As the other poster pointed out, there are games that update your position immediately, before getting an OK from the server. However, that's not the point in my post. The point in my post is that games frequently have a display that will tell you your latency, and I believe that this number is usually just the time it takes to send a ping from your computer to the server. So a game that would get 150ms latency in this test may only report 40 ms latency inside the game (I obviously just made up that number). In that case it is misleading to compare 150ms to 40ms, since actually those two numbers might indicate the same performance for the game since they aren't measuring the same thing.
I humbly suggest that you might want to re-read that section of the book. His point was that these children did not do exceptionally well - that they were not the kind of outliers his book was about. He then argues that if IQ was all you needed for astounding, astonishing success (=outlier), then some of these children should have been the kind of outliers he was writing about. Yet they weren't so IQ on its own isn't enough to be an outlier as defined by Malcolm Gladwell. That certainly does not imply that these children did no better on average than the average person.
All of the items listed suggest that IQ is moderately to strongly correlated with performance, but almost all of them couch that by saying the correlations are weak, or that there are other influencing factors.
So we agree that IQ does indeed correlate moderately or strongly with performance in many jobs, but not all. Right?
Full Scale IQ scores are the product of scales that evaluate Verbal and Non-Verbal working memory. Both of these scales are informed by more specific sub-scales below them. Think of it like a pyramid with IQ sitting at the top; you cannot arrive at an IQ score without assessing the many different types of intelligence. The Mathematical and logical questions are not interchangeable, both of these types (and others) are required to arrive at anything like a reliable IQ score.
I don't doubt that many IQ tests are constructed that way, and I imagine it is a fine way to structure an IQ test. That the tests you know about are structured that way does not imply that it is necessary to do it that way to measure IQ.
I myself have taken the Raven's Advanced Progressive Matrices test which contains only questions of the type "pick the missing shape". I didn't notice any mathematical themes in the questions. You can read about it here.
I maintain that if you give people a huge battery of diverse questions about all manner of things, you'll have a good measurement of IQ. I have this straight out of the g factor, though I don't have the book handy to give a quote. There is no need to measure sub-components - in fact you need more questions to be able to do so.
I know I already quoted this bit but I want to hit on it again. One of the classic criticism of any form of standardized testing are the biases of the test writers. If I know nothing about duck hunting, and you give me an IQ test that consists largely of questions about duck hunting, I am going to score very low. This will not be a valid measure of my IQ. This is why determining an accurate IQ score is so difficult.
I didn't mean to suggest that a test just about duck hunting would measure IQ well. What you are pointing out is exactly the problem with asking a duck hunting question in an IQ test. Your ability to answer the question does correlate with your IQ, but it does so weakly because lots of people just happen to be uninterested in duck hunting and so they have little chance to know the right answer regardless of their IQ. So if you are going to ask questions like that, you are going to have to ask hundreds or thousands of questions about all kinds of things (certainly not just about duck hunting) to get a reliable indicator. The point is that IQ tests look the way they do not because IQ is just about logical puzzles or something like that (I know that you have not stated that it is), but because logical puzzles happen to be the fastest way to get a reliable measurement. IQ tests could look completely different from the IQ tests that we know, it's just that then they would consist of a much larger number of questions that would take longer to answer and grade, and they would probably also be more culturally dependent.
I'm not saying Google's test is not IQ loaded, on the contrary, it's very loaded. Maybe google will let someone test it, but still...
But the main problem is that, to get to the point in the interview where only IQ matters you need to know A LOT of things. Think of it like doing a test in a foreign language.
I'd say that having a high IQ also increases your ability to get to know all those things, but that's a detail. I agree with what you are saying here.
Costs? The video is already being compressed online just for you. Eye tracking software is already available. It only requires people to have a webcam, and you don't have to use it if you don't have a webcam. The only problem is the 150ms latency, so the quality spot wouldn't keep up if you are moving your eyes continually.
[IQ] is an arbitrary number that is generated by an individuals performance on a series of abstract tests.
Hard puzzles as found in IQ tests are a very quick and convenient way to measure IQ, and that is why they are used. You could have IQ tests with not a single mathematical or logical question in it, asking questions about botany and duck hunting and all kinds of different things. In the end the test would measure the same thing and have the same results, it's just that it would require many more questions to do it that way and so that's not the way it's done.
If IQ is an arbitrary then why does it correlate positively with so many desirable outcomes? I'll give you an arbitrary number: 23. I guarantee you that IQ is going to be more useful in predicting your performance than my arbitrary number is.
That book is primarily based on interesting anecdotes that don't aggregate to data. Doesn't mean he is wrong, but even if he is right, his contention is that success requires luck. His contention is not that IQ is irrelevant. On top of that, his book is about outliers, people whose outcomes are not the norm. His book is not about the great, vast majority of people.
Citation needed on Google tests being way better than just an IQ test
Maybe because one has to do with skills related to the job and the other doesn't. Do you even know what kind of questions google asks? Well, I do.
IQ tests can sometimes outperform subject-specific tests even for determining future performance in that specific subject. This occurs especially when everyone tested is known to already posses the basic knowledge of a field.
It's surprising until you think about it for a bit. People can increase their knowledge if they have a high IQ, but people cannot increase their IQ no matter how much they know of their subject.
and yet IQ tests still predict performance very well in many jobs. It's both fantastic
Depends on the job. Get out of the lab and onto the real world.
Yes, it is generally the case that each job has a certain level of IQ beyond which further IQ doesn't help you very much. There is a limit to how much sweeping the street can be improved with brainpower. I don't have a number for programming, but it is a complicated technical field where productivity differs wildly among individuals. That's exactly the kind of thing that lends itself to IQ mattering into very high levels. It's true that high IQ doesn't help if the guy then doesn't show up for work at all or spends all his time at work playing World of Warcraft, but that is true of anyone so it isn't something specific to people with a high IQ.
Telling me to get onto the real world does nothing for your argument. I could tell you that knowledge of IQ would probably increase your ability to talk about it, but that would equally do nothing for my argument.
Google had every reason to use IQ for hiring and still it doesn't. It's about 70% 'knowing stuff'.
You seem to think that a question about "knowing stuff" cannot have an IQ loading. It doesn't work that way and the more so the harder the questions are. The way to know if the Google hiring process is highly IQ loaded is to measure people's IQ and compare that to their interview score. If you think IQ isn't relevant I think you'd be surprised.
To display a frame you have to wait for that frame to have been transferred to you. We have to know the size of a frame to know the impact, but I'm guessing it's non-trivial. In your comment you seemed to only be accounting for the communication of key presses from the client to the server. I'm pointing out the much larger amount of information flowing the other way, and the transfer of this information was included in their latency measurements.
Basing emplyment on IQ is pointless as it doesn't actually predict "real-world" performance.
Citation needed in the same way that citation is needed for water not being wet.
This is similar to college only accepting students with a score in the top 1% on the ACT/SAT - they can do well on a test, but that doesn't mean they are a good student.
That's true, though it also does not mean that they won't do well. It's a correlation - it doesn't automatically imply any specific outcome, just makes it more likely.
What do you know? Valleywag got everything wrong. Google is hiring, not laying off. Also, our interview scores actually correlate very well with on-the-job performance. Peter Seibel asked me if there was anything counterintuitive about the process and I said that people who got one low score but were hired anyway did well on-the-job. To me, that means the interview process is doing very well, not that it is broken. It means that we don't let one bad interview blackball a candidate. We'll keep interviewing, keep hiring, and keep analyzing the results to improve the process. And I guess Valleywag will keep doing what they do... - Peter Norvig from Bookmarklet
It's true that it's not all you need to do well. Citation needed on it being almost useless, in the same way that citation is needed on water not being wet.
Google tests are (way) better than IQ, but guess what Google found out: the best performers are the ones who have the lowest scores on their interviews.
The best performers are those that were hired in spite of having a low score in one interview out of several. These are people that are so impressive for some reason or other that even a low score in an interview does not rule them out. Citation needed on Google tests being way better than just an IQ test - I only know that they are more laborious, not that they outperform 100 years of research into IQ. If they do I expect it's because they include either an actual IQ test or an IQ test by proxy such as riddles or hard subject-specific questions you can't just memorize ahead of time. In any case, citation needed.
IQ is not concerned with
- the candidate knows about the job
- the candidate has good (enough) people skills
- the candidate showers, shaves, etc
... and yet IQ tests still predict performance very well in many jobs. It's both fantastic and fantastically politically unacceptable.
150ms doesn't mean 150ms in this case
on
OnLive Latency Tested
·
· Score: 2, Insightful
This number of 150 ms latency may be true but it very likely does not mean at all what you think it means - it is not to be compared to the network latency that multiplayer games often report. The latency you normally see in a game is just the network latency - the amount of time it takes for a small packet to go from your computer to the server. The 150 ms latency includes the time it takes for a packet to go to the server, for the game to process that packet, and then send a frame of video back to you. So the server has registered your action long before the 150 ms are up. Also, normal lag does not include the time it takes for the game to process your command, which can be even more lengthy than your network latency, but that time is included in the 150 ms. Unless you are aware of these things, then the 150 ms number is completely meaningless to you and if you compare it to the latency number from some game you've played before then you are doing it wrong.
What they should have done to get a meaningful comparison is to do the exact camera setup thing they did, but also do it for a game running locally and then over the net. Only then can you meaningfully compare the numbers and know that you got it right.
Actually I'm guessing that video frames are a lot larger than 5kb. If they are e.g. 1 mb then the time to transfer may even be more than the latency of the connection itself, depending on your bandwidth. So the latency as measured by the time from pressing a key and the man on your screen doing something is most certainly impacted by bandwidth in many cases. This effect is not limited to when the game can't finish sending a frame before the next one is ready, because even when that doesn't happen you still have to wait to receive the frame before it can be displayed.
Bandwidth can reduce latency all other things being equal. If you need to send a packet of 5kb then the latency of that packet will be lots of things that are independent of the bandwidth, but it also does include the time it takes to transfer the data in the packet itself over the wire, and higher bandwidth reduces that time. Not that it is going to be a lot of time for a 5 kb packet, but there can indeed be an impact of bandwidth to latency.
The only thing i'm not sure I like at the moment is the some of the minor artifacting you'll see due to the video compression.
What would be cool is if they could track your eyes using a webcam and increase the quality at the part of the screen you are looking at. Won't help if your eyes are twitching all over the place, but if there is anything you want to see in full quality, just look at it during the latency of those 230 or so milliseconds and the artifacts will disappear where you are looking.
We are not talking about increasing the resolution of the designer's screen. We are talking about increasing the resolution of the image. That won't increase the detail in that image, but it will increase the potential for adding smaller details. The designer will likely need to redo the image completely in the new resolution.
The designer zoomed into the level of zoom that the designer needed. That just randomly happened to be the pixel level, and that was precisely the problem he wanted help with.
Are you honestly suggesting that, by having more pixels on the screen, the picture won't be blocky when they zoom to the pixel level?
No, I'm suggesting that by increasing the resolution the designer will be able to draw smaller shapes. This solves the designer's problem of "I need to be able to get a smaller shape in here but it's all too blocky."
Also, a standardized connector would let third parties come in and start making accessories and replacement bricks for a lot less than the highway robbery prices that the brand names charge.
That's exactly the problem. The brand names are both in charge of choosing the connector design and of earning the income from selling their proprietary power bricks.
Undertaking PhD studies is about learning how to do research, it's not about already knowing how to do everything the right way and then just doing that for three years.
This is to be read as "it's not about already knowing what to do and then doing that for three years." It's not intended to be a statement about how you should be not knowing what to do for 3 years - even if it can take quite a while to figure it all out.
Slashdot Mirror
As the other poster pointed out, there are games that update your position immediately, before getting an OK from the server. However, that's not the point in my post. The point in my post is that games frequently have a display that will tell you your latency, and I believe that this number is usually just the time it takes to send a ping from your computer to the server. So a game that would get 150ms latency in this test may only report 40 ms latency inside the game (I obviously just made up that number). In that case it is misleading to compare 150ms to 40ms, since actually those two numbers might indicate the same performance for the game since they aren't measuring the same thing.
I humbly suggest that you might want to re-read that section of the book. His point was that these children did not do exceptionally well - that they were not the kind of outliers his book was about. He then argues that if IQ was all you needed for astounding, astonishing success (=outlier), then some of these children should have been the kind of outliers he was writing about. Yet they weren't so IQ on its own isn't enough to be an outlier as defined by Malcolm Gladwell. That certainly does not imply that these children did no better on average than the average person.
All of the items listed suggest that IQ is moderately to strongly correlated with performance, but almost all of them couch that by saying the correlations are weak, or that there are other influencing factors.
So we agree that IQ does indeed correlate moderately or strongly with performance in many jobs, but not all. Right?
Full Scale IQ scores are the product of scales that evaluate Verbal and Non-Verbal working memory. Both of these scales are informed by more specific sub-scales below them. Think of it like a pyramid with IQ sitting at the top; you cannot arrive at an IQ score without assessing the many different types of intelligence. The Mathematical and logical questions are not interchangeable, both of these types (and others) are required to arrive at anything like a reliable IQ score.
I don't doubt that many IQ tests are constructed that way, and I imagine it is a fine way to structure an IQ test. That the tests you know about are structured that way does not imply that it is necessary to do it that way to measure IQ.
I myself have taken the Raven's Advanced Progressive Matrices test which contains only questions of the type "pick the missing shape". I didn't notice any mathematical themes in the questions. You can read about it here.
I maintain that if you give people a huge battery of diverse questions about all manner of things, you'll have a good measurement of IQ. I have this straight out of the g factor, though I don't have the book handy to give a quote. There is no need to measure sub-components - in fact you need more questions to be able to do so.
I know I already quoted this bit but I want to hit on it again. One of the classic criticism of any form of standardized testing are the biases of the test writers. If I know nothing about duck hunting, and you give me an IQ test that consists largely of questions about duck hunting, I am going to score very low. This will not be a valid measure of my IQ. This is why determining an accurate IQ score is so difficult.
I didn't mean to suggest that a test just about duck hunting would measure IQ well. What you are pointing out is exactly the problem with asking a duck hunting question in an IQ test. Your ability to answer the question does correlate with your IQ, but it does so weakly because lots of people just happen to be uninterested in duck hunting and so they have little chance to know the right answer regardless of their IQ. So if you are going to ask questions like that, you are going to have to ask hundreds or thousands of questions about all kinds of things (certainly not just about duck hunting) to get a reliable indicator. The point is that IQ tests look the way they do not because IQ is just about logical puzzles or something like that (I know that you have not stated that it is), but because logical puzzles happen to be the fastest way to get a reliable measurement. IQ tests could look completely different from the IQ tests that we know, it's just that then they would consist of a much larger number of questions that would take longer to answer and grade, and they would probably also be more culturally dependent.
I'm not saying Google's test is not IQ loaded, on the contrary, it's very loaded. Maybe google will let someone test it, but still...
But the main problem is that, to get to the point in the interview where only IQ matters you need to know A LOT of things. Think of it like doing a test in a foreign language.
I'd say that having a high IQ also increases your ability to get to know all those things, but that's a detail. I agree with what you are saying here.
Costs? The video is already being compressed online just for you. Eye tracking software is already available. It only requires people to have a webcam, and you don't have to use it if you don't have a webcam. The only problem is the 150ms latency, so the quality spot wouldn't keep up if you are moving your eyes continually.
[IQ] is an arbitrary number that is generated by an individuals performance on a series of abstract tests.
Hard puzzles as found in IQ tests are a very quick and convenient way to measure IQ, and that is why they are used. You could have IQ tests with not a single mathematical or logical question in it, asking questions about botany and duck hunting and all kinds of different things. In the end the test would measure the same thing and have the same results, it's just that it would require many more questions to do it that way and so that's not the way it's done.
If IQ is an arbitrary then why does it correlate positively with so many desirable outcomes? I'll give you an arbitrary number: 23. I guarantee you that IQ is going to be more useful in predicting your performance than my arbitrary number is.
The smartest man in the world works as a bouncer.
The oldest person to ever live smoked till she was 115.
Citation needed on Google tests being way better than just an IQ test
Maybe because one has to do with skills related to the job and the other doesn't. Do you even know what kind of questions google asks? Well, I do.
IQ tests can sometimes outperform subject-specific tests even for determining future performance in that specific subject. This occurs especially when everyone tested is known to already posses the basic knowledge of a field.
It's surprising until you think about it for a bit. People can increase their knowledge if they have a high IQ, but people cannot increase their IQ no matter how much they know of their subject.
and yet IQ tests still predict performance very well in many jobs. It's both fantastic
Depends on the job. Get out of the lab and onto the real world.
Yes, it is generally the case that each job has a certain level of IQ beyond which further IQ doesn't help you very much. There is a limit to how much sweeping the street can be improved with brainpower. I don't have a number for programming, but it is a complicated technical field where productivity differs wildly among individuals. That's exactly the kind of thing that lends itself to IQ mattering into very high levels. It's true that high IQ doesn't help if the guy then doesn't show up for work at all or spends all his time at work playing World of Warcraft, but that is true of anyone so it isn't something specific to people with a high IQ.
Telling me to get onto the real world does nothing for your argument. I could tell you that knowledge of IQ would probably increase your ability to talk about it, but that would equally do nothing for my argument.
Google had every reason to use IQ for hiring and still it doesn't. It's about 70% 'knowing stuff'.
You seem to think that a question about "knowing stuff" cannot have an IQ loading. It doesn't work that way and the more so the harder the questions are. The way to know if the Google hiring process is highly IQ loaded is to measure people's IQ and compare that to their interview score. If you think IQ isn't relevant I think you'd be surprised.
To display a frame you have to wait for that frame to have been transferred to you. We have to know the size of a frame to know the impact, but I'm guessing it's non-trivial. In your comment you seemed to only be accounting for the communication of key presses from the client to the server. I'm pointing out the much larger amount of information flowing the other way, and the transfer of this information was included in their latency measurements.
IQ is an interesting measure but research into correlation with ability to perform at certain jobs is appropriately lacking.
What makes you believe this? IQ research has been ongoing for 100 years.
Basing emplyment on IQ is pointless as it doesn't actually predict "real-world" performance.
Citation needed in the same way that citation is needed for water not being wet.
This is similar to college only accepting students with a score in the top 1% on the ACT/SAT - they can do well on a test, but that doesn't mean they are a good student.
That's true, though it also does not mean that they won't do well. It's a correlation - it doesn't automatically imply any specific outcome, just makes it more likely.
What do you know? Valleywag got everything wrong. Google is hiring, not laying off. Also, our interview scores actually correlate very well with on-the-job performance. Peter Seibel asked me if there was anything counterintuitive about the process and I said that people who got one low score but were hired anyway did well on-the-job. To me, that means the interview process is doing very well, not that it is broken. It means that we don't let one bad interview blackball a candidate. We'll keep interviewing, keep hiring, and keep analyzing the results to improve the process. And I guess Valleywag will keep doing what they do... - Peter Norvig from Bookmarklet
http://friendfeed.com/peternorvig/7a110005/google-broken-hiring-process-gawker?embed=1
IQ is highly overrated
In practice, it's almost useless...
It's true that it's not all you need to do well. Citation needed on it being almost useless, in the same way that citation is needed on water not being wet.
Google tests are (way) better than IQ, but guess what Google found out: the best performers are the ones who have the lowest scores on their interviews.
The best performers are those that were hired in spite of having a low score in one interview out of several. These are people that are so impressive for some reason or other that even a low score in an interview does not rule them out. Citation needed on Google tests being way better than just an IQ test - I only know that they are more laborious, not that they outperform 100 years of research into IQ. If they do I expect it's because they include either an actual IQ test or an IQ test by proxy such as riddles or hard subject-specific questions you can't just memorize ahead of time. In any case, citation needed.
IQ is not concerned with - the candidate knows about the job - the candidate has good (enough) people skills - the candidate showers, shaves, etc
... and yet IQ tests still predict performance very well in many jobs. It's both fantastic and fantastically politically unacceptable.
If you are up in arms about IQ, then just wait till you read about the general fitness factor. This is the first link I found on google: http://www.eric.ed.gov/ERICWebPortal/search/detailmini.jsp?_nfpb=true&_&ERICExtSearch_SearchValue_0=EJ698164&ERICExtSearch_SearchType_0=no&accno=EJ698164
This number of 150 ms latency may be true but it very likely does not mean at all what you think it means - it is not to be compared to the network latency that multiplayer games often report. The latency you normally see in a game is just the network latency - the amount of time it takes for a small packet to go from your computer to the server. The 150 ms latency includes the time it takes for a packet to go to the server, for the game to process that packet, and then send a frame of video back to you. So the server has registered your action long before the 150 ms are up. Also, normal lag does not include the time it takes for the game to process your command, which can be even more lengthy than your network latency, but that time is included in the 150 ms. Unless you are aware of these things, then the 150 ms number is completely meaningless to you and if you compare it to the latency number from some game you've played before then you are doing it wrong.
What they should have done to get a meaningful comparison is to do the exact camera setup thing they did, but also do it for a game running locally and then over the net. Only then can you meaningfully compare the numbers and know that you got it right.
They also have to send you back frames of video.
Actually I'm guessing that video frames are a lot larger than 5kb. If they are e.g. 1 mb then the time to transfer may even be more than the latency of the connection itself, depending on your bandwidth. So the latency as measured by the time from pressing a key and the man on your screen doing something is most certainly impacted by bandwidth in many cases. This effect is not limited to when the game can't finish sending a frame before the next one is ready, because even when that doesn't happen you still have to wait to receive the frame before it can be displayed.
Bandwidth can reduce latency all other things being equal. If you need to send a packet of 5kb then the latency of that packet will be lots of things that are independent of the bandwidth, but it also does include the time it takes to transfer the data in the packet itself over the wire, and higher bandwidth reduces that time. Not that it is going to be a lot of time for a 5 kb packet, but there can indeed be an impact of bandwidth to latency.
The only thing i'm not sure I like at the moment is the some of the minor artifacting you'll see due to the video compression.
What would be cool is if they could track your eyes using a webcam and increase the quality at the part of the screen you are looking at. Won't help if your eyes are twitching all over the place, but if there is anything you want to see in full quality, just look at it during the latency of those 230 or so milliseconds and the artifacts will disappear where you are looking.
We are not talking about increasing the resolution of the designer's screen. We are talking about increasing the resolution of the image. That won't increase the detail in that image, but it will increase the potential for adding smaller details. The designer will likely need to redo the image completely in the new resolution.
The designer zoomed into the level of zoom that the designer needed. That just randomly happened to be the pixel level, and that was precisely the problem he wanted help with.
Are you honestly suggesting that, by having more pixels on the screen, the picture won't be blocky when they zoom to the pixel level?
No, I'm suggesting that by increasing the resolution the designer will be able to draw smaller shapes. This solves the designer's problem of "I need to be able to get a smaller shape in here but it's all too blocky."
Sorry to poop on the joke here, but it's a perfectly reasonable request and you do it by increasing the resolution.
I personally think internet anonymity is a good thing. It forces people to attack each other's arguments rather than resorting to ad hominems,
That's not been my experience.
Also, a standardized connector would let third parties come in and start making accessories and replacement bricks for a lot less than the highway robbery prices that the brand names charge.
That's exactly the problem. The brand names are both in charge of choosing the connector design and of earning the income from selling their proprietary power bricks.
Undertaking PhD studies is about learning how to do research, it's not about already knowing how to do everything the right way and then just doing that for three years.
This is to be read as "it's not about already knowing what to do and then doing that for three years." It's not intended to be a statement about how you should be not knowing what to do for 3 years - even if it can take quite a while to figure it all out.