Forgot your password?

Comment: Re:classroom tools (Score 1) 210

by ph0rk (#45722537) Attached to: Datawind Not Blowing Smoke: $38 Tablet Coming To the US

Graduate students and professors need to "publish or perish". I'm hoping that at least some of them will use at least some of their publishing time to write free textbooks.

It is unlikely that a free textbook (or any textbook, really) will count for tenure, in either the social or physical sciences, barring very high level technical textbooks.

So, yes, they would be wasting their time unless already well established. That rules out grad students and early career professors.

Comment: Re:Apple made the same mistake (Score 2) 390

by ph0rk (#45298927) Attached to: Smartphone Sales: Apple Squeezed, Blackberry Squashed, Android 81.3%

However, people are getting more educated and tech-savvy in general.

That is false: familiarity with facebook does not mean tech-savvy.

A surprising portion of even the very best and brightest 18-22 year olds would still hold a floppy disk completely level if you told them the bits might fall off.

Comment: Re:Apple made the same mistake (Score 2) 390

by ph0rk (#45298909) Attached to: Smartphone Sales: Apple Squeezed, Blackberry Squashed, Android 81.3%
It has always seemed to me that an iPhone only really made sense if you were already an almost completely Apple shop. If you already use OSX everywhere, have a few apple TVs or airport speakers lying around, an iPhone is tops for integration.

I've never understood why anyone would buy one that didn't already have at least one OSX machine.

Comment: Re:Med students (Score 1) 446

by ph0rk (#43824845) Attached to: Med Students Unaware of Their Bias Against Obese Patients

Even if X is often correlated with Y, it doesn't justify the assumption that X always implies Y.

While that is true, the safe bet is still going to be that X implies Y.

If the first premise is true (X is highly correlated with Y), then to expect Y when one finds X is only natural (and takes less processing time).

Now, if we had some clear cases where X doesn't lead to Y, for example when Z is present, then we can solve the problem of unfairly expecting Y by also looking for Z. Hunting for Z will probably be more fruitful in the long run than trying to train people to ignore stereotypes that have evidentiary support.

Comment: Re:Big enough sample size (Score 2) 578

by ph0rk (#43761407) Attached to: Of 1000 Americans Polled, Most Would Ban Home Printing of Guns
Wasting mod points to post, but: US Americans are not that heterogenous. What specific groups (with dissenting views relevant to the matter at hand) are systematically excluded from the sample?

They offer up their sampling procedures and methodology here.

A larger sample size is not inherently better. 1000 isn't much different from 10,000 or 10 million. If the sampling method would be unrepresentative with 1000 cases, it wouldn't be any better with more.

Comment: Re:not where from, where to? (Score 1) 523

I've played games that let the old content sit untouched.

Hardly anyone played it, so you had to essentially con top level players to milk run you through content to get geared - unlikely in cases where there wasn't already a real-world friendship. Once a game hits that point the influx new players that stick with the game long enough to raid drop to nearly zero, and the game ossifies, then dies. There is no way that it is a "better" solution.

"The value of marriage is not that adults produce children, but that children produce adults." -- Peter De Vries