I'm pretty happy when I feel like I understand a single interrupt routine 100%, even when it is only a few lines and I wrote it!
Especially because it is usually in C, and who knows what the compiler actually did! If I had time to "understand" what actually happened to 100%, I'd have time to write all my code in ASM to begin with. Actually, that would be quicker for that goal...
And even then my understanding is only an approximation; all hardware is analog under the hood, after all!
I know for sure I haven't memorized all the stuff I'm supposed to "know" to do the job; and I'm planning from the start to do the whole job with the manual open.
And yet, I'm expected to communicate with others using strong, certain language, that might imply I already "know" everything I need to know! But really, I only know what the sources of information are that are needed to acquire the knowledge as I do the work. Oh, I know lots of stuff about the work, I've done very similar work before; but I sure didn't memorize it all. I couldn't pass an exam on the work I did last year, I might have mixed feelings about being forced to claim I "do" or "don't" "know" everything involved in doing it!
Even if we presume that an expert has accurate self knowledge, the mismatch between the communication needs and the actual problem-solving process is such that they still might reasonably feel like a "fraud." Not because there is anything wrong with what they're doing, but because they have to communicate about the process in a simplified way that they know is inaccurate. If the work is very precise work, like most technical jobs, then the people who are good at it might also be likely to notice that different and be bothered by it sometimes.
For example, I don't memorize APIs. I've been using the C stdlib for 20 years, but I still open up the man page for sprintf at least twice a week. I don't feel like a fraud to start with on that, but if you asked if I "know to use sprintf" and I said "yes," I'd have mixed feelings about it! Basically, anything that you're demanded to claim to "know," you should probably have mixed feelings about claiming it. Especially when you consider the experiences of Socrates after the Oracle claimed he was Wise!
It kind of bothers me a little to see so many people talk about imposter syndrome (AKA Dunning-Kruger).
No. IS and DK are OPPOSITES.
Imposter Syndrome: You think are dumber than you really are.
Dunning-Kruger: You think you are smarter than you really are.
They could be different points on the same curve for all you know.
If you don't know what you do know, and what you don't know, you're pretty much a perpetual fraud.
The words "smarter" and "dumber" in your comment doesn't even have meaning; it is just means you think they're not as Virtuous as you are. You're using it as if they were inherent attributes, but what is being measured is something different than that in both cases.
It could even be something as simple as, after a person gets past the Dunning-Kruger peak, they feel like an "impostor" unless they're a psychopath(*), right up until they reach a level of extreme knowledge that most people will never even attempt to reach in their life. So you start out new and overconfident, and then after you achieve some minimal competence you're doomed to always know that you're (still!) less competent than you used tell people you were! There is always this lurking sense of embarrassment and shame related to having realized that you were standing on Mt. Stupid, and then also continued moving forwards down the far side.
*) Please note that sociopath is a deprecated synonym for psychopath.
That's not an extraordinary claim, it is a mainstream 101-level claim that anybody claiming enough subject knowledge to demand a citation should already have.
Take some responsibility for your ignorance and attempt to look shit up before you tell people it needs a citation.
Because I refuse to do that for you, I'll instead give an anecdote: In my early 20s I worked at a plywood mill, and they often wanted to increase production seasonally. They would always (always!) over-work everybody. What they would do is go from 40 hours a week to 50 the next week, then they'd add Saturday to get to 60, and then on Monday morning they'd be really happy because they were so far "ahead," and then production would be so low all week that by the end of the week we were barely getting 50% out. They'd try to increase it by pushing people harder, but then it would go down further because of how much work had to be redone. Then eventually one of the people from the head office would get mad enough to come over and find out what was going on, and force them back down to 50 hours a week. Then over 2 weeks production would climb back to 85%. They never seemed to figure out that 85% is less than 100%, though. They also never enacted any policy changes that would prevent the pattern from repeating; every time that a production increase was desired, the same pattern would play out, with reduced productivity.
Eventually they added a third shift, which prevented the ability to work more hours in a day, and waste went down a lot. The only reason they didn't go out of business was that they had a lot of warehouse space, so they could still sell more product during the seasonal peaks than the rest of the year. That helped to hide the related drops in output.
It seems to be the natural result of letting managers teach themselves their job by their bootstraps.
This is a common misunderstanding about statistics. You do NOT need a lot of samples for statistical significance, and the number of datapoints needed is usually far below what people intuitively expect.
It is a natural mistake, because people presume that "significance" as technical jargon would be even more significant than the literary word, but actually it only means "probably not a math error." It tells you nothing about if your results are "significant" in the literary sense; the vast majority of scientific results that are "statistically significant" are still worthless and misleading, because math errors are not even the main challenge to overcome! The difficulty in predicting unknown explanations for results is really the problem that experiments attempt to solve. And it takes a lot of work. Being 95% sure that you didn't have a rounding error is not exactly a result that is likely to have significant implications, even if it is statistically significant.
If they'd just rename it to Statistically Cromulent people would be better able to understand it.
I'm not sure, I was taught that the whole modern "scientific method" was based on Newton, et al., rejecting the publishers as gatekeepers and publishing through the newsletters of their local Philosophic Societies. The main thing that was supposedly holding back Natural Philosophy (what science was called then) was the gatekeepers! It worked fine for a long time. At first the "Journals" were just a type of group publishing, just a like a Philosophic Society newsletter. But once they started assigning a system of Peers as gatekeepers, the whole thing was instantly a farce; the same thing that open publishing had already replaced once!
Peer review was valuable, historically, because it was done by your peers, accomplished through open access. Naming a class of Very Important People as Peers does not in any way achieve the same thing.
I guess I agree publishing companies provided a valuable service in the past, I'm just saying, the time when that was true was pre-Newton!
You've got some word salad there, maybe you should just keep playing with your food until it spells out whatever crazy bullshit is in your head.
You understand, little babies get born, it's been happening for millions of years. When I observe that, and you can't figure it out, and want to argue with it, you're not going to convince me that you're actually trying to communicate with the other humans. You're just playing with your word salad, and you have no idea even that the words you don't get to choose are coming from other humans!
That's very hand-wavy, not only am I not convinced of your conclusions, I'm not even convinced that you actually offered any logical reason for a person to even start considering what you said. Just pure assertion, and without any concept of which parts are opinion, which parts are purported facts, and which parts are generally accepted truisms.
In fact, I'm not even convinced that you understood that I was claiming that a different field, called QA, might actually be the field with the expert knowledge, and they might have a different truism than the developers have!
You don't seem to have even noticed what I was talking about, but you know you disagree. You know I'm doing it wrong, but you don't even know what I'm doing, what the use case is! Perhaps the answer is different depending on the size of the project? Perhaps it is different depending on the consequences of bugs? Perhaps it varies based on those things, and also the product lifecycle? Your insistence that there Must Be One True Answer proves that you're only considering the least-important categories of concerns, like those that web developers have.
My comment was not over my head, but perhaps your own comment was over your own head?
You made a movie reference. I bridged the gap between what happened in the movie, and what happened in the actual fucking story we're talking about. Those are my comments, not yours. If they went over somebody's head, why would you start with me? I might have even understand what I was saying!
Right, when somebody tells you that their magic process will save the world, you just assume that means it is just like seatbelts, and there is no need to find out if it is true or not.
What if it turns out that there is data suggesting that adopting a new programming process will not improve your software quality? Would adopting Virtuous processes still be the same as a seatbelt, or not?
Or are you suggesting that the value in seatbelts is in the feeling of Virtue that you get in wearing it, and that they're not actually effective?
Does calling a feature a "safety feature" actually tell me that it improves safety, or does it merely communicate your intent without telling me anything at all about the results of adopting your process?
I was always taught that tests should be written by a QA developer instead of the application developer, because if you write your own tests you have symmetrical oversights and logical mistakes.
Also, I was taught to be suspicious of anything that wants to "drive" the development process that isn't the use case! The goal is to match the implementation to the semantics of the problem domain, not to the semantics of the testing framework.
TDD is great for web apps, since there is unlikely to be enough profit to afford QA, but why do those even need a lot of thrash?
It's called an "IDE," an Integrated Development Environment.
They're popular, but they also don't really achieve anything. The devil remains in the details.
You can make it as easy as lego, and yet as with lego, anybody can stuff a few blocks together, but building something interesting to other people is much harder. Now, build something life-sized that solves some sort of problem that some human is describing to you...
Slashdot Mirror
Man up and stop being wish-washy. If you think you've succeeded beyond your merit, then be better.
OK, now define "merit."
You can do that, right? You're not some sort of impostor, are you?
I'm pretty happy when I feel like I understand a single interrupt routine 100%, even when it is only a few lines and I wrote it!
Especially because it is usually in C, and who knows what the compiler actually did! If I had time to "understand" what actually happened to 100%, I'd have time to write all my code in ASM to begin with. Actually, that would be quicker for that goal...
And even then my understanding is only an approximation; all hardware is analog under the hood, after all!
I know for sure I haven't memorized all the stuff I'm supposed to "know" to do the job; and I'm planning from the start to do the whole job with the manual open.
And yet, I'm expected to communicate with others using strong, certain language, that might imply I already "know" everything I need to know! But really, I only know what the sources of information are that are needed to acquire the knowledge as I do the work. Oh, I know lots of stuff about the work, I've done very similar work before; but I sure didn't memorize it all. I couldn't pass an exam on the work I did last year, I might have mixed feelings about being forced to claim I "do" or "don't" "know" everything involved in doing it!
Even if we presume that an expert has accurate self knowledge, the mismatch between the communication needs and the actual problem-solving process is such that they still might reasonably feel like a "fraud." Not because there is anything wrong with what they're doing, but because they have to communicate about the process in a simplified way that they know is inaccurate. If the work is very precise work, like most technical jobs, then the people who are good at it might also be likely to notice that different and be bothered by it sometimes.
For example, I don't memorize APIs. I've been using the C stdlib for 20 years, but I still open up the man page for sprintf at least twice a week. I don't feel like a fraud to start with on that, but if you asked if I "know to use sprintf" and I said "yes," I'd have mixed feelings about it! Basically, anything that you're demanded to claim to "know," you should probably have mixed feelings about claiming it. Especially when you consider the experiences of Socrates after the Oracle claimed he was Wise!
It kind of bothers me a little to see so many people talk about imposter syndrome (AKA Dunning-Kruger).
No. IS and DK are OPPOSITES.
Imposter Syndrome: You think are dumber than you really are.
Dunning-Kruger: You think you are smarter than you really are.
They could be different points on the same curve for all you know.
If you don't know what you do know, and what you don't know, you're pretty much a perpetual fraud.
The words "smarter" and "dumber" in your comment doesn't even have meaning; it is just means you think they're not as Virtuous as you are. You're using it as if they were inherent attributes, but what is being measured is something different than that in both cases.
It could even be something as simple as, after a person gets past the Dunning-Kruger peak, they feel like an "impostor" unless they're a psychopath(*), right up until they reach a level of extreme knowledge that most people will never even attempt to reach in their life. So you start out new and overconfident, and then after you achieve some minimal competence you're doomed to always know that you're (still!) less competent than you used tell people you were! There is always this lurking sense of embarrassment and shame related to having realized that you were standing on Mt. Stupid, and then also continued moving forwards down the far side.
*) Please note that sociopath is a deprecated synonym for psychopath.
That's not an extraordinary claim, it is a mainstream 101-level claim that anybody claiming enough subject knowledge to demand a citation should already have.
Take some responsibility for your ignorance and attempt to look shit up before you tell people it needs a citation.
Because I refuse to do that for you, I'll instead give an anecdote: In my early 20s I worked at a plywood mill, and they often wanted to increase production seasonally. They would always (always!) over-work everybody. What they would do is go from 40 hours a week to 50 the next week, then they'd add Saturday to get to 60, and then on Monday morning they'd be really happy because they were so far "ahead," and then production would be so low all week that by the end of the week we were barely getting 50% out. They'd try to increase it by pushing people harder, but then it would go down further because of how much work had to be redone. Then eventually one of the people from the head office would get mad enough to come over and find out what was going on, and force them back down to 50 hours a week. Then over 2 weeks production would climb back to 85%. They never seemed to figure out that 85% is less than 100%, though. They also never enacted any policy changes that would prevent the pattern from repeating; every time that a production increase was desired, the same pattern would play out, with reduced productivity.
Eventually they added a third shift, which prevented the ability to work more hours in a day, and waste went down a lot. The only reason they didn't go out of business was that they had a lot of warehouse space, so they could still sell more product during the seasonal peaks than the rest of the year. That helped to hide the related drops in output.
It seems to be the natural result of letting managers teach themselves their job by their bootstraps.
This is a common misunderstanding about statistics. You do NOT need a lot of samples for statistical significance, and the number of datapoints needed is usually far below what people intuitively expect.
It is a natural mistake, because people presume that "significance" as technical jargon would be even more significant than the literary word, but actually it only means "probably not a math error." It tells you nothing about if your results are "significant" in the literary sense; the vast majority of scientific results that are "statistically significant" are still worthless and misleading, because math errors are not even the main challenge to overcome! The difficulty in predicting unknown explanations for results is really the problem that experiments attempt to solve. And it takes a lot of work. Being 95% sure that you didn't have a rounding error is not exactly a result that is likely to have significant implications, even if it is statistically significant.
If they'd just rename it to Statistically Cromulent people would be better able to understand it.
One is a syndrome, the other is an effect that you don't quite understand as well as you think you do. ;)
I can haz cloud? Or I can go outside? I can haz 364? I can haz error massage?
Most of that is analysis. Just so you know.
But Whatabout Obama?
Yo, cluestick, your MAGA hat is crooked.
/. is the only place I know where you can tell a silly joke and people will start analyzing and overthinking it...
That proves it; slashdot is the only place you know!
Yeah, it seems pretty clear-cut to me; either the guy did "marry" a 12 year old child, or he didn't.
The accusation contains a considerably high level of specificity regarding what is accused. Not exactly a he-said, he-said.
I'm not sure, I was taught that the whole modern "scientific method" was based on Newton, et al., rejecting the publishers as gatekeepers and publishing through the newsletters of their local Philosophic Societies. The main thing that was supposedly holding back Natural Philosophy (what science was called then) was the gatekeepers! It worked fine for a long time. At first the "Journals" were just a type of group publishing, just a like a Philosophic Society newsletter. But once they started assigning a system of Peers as gatekeepers, the whole thing was instantly a farce; the same thing that open publishing had already replaced once!
Peer review was valuable, historically, because it was done by your peers, accomplished through open access. Naming a class of Very Important People as Peers does not in any way achieve the same thing.
I guess I agree publishing companies provided a valuable service in the past, I'm just saying, the time when that was true was pre-Newton!
So like, New Scientist magazine, but for OA?
It might be more likely to work as a free service with a donation model.
You've got some word salad there, maybe you should just keep playing with your food until it spells out whatever crazy bullshit is in your head.
You understand, little babies get born, it's been happening for millions of years. When I observe that, and you can't figure it out, and want to argue with it, you're not going to convince me that you're actually trying to communicate with the other humans. You're just playing with your word salad, and you have no idea even that the words you don't get to choose are coming from other humans!
I can haz adblock? I can haz extension? I can haz personal choose? Yaong?! Nyan?! Meow?!
That's very hand-wavy, not only am I not convinced of your conclusions, I'm not even convinced that you actually offered any logical reason for a person to even start considering what you said. Just pure assertion, and without any concept of which parts are opinion, which parts are purported facts, and which parts are generally accepted truisms.
In fact, I'm not even convinced that you understood that I was claiming that a different field, called QA, might actually be the field with the expert knowledge, and they might have a different truism than the developers have!
You don't seem to have even noticed what I was talking about, but you know you disagree. You know I'm doing it wrong, but you don't even know what I'm doing, what the use case is! Perhaps the answer is different depending on the size of the project? Perhaps it is different depending on the consequences of bugs? Perhaps it varies based on those things, and also the product lifecycle? Your insistence that there Must Be One True Answer proves that you're only considering the least-important categories of concerns, like those that web developers have.
My comment was not over my head, but perhaps your own comment was over your own head?
You made a movie reference. I bridged the gap between what happened in the movie, and what happened in the actual fucking story we're talking about. Those are my comments, not yours. If they went over somebody's head, why would you start with me? I might have even understand what I was saying!
Right, when somebody tells you that their magic process will save the world, you just assume that means it is just like seatbelts, and there is no need to find out if it is true or not.
What if it turns out that there is data suggesting that adopting a new programming process will not improve your software quality? Would adopting Virtuous processes still be the same as a seatbelt, or not?
Or are you suggesting that the value in seatbelts is in the feeling of Virtue that you get in wearing it, and that they're not actually effective?
Does calling a feature a "safety feature" actually tell me that it improves safety, or does it merely communicate your intent without telling me anything at all about the results of adopting your process?
DoubleClick finally took over the zoo.
All it says is they're considering making some rules, it doesn't actually say anything about forcing the Indian people to choose locally-owned apps.
A penny spar'd is twice got.
I was always taught that tests should be written by a QA developer instead of the application developer, because if you write your own tests you have symmetrical oversights and logical mistakes.
Also, I was taught to be suspicious of anything that wants to "drive" the development process that isn't the use case! The goal is to match the implementation to the semantics of the problem domain, not to the semantics of the testing framework.
TDD is great for web apps, since there is unlikely to be enough profit to afford QA, but why do those even need a lot of thrash?
It's called an "IDE," an Integrated Development Environment.
They're popular, but they also don't really achieve anything. The devil remains in the details.
You can make it as easy as lego, and yet as with lego, anybody can stuff a few blocks together, but building something interesting to other people is much harder. Now, build something life-sized that solves some sort of problem that some human is describing to you...
Yeah, you're confused, in the movie it was apes, in the story it was beagles.