rpackard

I am really interested in how this is measured. I have found this info on free sources so far

"The FIT Treadmill Score, calculated as [percentage of maximum predicted heart rate + 12(metabolic equivalents of task) – 4(age) + 43 if female]"

And this from another interview.

"In addition to age and gender, the formula factors in peak heart rate reached during intense exercise and the ability to tolerate physical exertion as measured by so-called metabolic equivalents, or METs, a gauge of how much energy the body expends during exercise. More vigorous activities require higher energy output (higher METs), better exercise tolerance and higher fitness level. An activity such as slow walking equals two METs, compared with eight for running."

If anyone has access to the paper and could better show the algorithm I would be super thankful.

EDIT: I think I figured out more but still not positive about MET unless the 2 for if heart rate hit max walking and 8 for if heart rate hit max with running. [(Measured Heart Rate Max/(220-age))100 + 12MET - 4*age + (43 if female)]

http://www.codeschool.com/courses/try-r

Codeschool recently launched a free tutorial class for R thats interactive. They do a great job with intro level learning.

This may actually be yet another way they are protecting users privacy.

For anyone that hasn't read the news: mega is securing users uploaded files from Mega itself by forcing users to encrypt the uploads. This encryption prevents mega from policing the content that their users upload and thus they can't be held accountable.

If they are really on the users side, Then this might be a way for mega to police copyrighters and DMCA people from creating scripts and application that publicly crawl mega.com looking for infringers.

Adding client side validations offloads some work from your servers. It could reduce the number of total requests that hit your server and overall increase your speed.

This may be premature optimization for a lot of cases but for form heavy apps this could be a useful optimization.

Also saving a round trip to the server can make the UX seem snappier and overall give a better user experience.

Started in highschool with visual basic and created a aol instant messenger punting program.

No classes in HS. In college was chem engineer and learned matlab and fortran for scientific computing and data analysis. Linux admin and bash scripting.

Finally, quit my job at a large consumer packaged goods company. And have spent the last few months learning ruby and ruby on rails to try and build my first business.

So no cs degree. Just a math heavy background and curiosity.

I worked R & D at a big consumer packaged goods company and ran into a lot of these 'excel messes' especially dealing with raw materials specifications. Here are my suggestions:

The 'R' programming language is pretty useful for statistics and could be used to aggregate a lot of files (and its free). Also if you want to generate dynamic reports using templates then you should look into sweave which is a package in 'R'. (http://www.stat.uni-muenchen.de/~leisch/Sweave/) You could build beautiful PDF reports that you simply run a script to generate if they are easily created into a template. But that would also require learning Latex for creating documents, something that is used heavily in the academic world but I have yet to see it in the business world.

Another approach which I would try is to use a programming language to dump it into a relational database and then use SQL to access the info (which is heavily used in business setting). This option also also allows for others to use it with familiar business tools. My unconventional tools of choice would be Ruby and using Rails to build an interface.

I think that you are mostly correct. In that, you do not want to lead your customers on too much and with simply testing one thing you will not be sure if the idea as a whole is bad. But there is a time and a place to ask a customer to decide between two choices. Usually this is done using discrete choice models and is pretty useful for ranking feature sets and even can help understand pricing.

I found a random white paper from a software company that I have never heard of before that actually gives a pretty simple example and explanation. (http://sawtoothsoftware.com/download/techpap/undca15.pdf)

In a SaaS application this could be extremely useful in seeing how much perceived utility people get from X GB more space or Y more users...

For the OP it might be cool to extend the idea to having multiple binary choices and performing the necessary math to understand the utility for features (multi-linear regression).