Interesting links in July 2013

Fortunately some of the top companies (like Google) understood that giving people brainteasers and other hard (and artificial) tasks does not return great software developer from an interview. Personally I have to admit that I suck at those tricky algorithmic questions. I feel too stressed and I hardly find solution in a given time. But... do you often solve such problems at your work? I have some experience from various companies, projects and technologies and in only a small part of the code you have to invent your own awesome algorithms. The rest is "quite simple" daily programming work. So at the interview it is better to ask for answers regarding those "simple" problems, that make the 99% percent of the tasks. 

Hopefully we will see some improvements in such interviews in the near future.

This post caught my attention because of the above sentence. Is this the future of GPU Apis? Recently we got new, a bit refreshed, versions of DX11.2 and OpenGL 4.4 and we see that those Apis become more and more programmable. More and more we can do on the GPU side only. I think that, as in most cases, we need to find a golden mean: one side we have one unified API (for all different vendors) and the second extreme there is no overhead from the driver. 

This is an interesting discussion on the http://programmers.stackexchange.com related to visible fact that Microsoft is doing a great job to bring more popularity to C++. With the raise of low powered devices C++ fits perfectly as a universal and powerful programming language. We can use Java or .NET but obviously they need to much resource to run. C++ being a native language gives us a lot of freedom to develop interactive apps. C++11 is becoming more and more accepted so it is even easier to code. .NET will not be replaced by C++ WinRT, but the reverse is also not true: .NET Framework will not replace C++. Having WinRT we got another option to build applications.

Since I have some experience with EEG and biofeedback I was quite interested in this article. For sure during next five years we will still need a mouse and a keyboard, but maybe after those years there will be something different? 

In general perceptual computing means using human body: gestures, facial expressions, muscle tension, electric brain activity, and even more of our actions to create more natural communication with computers and electronic devices. 

I have seen pretty impressive results of using EEG to control computer games. It can be used not only for fun but for medical treatment and improving your mental abilities as well (like stress reduction treatment). Although some of techniques are not that new (early biofeedback studies comes from 70s, but it widespread probably in late 90s). Now computers are so powerful that even a cellphone can 'compute' and generate interesting results. Moreover we can now combine EEG with facial recognition (and some more features) to get more accurate effects.

For sure we will be seeing more and more advanced usage of it.