Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
Slashdot Deals: Deal of the Day - Pay What You Want for the Learn to Code Bundle, includes AngularJS, Python, HTML5, Ruby, and more. ×

Comment Re: The problem with data driven science.. (Score 1) 55

I agree that defining terms is important. FYI, I think you should consult a external reference for a definition of meaningful, as you did much higher up the thread with "data". Your usage of the word seems unusually specific, specifically the "automatically" part referring to data without further analysis. Collecting and analyzing data is what researchers do, so requiring data to be useful without analysis is baffling. Several commenters have tried to correct your definition.

Even data without pristine integrity can be collected and analyzed to create something meaningful. For an example of that, how about neurosynth.org - this uses an automatic scraper to collect coordinates from published neuroimaging papers. The scraper does as scrapers do and collects all sorts of false positives, so many of the "coordinates" are in fact not. The overall result produces very strong results.

Comment Re:The problem with data driven science.. (Score 1) 55

Some people use their accelerometer data as a pedometer, some don't. Is the data meaningful or not? Whether data is meaningful, or useful or significant (all are fine with me) is subjective. Accelerometer data is not meaningful to you? Fine. But, you are not the arbiter of what is meaningful.

On the other hand, if you'd like an example of data which is (nearly?) universally recognized as meaningless, how about /dev/random > data.txt?

"When it comes to humility, I'm the greatest." -- Bullwinkle Moose