left with either repeating myself, or attempting to debunk the appeal to authority. I went with the second. I mostly could care less about Joel, but I'm pretty strongly against logical fallacy.
As I suggested in response to others, if you're a big joel fan, you're not bright enough to work for me. I have to ask: have you thought about why you have this problem? Does the fact that you are passing around naked strings representing both sql and html not bother you? Is the obviously better solution not obvious enough? Every language I have seen has a straightforward solution to this problem, including c, c++, java (the primary languages for which I see hungarian prescribed).
The problem with hungarian notation is that it conveys the most useless information about variables. Information which is guaranteed to be documented elsewhere. It harms subsequent maintainers because it encourages you to store useless information in variable names which of course winds up discouraging you from storing useful information in those variable names instead.
I'm in complete agreement with everything you said, except that Hungarian notation is a good tool for what you want. It's a bad, bad tool. Bad for developers, bad for maintainers.
Heh. The discussion you reference is linking Joel on software. I send beginners there and ask: give me a list of all the things Joel is obviously wrong about. I keep the ones who are smart enough to come back and say: I don't want to spend that much time writing it all down.
You're making my case for me: you're asking programmers to put overhead time into their work to make hungarian notation work (and demanding that they not make mistakes!). Hungarian notation is error-prone. It's not going to help your blind coworker if it is wrong.
We just need to build big boxes to put them in, equipped with xbox360s and a food supply. Call them 'the forcefully retired'. Wait for them to die. Problem solved.
That's building a tool (understanding scope and type) into your coding style. Always a bad idea. Build the tools around your coding style, and keep your style as elegant and simple as possible.
As a relatively trivial example of where this goes wrong, refactoring such a variable can trivially result in the code lying to you about the type and scope of a variable. If you instead have a tool that will tell you the scope and type based on inspection, it will never lie to you.
Hungarian notation was a bad bad idea created by someone with a poor understanding of and lacking insight into the problem they were trying to solve.
I'm always shocked by the number of people who report this. I mean, surely if you're not a rebate mailer, you learn at some point not to factor the rebate into the purchase price? I've also never had a problem with a rebate. Maybe because I have easy access to a photocopier, and always include a note that indicates that I kept a photocopy of everything in the event that there is a problem with the rebate. I've done about 20 rebates of $10 or more over the last 5 years with zero failures.
Well, a few, intermittently. Not what I would describe as 'a lot'. I guess we're firing a lot of bullets over there. But not near as many as we're firing over here, and we're killing not nearly as many of them as we are of us, and I'd hate to think we're at war with us. http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5002a1. htm
In fairness to you, I'm quite confident there's zero chance the original statement wasn't erroneous. But it was an old hobby of mine to try to come up with rational explanations for the similar garbage in star trek episodes, and it has become something of a habit to try to explain away these kinds of mistakes.:-)
Slashdot Mirror
My guess is that what you're seeing is all those people who said during the boom: I know I'll make a career out of making web pages by hand. So they learned to create web pages, by hand, in html. Now they're out of work, and calling themselves programmers, because, you know, they work with computers after all.
Ok, please talk to us after you've crunched. How long have you been there? Have you shipped a title yet? You probably won't be able to get a better job if you don't ship a title.
I think everyone understands that there are good times in the game business. It's just that they don't make up for the bad in the medium run.
Algorithm actually has the same issue. Note that you said:
N(n) = n mod 10, n>=0
What's n? What's mod? What's greater than?
All of that requires the backing of a language. And lest you call for the language of mathematics, that's the language with the single largest definition in existence.
In my custom language, I can express the same algorithm as:
Q
Pick any other sequence, and I'll express it as a single character too. It's a very poor definition of the problem.
An excellent point. On reflecting on this comment, I realized it would be relatively trivial to design a language for which the shortest program expressing any given sequence was a single character.
Kudzu doesn't have nice oily seeds that can be trivially pressed for oil.
Algae production requires a lot more water and shit than jatropha. They're complementary for different locations of the world.
Of course you can, it's a plant with leaves. Whether or not you'd survive the experience, or get high, I leave to the botanists in the crowd.
Indeed, and the open question is whether quantum mechanics is the bottom, and there is no more we can know that would allow us to predict quantum mechanical outcomes, or if at some deeper level, it all becomes fixed again.
The problem everyone has with that definition is that it's unmeasurable, because we don't know how to generate the shortest program for a given sequence.
Indeed, there is very little to possibly zero actual randomness in the universe. Many scientists think there may be no actual randomness (just stuff for which we don't know how to examine the internal state yet).
I think you misunderstood me, and that misunderstanding is probably my fault: in all statically typed languages (the languages for which hungarian is promoted) the type of a variable or function is always available somewhere. I wasn't claiming that someone wrote a comment, or whatever ('documented'). Just that the type information is guaranteed to be discoverable in other ways.
I assumed people were linking Joel because they're making an appeal to authority:
http://en.wikipedia.org/wiki/Appeal_to_authority
In spite of the fact that I already debunked the actual arguments made in that article. If I've already poked holes in the content, what's the point of linking him other than appeal to authority? I'm left with either repeating myself, or attempting to debunk the appeal to authority. I went with the second. I mostly could care less about Joel, but I'm pretty strongly against logical fallacy.
As I suggested in response to others, if you're a big joel fan, you're not bright enough to work for me.
I have to ask: have you thought about why you have this problem? Does the fact that you are passing around naked strings representing both sql and html not bother you? Is the obviously better solution not obvious enough? Every language I have seen has a straightforward solution to this problem, including c, c++, java (the primary languages for which I see hungarian prescribed).
The problem with hungarian notation is that it conveys the most useless information about variables. Information which is guaranteed to be documented elsewhere. It harms subsequent maintainers because it encourages you to store useless information in variable names which of course winds up discouraging you from storing useful information in those variable names instead.
I'm in complete agreement with everything you said, except that Hungarian notation is a good tool for what you want. It's a bad, bad tool. Bad for developers, bad for maintainers.
Heh. The discussion you reference is linking Joel on software. I send beginners there and ask: give me a list of all the things Joel is obviously wrong about. I keep the ones who are smart enough to come back and say: I don't want to spend that much time writing it all down.
You're making my case for me: you're asking programmers to put overhead time into their work to make hungarian notation work (and demanding that they not make mistakes!). Hungarian notation is error-prone. It's not going to help your blind coworker if it is wrong.
We just need to build big boxes to put them in, equipped with xbox360s and a food supply. Call them 'the forcefully retired'. Wait for them to die. Problem solved.
That's building a tool (understanding scope and type) into your coding style. Always a bad idea. Build the tools around your coding style, and keep your style as elegant and simple as possible.
As a relatively trivial example of where this goes wrong, refactoring such a variable can trivially result in the code lying to you about the type and scope of a variable. If you instead have a tool that will tell you the scope and type based on inspection, it will never lie to you.
Hungarian notation was a bad bad idea created by someone with a poor understanding of and lacking insight into the problem they were trying to solve.
I would guess the total was about $400-$500. Or about $8/month overall.
I'm always shocked by the number of people who report this. I mean, surely if you're not a rebate mailer, you learn at some point not to factor the rebate into the purchase price?
I've also never had a problem with a rebate. Maybe because I have easy access to a photocopier, and always include a note that indicates that I kept a photocopy of everything in the event that there is a problem with the rebate. I've done about 20 rebates of $10 or more over the last 5 years with zero failures.
Ouch that's expensive. And big (in physical dimensions) compared to:i d=5623741&affid=10000483
http://www.computers4sure.com/product.asp?product
I guess it may be somewhat faster, but both are approaching the limits of what you can push through a sata interface.
http://download.micron.com/pdf/datasheets/flash/na nd/4gb_nand_m40a.pdf
promises data retention of 10 years. I would guess that it will function longer than that, but only if you refresh the data.
http://download.micron.com/pdf/datasheets/flash/na nd/4gb_nand_m40a.pdf
Says 100,000 program/erase cycles right on the first page (though I do note they only 'guarantee' 1k writes for the first block).
Well, a few, intermittently. Not what I would describe as 'a lot'. I guess we're firing a lot of bullets over there. But not near as many as we're firing over here, and we're killing not nearly as many of them as we are of us, and I'd hate to think we're at war with us.. htm
http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5002a1
In fairness to you, I'm quite confident there's zero chance the original statement wasn't erroneous. But it was an old hobby of mine to try to come up with rational explanations for the similar garbage in star trek episodes, and it has become something of a habit to try to explain away these kinds of mistakes. :-)