>>Repeat 10 times and check which number is more stable (that is, has the smaller variance).
Yeah, it'd probably be less susceptible to interference from load or other confounding factors. But sometimes we do need to see those effects in the timings... in a large parallel system, for example, if one computer occasionally bugs out and runs 10x as slow, and you have a barrier in your code, meaning everyone has to wait on the slowpoke, we want to capture that, so we can deal with the problem, whereas a minimum timing would ignore it.
If ever, you mean "in the next couple years", then yeah, maybe.
Pedantic Note: Except we've been getting new Xbox hardware. Look at the differences in the different revs of the 360. They've migrated both the CPU and GPU from 90 to 65nm, and reduced power consumption by about 50W.
No game generation has ever been "for forever". It'd be kind of silly to expect the 360 to be the last Xbox they ever make.
Do what I do and make friends with a Microsoft employee. They get copies for next to nothing from the company store. Failing that, go to a technical talk that Microsoft gives in your area. They usually give away quite a bit of free stuff. I got my first copy of Visual Studio that way back in college.
>>A material wich can absorb all the toxens out of the water and when it is full all we need to do is grind it back into little pieces and flush it down the toilet and all our problems are solve.
Narrator: Thus solving the problem for all time
Suzy: But...
Narrator: FOR ALL TIME.
Actually, my problem with this is the trade name. OBSORB sounds like OXYCLEAN or the SHAMWOW or all the other staples of late night TV.
So what is the energy that is moving about / creating the asymmetrical surfaces? In a vaccuum, gravity obviously exists, even if there is no energy being exchanged between two bodies.
>>The vendors who don't rent floor space are capitalizing on the attendees, who are their because of the efforts of CES and those vendors who rented floor space. Before anyone makes the analogy - this is NOT akin to filesharing or the alledged piracy of music or video. This is more akin to pirating someones' signal and replacing their content with your own.
That's a horrible analogy, even for Slashdot.
Everything these guys were doing complied with the law, and the pre-stated hotel policies (and longstanding tradition - NDA-only products are usually ONLY shown in hotel rooms).
Or as Shakespeare would say, Who am I to CES, or CES to me? Fucking cry me a river that they are capitalizing on an event. Are you going to complain that restaurants in the area are having increased sales as well?
>>Or even just forget about things that have already been written -- commission Doctorow or Stross (or someone) to create a TV miniseries based on new SF material.
I'm with you on that, except I can't stand either Doctorow or Charlie Strauss. Way too gee whiz / whiz bang, combined with horrendous characterization. And the whiz bang doesn't even make sense, either.
I'd prefer seeing more of what HBO is doing right now, making GRR Martin's Song of Fire and Ice into a TV show. I even have some hopes that they won't totally butcher it, since GRRM is involved in the production.
Some studios have been looking at Brandon Sanderson's Mistborn and Robert Jordan's Wheel of Time series, and that could be pretty cool.
But seriously, folks, learn from the horrible Star Trek remake and just let old shows die.
Eh, I have a Droid and I hate the physical keyboard. The keys are just too tiny - the on-screen keyboard keys are actually easier to hit.
I have an iPod Touch, too, and I'd agree that the iPod's screen is better. Just in terms of overall feel - the droid is actually more accurate when using the on-screen keyboard, but it's way too eager to click instead of scroll, meaning that when you're paging around through your contacts you'll accidentally dial people, and when dragging around inside of the browser, you'll accidentally follow links, etc.
>>If someone has some form of autism making him extremely good at something (music, math, extreme memory, collecting stamps,...), would this medicine affect his ability to do that?
In the study of "idiot" savants, studies have shown that curing their extreme social inability also "cures" their ability to be exceptional at math, or whatever.
>>He has got a point that Computer Science graduates do value logic and reason (or less charitably bullshit) over evidence and observation.
Do they? In my experience, you tend to both have a logic model for how your performance should behave, and you test extensively to see how it works out in practice. If you just have observation and no theory, you'll never be able to understand how sometimes when you add more CPUs to a system, you get better than linear (superlinear) speedup. It's supposed to be impossible!
And likewise, if all you have is theory, then all sorts of things will come back to bite you in the ass later, as ignoring constant time factors is fine for theory, but can be really really bad in practice.
Every time I'd finish a program of bloc of code, I'd run it hundreds or thousands of times with different parameters and configurations. Did I pull the number of runs out of my ass (as the article complains about)? Yes. Would the author be able to do better? No. He's assuming he knows a priori what all the timing measurements are, which is exactly what we're trying to figure out!
You can't use the "stddev" (which is bad stats anyway for non gaussian curves) to estimate the number of samples you need when you don't know this number yet.
>>Spiking vs non-spiking is something pretty easy to see when you glance at the data.
Yeah, in fact, the way that he presents it is bad statistics. =)
If the problem is that one out of 1000 queries is taking a minute to return instead of 0.1 seconds, then using the std deviation to describe the problem is nonsense. It is not a Gaussian distribution!
But of course someone who "has spent his life studying statistics and even R language" would know that, right?:p
Instead, as you point out, any programmer who did the same testing would see that one out of a thousand queries were taking far too long, and come to the same conclusion as him, without making the ghost of Gauss cry.
I just realized I was a nerd. I looked at the table of contents and closed it down, then realized I hadn't even looked at the short skirt-wearing protagonist.
Sigh...
But to answer the article's point, elementary statistics are very easy. Advanced statistics are very hard. It's kind of like how people think "knowing the difference between circles and squares" is geometry and so analytical geometry must be just more of the same, right? It's quite possible the programmers think they know statistics because they know they're vaguely supposed to do a run multiple times, and maybe average the results or something.
It's also possible the author of the article is a know-it-all douchebag who tries to solve problems with overwrought solutions.
From TFA: "Zed: Fuck! Fuck! I have eyes! You do not! See!? No?! Exactly! Because you can't fucking see because you have no fucking eyes! Arrggh!"
>>I mean seriously, passwords are among the weakest chain when it comes to security today
No, the weakest chain in security today is not the password. I recently found out that the idiot at Verizon typed in my password wrong when I bought my droid, so even though it'd autologin on my phone, I couldn't access my account (I was wanting to pull up the calendar) on my PC. There's no way of resetting the password without not logging in for a while, and my phone (of course), was always logging in, so it mean for going a number of days (weeks?) without having a phone, just to be able to reset my password.
After clicking around for a while, I found the weakest link: the password recovery password. Favorite food? Well... I won't tell you. It's a secret! Only I love these delicious cheese and pepperoni covered delicacies from Italy.
Why on earth we demand ridiculous password strengths on one hand but have the most mind-bogglingly easy to guess password recovery passwords on the other. They're easily the weakest link. IIRC, Palin's email got compromised by a similar means.
>>Pah! Trespasser was doing that back in 90's. Call me when there's a real revolution.
Heh heh, yeah that game was great. Watching your gun get caught in a chain link fence and then go sproinging hundreds of yards away was... well, "fun" is not exactly the right word for it.
>>I don't get the bit about the "six axes". I thought we had only three in meatspace.
Well, I only have one axe in meatspace.
I'm very excited about this controller though - I can't wait to see how inaccurate people will be in games when they actually have to aim, instead of just clicking on a point with their "sniper rifle" and pretending that's skill.
Being able to shoot yourself in the head in a FPS? That's revolutionary.
>>I wonder how you balance this with the fact that US courts don't care whether or not the people standing in front of them have been illegally abducted off foreign streets by intelligence agents, or whether the evidence against them was obtained by illegal means such as torture.
Which is precisely my point. It does bother me, but it doesn't seem to bother our legal system.
How could you say you care about illegal means used in law enforcement in one sentence and then say you don't worry about it happening in another?
If they tried to pull a Bush and claim that they couldn't reveal the source of information being used to convict me because it came from a privileged source, then it's a bit too late, isn't it. And the Bush administration did do these sorts of hijinx, ignoring the fact that you're always supposed to confront your accuser, as it were.
>>And INTERPOL isn't part of your law enforcement efforts either, except insomuch as they cooperate with national agencies that ARE law enforcement bodies and who must act within the law
So you're fine with some part of that chain not being one "who must act within the law"? Just curious.
I don't have anything against INTERPOL per se, and I in fact think they're a fine institution. However, no part of the legal system used for convictions should be outside the law. It makes sense, does it not?
Right, and that's why we saw things like the Cult of Reason, replacing the names of the week with atheist names, etc., because they were so few and uninfluential?
I've studied the French Revolution. I'd recommend you do the same.
Obviously it was a heterogeneous mixture, and different factions in the revolution had different takes on it. But it doesn't change the facts.
However, there's really no reason why we shouldn't be looking into solutions (geoengineering foremost among them) and switching to nuclear. There's a variety of Really Good Reasons why we should be on nuclear power, AGW just one of them.
>>From what I have read its around 45-60 hours for most people, decent value for the money IMHO
It took me about 80 to 90 in-game hours to beat it, with about 120 total spent in the game (I restarted and played through every pathway twice so I could see all the dialogue options and combats). Hours isn't an especially good way of measuring how fun a game is, and it did manage to keep my interest through the first three quarters of the game. But by the end, it was so easy that the combats went from being tactically interesting to trivially easy. (Two mages loaded up on crowd control spells will do that.) But the end of the game went by fast enough that it wasn't too bad.
I even decided to challenge myself by not using any of the army units you recruit throughout the game, and the final set of battles still were pretty easy.
I'm not sure if I'll get the expansion. They made similar claims "hours of extra gameplay" for the existing DLC, but both the Stone Prisoner and the Warden's Keep are less than an hour of gameplay each. A bit more if you play through all the different paths, and maybe a lot more if you want to solve the puzzle in the Stone Prisoner, but even still, not a lot of additional gameplay for the cost involved.
I'm holding off on uninstalling it because I want to see what sort of player generated campaigns people come up with. I sort of feel bad for the people that bought it for the console.
>>Oil and Coal use is perfectly fine if there aren't any environmental downsides to them, so there is absolutely no need to push for seeking replacements that aren't going to be implemented until they naturally would be anyway.
Besides, you know, the whole "middle east" dynamic going on?
Nuclear is a perfectly acceptable alternative to coal, but "it's scary" so we don't build nuclear plants any more. AGW issues aside, coal kills thousands of people every year. (And dumps lots of radiation into the atmosphere, but people never talk about that either.)
>>Recycling is another canard.
Agreed. What actually bothers me is that even with things like CRV in place, city governments can't run recycling programs without incurring a loss. In my town here, they've enrolled all apartment complexes in a "voluntary" recycling program (that they can't opt out of), put in recycling bins, and then charge them hundreds of dollars each month for the privilege of being in the program.
This is a town that has a serious problem with people going through trash looking for recycleables, in organized teams. Which means that obviously these people are turning a profit at it. So they outlawed these people. It boggles the mind.
Slashdot Mirror
>>Repeat 10 times and check which number is more stable (that is, has the smaller variance).
Yeah, it'd probably be less susceptible to interference from load or other confounding factors. But sometimes we do need to see those effects in the timings... in a large parallel system, for example, if one computer occasionally bugs out and runs 10x as slow, and you have a barrier in your code, meaning everyone has to wait on the slowpoke, we want to capture that, so we can deal with the problem, whereas a minimum timing would ignore it.
>>New Xbox hardware isn't going to happen. Ever.
If ever, you mean "in the next couple years", then yeah, maybe.
Pedantic Note: Except we've been getting new Xbox hardware. Look at the differences in the different revs of the 360. They've migrated both the CPU and GPU from 90 to 65nm, and reduced power consumption by about 50W.
No game generation has ever been "for forever". It'd be kind of silly to expect the 360 to be the last Xbox they ever make.
>>Windows 7, SP0, is actually pretty darned good - especially compared to that steaming pile of mediocrity (Vista) they put out last time.
Hey, have you heard about Microsoft Mojave?
It'll be coming out soon.
Do what I do and make friends with a Microsoft employee. They get copies for next to nothing from the company store. Failing that, go to a technical talk that Microsoft gives in your area. They usually give away quite a bit of free stuff. I got my first copy of Visual Studio that way back in college.
>>A material wich can absorb all the toxens out of the water and when it is full all we need to do is grind it back into little pieces and flush it down the toilet and all our problems are solve.
Narrator: Thus solving the problem for all time
Suzy: But...
Narrator: FOR ALL TIME.
Actually, my problem with this is the trade name. OBSORB sounds like OXYCLEAN or the SHAMWOW or all the other staples of late night TV.
I've just been reading the comments to see what the implications of this would be, if true.
So far, been sort of disappointed.
So what is the energy that is moving about / creating the asymmetrical surfaces? In a vaccuum, gravity obviously exists, even if there is no energy being exchanged between two bodies.
Right?
>>The vendors who don't rent floor space are capitalizing on the attendees, who are their because of the efforts of CES and those vendors who rented floor space. Before anyone makes the analogy - this is NOT akin to filesharing or the alledged piracy of music or video. This is more akin to pirating someones' signal and replacing their content with your own.
That's a horrible analogy, even for Slashdot.
Everything these guys were doing complied with the law, and the pre-stated hotel policies (and longstanding tradition - NDA-only products are usually ONLY shown in hotel rooms).
Or as Shakespeare would say, Who am I to CES, or CES to me? Fucking cry me a river that they are capitalizing on an event. Are you going to complain that restaurants in the area are having increased sales as well?
>>Or even just forget about things that have already been written -- commission Doctorow or Stross (or someone) to create a TV miniseries based on new SF material.
I'm with you on that, except I can't stand either Doctorow or Charlie Strauss. Way too gee whiz / whiz bang, combined with horrendous characterization. And the whiz bang doesn't even make sense, either.
I'd prefer seeing more of what HBO is doing right now, making GRR Martin's Song of Fire and Ice into a TV show. I even have some hopes that they won't totally butcher it, since GRRM is involved in the production.
Some studios have been looking at Brandon Sanderson's Mistborn and Robert Jordan's Wheel of Time series, and that could be pretty cool.
But seriously, folks, learn from the horrible Star Trek remake and just let old shows die.
Eh, I have a Droid and I hate the physical keyboard. The keys are just too tiny - the on-screen keyboard keys are actually easier to hit.
I have an iPod Touch, too, and I'd agree that the iPod's screen is better. Just in terms of overall feel - the droid is actually more accurate when using the on-screen keyboard, but it's way too eager to click instead of scroll, meaning that when you're paging around through your contacts you'll accidentally dial people, and when dragging around inside of the browser, you'll accidentally follow links, etc.
It's annoying, but I like the Droid anyway.
>>If someone has some form of autism making him extremely good at something (music, math, extreme memory, collecting stamps, ...), would this medicine affect his ability to do that?
In the study of "idiot" savants, studies have shown that curing their extreme social inability also "cures" their ability to be exceptional at math, or whatever.
>>In layman's terms a series of tubes is actually appropriate.
Yeah, it's a good enough analogy.
The problem was him "getting an internet". Even my 80+ year old grandfather knew better than that.
Well, he started a computer club at his retirement home, but still.
>>He has got a point that Computer Science graduates do value logic and reason (or less charitably bullshit) over evidence and observation.
Do they? In my experience, you tend to both have a logic model for how your performance should behave, and you test extensively to see how it works out in practice. If you just have observation and no theory, you'll never be able to understand how sometimes when you add more CPUs to a system, you get better than linear (superlinear) speedup. It's supposed to be impossible!
And likewise, if all you have is theory, then all sorts of things will come back to bite you in the ass later, as ignoring constant time factors is fine for theory, but can be really really bad in practice.
Every time I'd finish a program of bloc of code, I'd run it hundreds or thousands of times with different parameters and configurations. Did I pull the number of runs out of my ass (as the article complains about)? Yes. Would the author be able to do better? No. He's assuming he knows a priori what all the timing measurements are, which is exactly what we're trying to figure out!
You can't use the "stddev" (which is bad stats anyway for non gaussian curves) to estimate the number of samples you need when you don't know this number yet.
>>Spiking vs non-spiking is something pretty easy to see when you glance at the data.
Yeah, in fact, the way that he presents it is bad statistics. =)
If the problem is that one out of 1000 queries is taking a minute to return instead of 0.1 seconds, then using the std deviation to describe the problem is nonsense. It is not a Gaussian distribution!
But of course someone who "has spent his life studying statistics and even R language" would know that, right? :p
Instead, as you point out, any programmer who did the same testing would see that one out of a thousand queries were taking far too long, and come to the same conclusion as him, without making the ghost of Gauss cry.
A manga statistics book, eh?
I just realized I was a nerd. I looked at the table of contents and closed it down, then realized I hadn't even looked at the short skirt-wearing protagonist.
Sigh...
But to answer the article's point, elementary statistics are very easy. Advanced statistics are very hard. It's kind of like how people think "knowing the difference between circles and squares" is geometry and so analytical geometry must be just more of the same, right? It's quite possible the programmers think they know statistics because they know they're vaguely supposed to do a run multiple times, and maybe average the results or something.
It's also possible the author of the article is a know-it-all douchebag who tries to solve problems with overwrought solutions.
From TFA: "Zed: Fuck! Fuck! I have eyes! You do not! See!? No?! Exactly! Because you can't fucking see because you have no fucking eyes! Arrggh!"
Just throwing that theory out there.
>>I mean seriously, passwords are among the weakest chain when it comes to security today
No, the weakest chain in security today is not the password. I recently found out that the idiot at Verizon typed in my password wrong when I bought my droid, so even though it'd autologin on my phone, I couldn't access my account (I was wanting to pull up the calendar) on my PC. There's no way of resetting the password without not logging in for a while, and my phone (of course), was always logging in, so it mean for going a number of days (weeks?) without having a phone, just to be able to reset my password.
After clicking around for a while, I found the weakest link: the password recovery password. Favorite food? Well... I won't tell you. It's a secret! Only I love these delicious cheese and pepperoni covered delicacies from Italy.
Why on earth we demand ridiculous password strengths on one hand but have the most mind-bogglingly easy to guess password recovery passwords on the other. They're easily the weakest link. IIRC, Palin's email got compromised by a similar means.
>>Pah! Trespasser was doing that back in 90's. Call me when there's a real revolution.
Heh heh, yeah that game was great. Watching your gun get caught in a chain link fence and then go sproinging hundreds of yards away was... well, "fun" is not exactly the right word for it.
>>I don't get the bit about the "six axes". I thought we had only three in meatspace.
Well, I only have one axe in meatspace.
I'm very excited about this controller though - I can't wait to see how inaccurate people will be in games when they actually have to aim, instead of just clicking on a point with their "sniper rifle" and pretending that's skill.
Being able to shoot yourself in the head in a FPS? That's revolutionary.
>>I wonder how you balance this with the fact that US courts don't care whether or not the people standing in front of them have been illegally abducted off foreign streets by intelligence agents, or whether the evidence against them was obtained by illegal means such as torture.
Which is precisely my point. It does bother me, but it doesn't seem to bother our legal system.
How could you say you care about illegal means used in law enforcement in one sentence and then say you don't worry about it happening in another?
If they tried to pull a Bush and claim that they couldn't reveal the source of information being used to convict me because it came from a privileged source, then it's a bit too late, isn't it. And the Bush administration did do these sorts of hijinx, ignoring the fact that you're always supposed to confront your accuser, as it were.
>>The question is: would that legalize downloading / Sharing since the artists are supposed to get payed?
Of course not.
And would artists actually see any of this money?
Of course not.
>>And INTERPOL isn't part of your law enforcement efforts either, except insomuch as they cooperate with national agencies that ARE law enforcement bodies and who must act within the law
So you're fine with some part of that chain not being one "who must act within the law"? Just curious.
I don't have anything against INTERPOL per se, and I in fact think they're a fine institution. However, no part of the legal system used for convictions should be outside the law. It makes sense, does it not?
Right, and that's why we saw things like the Cult of Reason, replacing the names of the week with atheist names, etc., because they were so few and uninfluential?
I've studied the French Revolution. I'd recommend you do the same.
Obviously it was a heterogeneous mixture, and different factions in the revolution had different takes on it. But it doesn't change the facts.
It's probably not that bad:
http://www.realclimate.org/index.php/archives/2009/08/ups-and-downs-of-sea-level-projections/
However, there's really no reason why we shouldn't be looking into solutions (geoengineering foremost among them) and switching to nuclear. There's a variety of Really Good Reasons why we should be on nuclear power, AGW just one of them.
>>From what I have read its around 45-60 hours for most people, decent value for the money IMHO
It took me about 80 to 90 in-game hours to beat it, with about 120 total spent in the game (I restarted and played through every pathway twice so I could see all the dialogue options and combats). Hours isn't an especially good way of measuring how fun a game is, and it did manage to keep my interest through the first three quarters of the game. But by the end, it was so easy that the combats went from being tactically interesting to trivially easy. (Two mages loaded up on crowd control spells will do that.) But the end of the game went by fast enough that it wasn't too bad.
I even decided to challenge myself by not using any of the army units you recruit throughout the game, and the final set of battles still were pretty easy.
I'm not sure if I'll get the expansion. They made similar claims "hours of extra gameplay" for the existing DLC, but both the Stone Prisoner and the Warden's Keep are less than an hour of gameplay each. A bit more if you play through all the different paths, and maybe a lot more if you want to solve the puzzle in the Stone Prisoner, but even still, not a lot of additional gameplay for the cost involved.
I'm holding off on uninstalling it because I want to see what sort of player generated campaigns people come up with. I sort of feel bad for the people that bought it for the console.
>>Oil and Coal use is perfectly fine if there aren't any environmental downsides to them, so there is absolutely no need to push for seeking replacements that aren't going to be implemented until they naturally would be anyway.
Besides, you know, the whole "middle east" dynamic going on?
Nuclear is a perfectly acceptable alternative to coal, but "it's scary" so we don't build nuclear plants any more. AGW issues aside, coal kills thousands of people every year. (And dumps lots of radiation into the atmosphere, but people never talk about that either.)
>>Recycling is another canard.
Agreed. What actually bothers me is that even with things like CRV in place, city governments can't run recycling programs without incurring a loss. In my town here, they've enrolled all apartment complexes in a "voluntary" recycling program (that they can't opt out of), put in recycling bins, and then charge them hundreds of dollars each month for the privilege of being in the program.
This is a town that has a serious problem with people going through trash looking for recycleables, in organized teams. Which means that obviously these people are turning a profit at it. So they outlawed these people. It boggles the mind.