From watching my autism younger brother over the last half dozen years or so (when he's had access to treatments, drugs, etc. that help), they would likely gradually adapt. The therapies, drugs, etc. now help them to slowly build the abilities that they lack. I expect they won't just "wake up normal" one morning since they would be missing the cognitive development based on the experiences that they couldn't previously understand/assimilate/[insert your definition of autism here]. A "cure" would be enabling them to gain the capacity to learn and develop in ways that were previously unavailable to them.
Ask any parent whose child did the apparent regression around age 4 to autism [1]. It'd be the same thing in reverse, maybe more slowly.
Maybe you wouldn't want to cure all Asperger's but if you've known someone with severe autism there's no question that anything that can be done should be done.
Purely by source for the initial submission? That would be nice on the email quota. From my little corner of the world it seems like they want PDF or similar to send out to the reviewers and they'll take your LaTeX source later when/if it's accepted and you've made the changes.
The lack of Tex/Latex on these lists also seems to be causing some confusion. I can't see why anyone would send an initial submission in Latex source. Any Latex document will be compiled to PostScript or PDF for distribution by the writer. Accepting PostScript or PDF is accepting Latex/Tex.
At least in my field (operations research / math / cs / machine intelligence) the journals (e.g., INFORMS: http://templates.pubs.informs.org/) provide not only formatting guidelines but template and Latex style files. In that case it's just a matter of downloading the style file, sticking it in the folder with the latex source, and recompiling it. Most even provide instructions in case your run into trouble.
What about self-selection due to knowledge (or perception) of difficulty entering the field? It's quite plausible that there's a feedback mechanism such that women don't spend years studying fields that they've been socialized to think are "male".
It could also be that they start into IT/math/whatever and get tired of people being shocked when they tell them what they do.
Or am I the only one with that problem? It's really annoying to be frequently reminded that people need to be convinced that you can do what you spend your life doing. No one is surprised when I make good cookies or get a small child to stop crying! Why must they be shocked when I do good math?
The closely related phenomena is that people feel the need to tell me I'm good at this stuff. It's as if they expect me not to realize I'm good at it, since it was such a shock to them. I swear it's as if they caught their pet gerbil building a rocket, completely unthinkable but kind of cool.
It's not just price optimization that's relatively easy to learn. There's also location theory and optimization (e.g., retail, essential and emergency services, noxious facilities), routing and scheduling, order and inventory levels,... The math is relatively simple (most of the time) and if you can set up the problems there are some decent OS software packages with very good solvers. COIN-OR comes to mind but I'm sure there are several.
The real issues are getting the data and interpreting the results. I've got a land use decision model for urban fringe areas that's doing a reasonable job of presenting sets of potential solutions for decision-makers. The tough stuff is the pre-processing, e.g., defining what's feasible and desirable, and interpreting the results, e.g., what solutions are or aren't significantly different. For this type of work geographic data is readily available. I doubt that will be the case for most business decisions.
There's a reason that warriors have historically been male. If you kill off too many women you can't restore the loss in population of your tribe/country/whatever. If you kill off too many men the remaining ones are just very happy.
Consider the post-WWII baby boom, we might want to consider adjusting the draft to deal with modern society's post-war responses.
Forgive my lack of American knowledge but would there not be someone somewhere who knows what these ZIP+5 codes correspond to? Perhaps in the public domain or hands of the government? If I look up a Canadian postal code I can get the list of addresses corresponding to it. Surely finer granularity coupled with information like "number of units" would give a better, although still less than ideal, estimate?
IMHO, this problem is just begging for a decent geographic information system (think database with geographic attributes and analytical capabilities). Nice little georeferenced polygons (zip codes or whatever) with attributes maybe some points representing the information that's actually known. Not quite yet, we'll need to centralize the data, but I can check the elevation on a new build lot in my city whenever I feel like it or see which houses on my street have permits to build decks. Oh, and GIS people are used to dealing with scale issues including problems just like this one.
I can't help it, I've got to post this (from TA linked by parent):
The Straphangers Campaign and Transportation Alternatives said the bus - which carries frustrated riders across town from East 20th Street/Avenue C to West 23rd Street/12th Avenue - travels at an average speed of 3.4 mph during the afternoon rush - slower than a chicken, which travels at speeds up to 9 mph.
A king penguin can swim at an average rate of 5.3 mph, while the average person can walk at a speed of about 3 mph.
"There's something terribly wrong when the slowest city buses come in a distant third behind chickens and penguins and just ahead of the average pedestrian," said Gene Russianoff, staff lawyer for the Straphangers Campaign. I'd love to know who decided that the sample should be {bus, penguin, chicken}?
Now I use LaTeX Beamer and could not be happier. Maybe S5 would be great for talks that have few or no figures or equations and just bullet points, but that is not enough to help me. With Beamer I get a single pdf with everything and it looks the same regardless of what computer/OS I show it on. All done using nothing more than the free software I normally use. While I do use Beamer (and think it's great) it's not necessarily the greatest solution for talks with many figures. It's great with equations but having to define a grid and explicitly place figures (e.g., to have a column of text on half of the slide with an image next to it) is a pain most of the time. Unless you're comfortable with Pstricks, of course.
For anyone familiar with Latex it's well worth learning. Nearly any functionality that you can use in Latex can be used with Beamer. If you're presenting report that you've already written it's really nice to just cut, paste, parse, and edit it down.
But this is really no different than saying males do not have the same capacity for child birth, because, guess what, regardless of what science comes up with, females will still be better suited for this task. Actually it's quite distinctly different. Statistically women are less likely to be "good at math and science". Statistics is a generalization of the properties of a group and cannot be used to say anything about any individual. So you can pick a male and female at random and find the female is better at math and science than the male. The difference in child-bearing abilities isn't statistical, it's true of every person. Female implies potential for childbirth (barring certain medical issues) and male implies no potential for childbirth. Females aren't "better suited" for childbirth, they're capable and men aren't.
Ok, you probably didn't mean that but plenty of people don't realize they're committing the same fallacy.
We're different people, you and me. You see, I'm ok with a bad script and over-use of special effects, if it has Han Solo to enhance it. I'll agree but only because I'm female. Bad movies suck but there can be no bad movie with Harrison Ford. While you men sit around debating whether he should do Han or Indy again I'll just wish he'd done both 15 years ago.
Canadian Tire (something of a hardware store that increasingly thinks it's a department store) carries dimmer CFL bulbs. I believe they're from GE. There are also "soft white" CFL's (for those who find the typical CFL light harsh) and some with plastic "bulbs" surrounding the tube to soften and better disperse the light (also available at Ikea).
Dimmable CFL's are still fairly expensive though.
Let's pretend the odds of a person being tested being a terrorist are 1/10,000 (yes, that's high). Then in a population of 1,000,000 (100 terrorists total) would have:
85 true terrorists labelled as such
79992 innocent people labelled as terrorists (0.08*(1,000,000-100))
15 true terrorists not labelled as terrorists
919908 innocent people not being labelled as terrorists
Taking the "terrorist labelled" group we have 85/80077 or approximately 0.1% actual terrorists (where the total population had 0.01%). It gets worse if the base rate (# terrorists/# people) is lower.
Same thing applies to medical tests, even very high "accuracy" (i.e., low rate of false positives and high rate of true positives) can lead to tests where given a "positive" result you're actually not very likely to have the disease, if the actual rate of occurence is low enough.
Slashdot Mirror
From watching my autism younger brother over the last half dozen years or so (when he's had access to treatments, drugs, etc. that help), they would likely gradually adapt. The therapies, drugs, etc. now help them to slowly build the abilities that they lack. I expect they won't just "wake up normal" one morning since they would be missing the cognitive development based on the experiences that they couldn't previously understand/assimilate/[insert your definition of autism here]. A "cure" would be enabling them to gain the capacity to learn and develop in ways that were previously unavailable to them.
Ask any parent whose child did the apparent regression around age 4 to autism [1]. It'd be the same thing in reverse, maybe more slowly.
Maybe you wouldn't want to cure all Asperger's but if you've known someone with severe autism there's no question that anything that can be done should be done.
[1] Typical only of some types of autism.
Purely by source for the initial submission? That would be nice on the email quota. From my little corner of the world it seems like they want PDF or similar to send out to the reviewers and they'll take your LaTeX source later when/if it's accepted and you've made the changes.
The lack of Tex/Latex on these lists also seems to be causing some confusion. I can't see why anyone would send an initial submission in Latex source. Any Latex document will be compiled to PostScript or PDF for distribution by the writer. Accepting PostScript or PDF is accepting Latex/Tex.
At least in my field (operations research / math / cs / machine intelligence) the journals (e.g., INFORMS: http://templates.pubs.informs.org/) provide not only formatting guidelines but template and Latex style files. In that case it's just a matter of downloading the style file, sticking it in the folder with the latex source, and recompiling it. Most even provide instructions in case your run into trouble.
What about self-selection due to knowledge (or perception) of difficulty entering the field? It's quite plausible that there's a feedback mechanism such that women don't spend years studying fields that they've been socialized to think are "male".
It could also be that they start into IT/math/whatever and get tired of people being shocked when they tell them what they do.
Or am I the only one with that problem? It's really annoying to be frequently reminded that people need to be convinced that you can do what you spend your life doing. No one is surprised when I make good cookies or get a small child to stop crying! Why must they be shocked when I do good math?
The closely related phenomena is that people feel the need to tell me I'm good at this stuff. It's as if they expect me not to realize I'm good at it, since it was such a shock to them. I swear it's as if they caught their pet gerbil building a rocket, completely unthinkable but kind of cool.
It's not just price optimization that's relatively easy to learn. There's also location theory and optimization (e.g., retail, essential and emergency services, noxious facilities), routing and scheduling, order and inventory levels, ... The math is relatively simple (most of the time) and if you can set up the problems there are some decent OS software packages with very good solvers. COIN-OR comes to mind but I'm sure there are several.
The real issues are getting the data and interpreting the results. I've got a land use decision model for urban fringe areas that's doing a reasonable job of presenting sets of potential solutions for decision-makers. The tough stuff is the pre-processing, e.g., defining what's feasible and desirable, and interpreting the results, e.g., what solutions are or aren't significantly different. For this type of work geographic data is readily available. I doubt that will be the case for most business decisions.
There's a reason that warriors have historically been male. If you kill off too many women you can't restore the loss in population of your tribe/country/whatever. If you kill off too many men the remaining ones are just very happy.
Consider the post-WWII baby boom, we might want to consider adjusting the draft to deal with modern society's post-war responses.
Forgive my lack of American knowledge but would there not be someone somewhere who knows what these ZIP+5 codes correspond to? Perhaps in the public domain or hands of the government? If I look up a Canadian postal code I can get the list of addresses corresponding to it. Surely finer granularity coupled with information like "number of units" would give a better, although still less than ideal, estimate? IMHO, this problem is just begging for a decent geographic information system (think database with geographic attributes and analytical capabilities). Nice little georeferenced polygons (zip codes or whatever) with attributes maybe some points representing the information that's actually known. Not quite yet, we'll need to centralize the data, but I can check the elevation on a new build lot in my city whenever I feel like it or see which houses on my street have permits to build decks. Oh, and GIS people are used to dealing with scale issues including problems just like this one.
But why would the volcano god want all the virgins if no one else does?
At the same time?
A king penguin can swim at an average rate of 5.3 mph, while the average person can walk at a speed of about 3 mph.
"There's something terribly wrong when the slowest city buses come in a distant third behind chickens and penguins and just ahead of the average pedestrian," said Gene Russianoff, staff lawyer for the Straphangers Campaign.
I'd love to know who decided that the sample should be {bus, penguin, chicken}?
While I do use Beamer (and think it's great) it's not necessarily the greatest solution for talks with many figures. It's great with equations but having to define a grid and explicitly place figures (e.g., to have a column of text on half of the slide with an image next to it) is a pain most of the time. Unless you're comfortable with Pstricks, of course. For anyone familiar with Latex it's well worth learning. Nearly any functionality that you can use in Latex can be used with Beamer. If you're presenting report that you've already written it's really nice to just cut, paste, parse, and edit it down.
Either way they have to learn it, teaching just makes it more likely.
Canadian Tire (something of a hardware store that increasingly thinks it's a department store) carries dimmer CFL bulbs. I believe they're from GE. There are also "soft white" CFL's (for those who find the typical CFL light harsh) and some with plastic "bulbs" surrounding the tube to soften and better disperse the light (also available at Ikea). Dimmable CFL's are still fairly expensive though.
Let's pretend the odds of a person being tested being a terrorist are 1/10,000 (yes, that's high). Then in a population of 1,000,000 (100 terrorists total) would have:
85 true terrorists labelled as such
79992 innocent people labelled as terrorists (0.08*(1,000,000-100))
15 true terrorists not labelled as terrorists
919908 innocent people not being labelled as terrorists
Taking the "terrorist labelled" group we have 85/80077 or approximately 0.1% actual terrorists (where the total population had 0.01%). It gets worse if the base rate (# terrorists/# people) is lower.
Same thing applies to medical tests, even very high "accuracy" (i.e., low rate of false positives and high rate of true positives) can lead to tests where given a "positive" result you're actually not very likely to have the disease, if the actual rate of occurence is low enough.