"Methods to obtain "95% significance" out of a measly 14 observations of what could for all intents and purposes be a random walk are usually taught in politics 101."
Methods to misrepresent research are taught in politics 101.
We have MUCH more than 14 measurements. Each year's datapoint is an aggregation of multiple measurements.
"Not like in bogey-science, where opposing views are heretics, underlying data is top-secret, claims are even theoretically falsifiable only after waiting at least 30 years, all anecdotal evidence in favor is significant and all anecdotal evidence to the contrary is just coincidence."
You are stupid. Most of underlying climate data AND models are open. You can easily get several open independent datasets and use different models on them.
"Especially in mediums that are only (or at least primarily) going to be read by educated people, there's no reason to prevent debate - the facts will speak for themselves."
Only it's known that the medium IS NOT going to be read by educated HONEST people.
Do you know how many other independent datasets point to global warming? Has Glen Beck told you that even though CRU dataset might be questioned, other datasets still point to global warming?
"So then you go to what we do know, that temperatures have risen.12C a decade and ask, how much of that is due to CO2. If you dig into, for example, the IPCC report, you get the answer is that most of it is probably caused by CO2 (and other GHG). OK, fair enough, how do we know? Dig deeper into the IPCC report and it's based on climate models. OK, and how do we know the climate models are reliable? The answer is we don't, and if you keep digging, you will find that they aren't."
Buzz! Wrong. We calibrate models on historic data and check if their predictions match the reality.
"The contention of skeptics is that since the 'raw' data and the methodology for the normalization are not clearly known in some cases, this brings the data into question."
Bullshit. Data filtering procedures are clearly documented in MANY papers. They are not controversial at all.
We had the same exact situation here (Kiev, Ukraine). As far as I know, it's now a standard operating procedure for telecoms here to wait 1-2 months before turning on a base station after it is installed, so such complaints can be tossed out immediately.
All banking sites use HTTPS. So simple traffic listening won't help you.
You'll need to do man-in-the-middle attack, and that's not simple. On Windows you'll have to do it in the kernel level, probably even below the TDI. Doable, but extremely hard.
A pedestrian can slip and fall on the road. Or a dog might frighten them and make them jump.
A lot of can happen. A good AI should predict the worst possible cases and avoid them. That probably means you'll probably ride a bit slower in cities, but that's not really a problem if your car can compensate for it using intelligent planning. Also, if most of cars are automatic - you'll be able to say 'bye-bye' to traffic jams.
"If the car were driving itself, that would have been a nasty surprise for him to suddenly find that he hit something big."
So? Car safety systems won't go away. If anything, a car can tighten seatbelt and deploy airbags in advance if collision is imminent.
"I'd like to see added measures put into cars, like automatic braking when an object is detected in the lane. That works fine and dandy until you realize you're not always going straight on an empty road."
Hm. And what makes you think computers won't be able to evaluate this case? It's quite common.
BTW, automatic cars will probably have LIDARs and sonars to measure speed and direction of other cars, so they'll be able to tell that the other car is not on the projected collision path.
Yet self-driving cars are the future. I believe that it's possible to make a much safer automatic car, eventually surpassing safety of even very good human drivers.
And the best thing - I'll be able to read books while driving!
Small correction: 'we're allowed to have more than 2 Jews work in our department' should be 'we're NOT allowed to have more than 2 Jews working in our department'.
No, I'm talking about discrimination of 60-70-s. It was not 'line up Jews against the wall and shoot them' type of discrimination, but rather 'we're allowed to have more than 2 Jews work in our department'. Also, discrimination flared up when emigration to Israel/USA had started.
There was a joke which goes like this - a Jewish candidate with perfect resume wants to work in a lab and is refused: - Why don't you want me to work here? - Because you'll emigrate soon. - But I won't emigrate, I love the Party and the USSR! - Even worse, we don't need idiots here.
Actually, Jews were discriminated against in the USSR at that time (it was semi-official policy). So he might have felt the effects of this discrimination. Or his parents did.
Slashdot Mirror
"Methods to obtain "95% significance" out of a measly 14 observations of what could for all intents and purposes be a random walk are usually taught in politics 101."
Methods to misrepresent research are taught in politics 101.
We have MUCH more than 14 measurements. Each year's datapoint is an aggregation of multiple measurements.
We also check how predictions of models compare with reality.
So far, their predictions are correct.
"Not like in bogey-science, where opposing views are heretics, underlying data is top-secret, claims are even theoretically falsifiable only after waiting at least 30 years, all anecdotal evidence in favor is significant and all anecdotal evidence to the contrary is just coincidence."
You are stupid. Most of underlying climate data AND models are open. You can easily get several open independent datasets and use different models on them.
There's no conspiracy.
"Especially in mediums that are only (or at least primarily) going to be read by educated people, there's no reason to prevent debate - the facts will speak for themselves."
Only it's known that the medium IS NOT going to be read by educated HONEST people.
Do you know how many other independent datasets point to global warming? Has Glen Beck told you that even though CRU dataset might be questioned, other datasets still point to global warming?
No? Why not?
"So then you go to what we do know, that temperatures have risen .12C a decade and ask, how much of that is due to CO2. If you dig into, for example, the IPCC report, you get the answer is that most of it is probably caused by CO2 (and other GHG). OK, fair enough, how do we know? Dig deeper into the IPCC report and it's based on climate models. OK, and how do we know the climate models are reliable? The answer is we don't, and if you keep digging, you will find that they aren't."
Buzz! Wrong. We calibrate models on historic data and check if their predictions match the reality.
"The contention of skeptics is that since the 'raw' data and the methodology for the normalization are not clearly known in some cases, this brings the data into question."
Bullshit. Data filtering procedures are clearly documented in MANY papers. They are not controversial at all.
The most commonly used model is here: http://data.giss.nasa.gov/gistemp/ . Reimplementations and other models exist as well. Go on and play with this one, for example: http://code.google.com/p/ccc-gistemp/
I'm sorry, you are jack shit.
All the relevant data and algorithms were open since, well, forever. I personally worked with several global datasets using open source models.
Go on, educate yourself: http://www.realclimate.org/index.php/data-sources/
Uhm. That's what statistics 101 is all about.
"While higher CO2 and temperatures might lead to higher crop yields in some regions"
It won't. Higher CO2 concentration doesn't increase crop productivity (though it somewhat increases tolerance to extreme conditions).
"Linux distributions release entire new packages - not patches."
Uhm. Have you ever looked inside Windows patches? They are in essence what Debian Stable updates are. I.e. new versions of affected software.
They are definitely NOT patches in original sense (i.e. a diff between two versions).
This story is typical.
We had the same exact situation here (Kiev, Ukraine). As far as I know, it's now a standard operating procedure for telecoms here to wait 1-2 months before turning on a base station after it is installed, so such complaints can be tossed out immediately.
"Example: If you take a look at the last major Microsoft conference (MIX10) then you'll see there we no new WPF features added"
WPF is definitely not going away. Look at VisualStudio 2010, or to be precise, look at its editor. Yes, it's WPF.
Not much new features are added to it because it's in the stabilization phase for now. So yes, it's just resting.
Not having a WPF implementation created a one-way street - Mono applications will work on Windows, but not vice versa.
That doesn't help Linux.
Are you sure they haven't just shipped Debian on these sticks? :)
WTF?!?
WPF is VERY alive. You're confusing it with WinForms (which is dead).
Aaaand Silverlight just uses a subset of WPF for drawing. So you're in essence saying that WPF is dead because its subset is alive :)
How about WPF, then? When are you going to develop it? I know: never. The scope of it is way too large for your team.
Same about complete WCF, WWF (and other WTFs).
So in reality, a Mono application will probably work on Windows, but almost none of Windows C# applications will work on Mono.
All banking sites use HTTPS. So simple traffic listening won't help you.
You'll need to do man-in-the-middle attack, and that's not simple. On Windows you'll have to do it in the kernel level, probably even below the TDI. Doable, but extremely hard.
Uhm. What's so complex about it?
Store a database of engine-power vs. GPS coordinates and use it to anticipate conditions.
Also, you can actually SHARE this database so your car will be able to anticipate slopes even if you've never used that road.
A pedestrian can slip and fall on the road. Or a dog might frighten them and make them jump.
A lot of can happen. A good AI should predict the worst possible cases and avoid them. That probably means you'll probably ride a bit slower in cities, but that's not really a problem if your car can compensate for it using intelligent planning. Also, if most of cars are automatic - you'll be able to say 'bye-bye' to traffic jams.
"If the car were driving itself, that would have been a nasty surprise for him to suddenly find that he hit something big."
So? Car safety systems won't go away. If anything, a car can tighten seatbelt and deploy airbags in advance if collision is imminent.
"I'd like to see added measures put into cars, like automatic braking when an object is detected in the lane. That works fine and dandy until you realize you're not always going straight on an empty road."
Hm. And what makes you think computers won't be able to evaluate this case? It's quite common.
BTW, automatic cars will probably have LIDARs and sonars to measure speed and direction of other cars, so they'll be able to tell that the other car is not on the projected collision path.
Why do you think a car can't identify playing children? Or for that matter, any pedestrian walking nearby.
Yes, self-driving cars will need a lot of advances in computer vision. But I don't think it will be really impossible.
Yet self-driving cars are the future. I believe that it's possible to make a much safer automatic car, eventually surpassing safety of even very good human drivers.
And the best thing - I'll be able to read books while driving!
We've discussed this in the past: http://news.slashdot.org/article.pl?sid=08/07/24/220225
Small correction: 'we're allowed to have more than 2 Jews work in our department' should be 'we're NOT allowed to have more than 2 Jews working in our department'.
No, I'm talking about discrimination of 60-70-s. It was not 'line up Jews against the wall and shoot them' type of discrimination, but rather 'we're allowed to have more than 2 Jews work in our department'. Also, discrimination flared up when emigration to Israel/USA had started.
There was a joke which goes like this - a Jewish candidate with perfect resume wants to work in a lab and is refused:
- Why don't you want me to work here?
- Because you'll emigrate soon.
- But I won't emigrate, I love the Party and the USSR!
- Even worse, we don't need idiots here.
You can. Although in this cases the other company might sue you if they find that your claims are bogus (which they almost certainly are).
Actually, Jews were discriminated against in the USSR at that time (it was semi-official policy). So he might have felt the effects of this discrimination. Or his parents did.