theatraine

This article gives a good overview in the USA: http://en.wikipedia.org/wiki/Largest_cities_in_the_United_St...

http://www.nextmicrosoftceo.com/ presents some good candidates. However, I doubt that the first contender, Gabe Newell would leave Valve, and that the second-place contender, RMS, would be a good business decision.

Operates in slow motion on Metro IE11, interestingly, desktop IE11 performs well although it starts off slow. It flies in FF and Chrome though. I suppose the Metro IE11 is missing some WebGL optimizations. Very cool!

Right, from the revenue report: "Commercial cloud revenue grew 103%". This appears to be the largest section of growth reported, so it looks like Microsoft is doing pretty well in the cloud space.

There's no compression. It's expansion, since null bytes were added to increase the digit length of the resulting prime, making it long enough to publish. For an arbitrary binary, the original (most likely non-prime) zipped representation is more useful since it's smaller.

Sure, you could pre-calculate your filter coefficients and use a bandpass for each frequency band then do the math in Javascript however it's slower, and more involved. Also, unless you're going to be using WAVs, there's going to be some overhead for the decoding of the sound. Also, the filter won't be very high order since you're going to be doing at least n*44100 operations per second (at the standard sampling rate). (Web Audio API filter nodes are only 2nd order but you can cascade them.) Do you have examples of the above working in Firefox? I'd be curious to see the implementation.

Right, the EQ controls the volume of each of the 10 (each one corresponds to a slider) sounds. (Every sound has been previously decomposed into its frequency components.) The Web Audio API is used for browsers that have it present, else HTML5 audio elements are used. The slider just adjusts the volume of the sound (exponentially of course!), leading to an equalizer effect with no filtering required. (Garnered from looking at the code.)

Interestingly, this could be done using one file (rather than ten) using a Web Audio API filter node. This would limit the site to only Web Audio API compatible browsers. Unfortunately, at this time, that's only newer versions of Chrome, Safari, and I believe FF nightly (slight API variations). However there is some mobile support through recent Chrome and Safari implementations.

I wonder how Microsoft's Drawbridge OS (http://research.microsoft.com/en-us/projects/drawbridge/) will compare to LXC, and the Docker APIs? Currently Drawbridge looks like it's lacking adoption, and doesn't seem to be widely available. Regardless, the container model looks like it solves a lot of PaaS security issues without the overhead of VMs (Iaas).

The Onion's parody on Facebook's CIA ownership is now sounding eerily accurate. For those who haven't seen it: http://www.youtube.com/watch?v=cqggW08BWO0

Even if you don't have an American degree, it helps for immigration purposes. For example, I have a Canadian MS, allowing me to apply for a green card under the EB2 category. The EB2 category has an expected wait time of about a year as opposed to the EB3 category (if I only had an undergraduate degree), which would have a wait time of around 5 years.

Bear in mind that Azure storage includes "Blobs, Disks/Drives, Tables, and Queues" however S3 is only blobs (other services like Amazon's DynamoDB would be analogous to table storage). Hence, it's not an apples-to-apples comparison.

I would be interested to see the Summly summarization of this article! I think it would read something like this:

Yahoo bought Summly, a startup with an app that not many people used, made no money, and employed no technical geniuses (since it wasn't actually built by them), because Marissa Mayer believes summarization technology is "going to be huge for Yahoo".

This seems like a violation of the first amendment’s freedom of speech protections. It's offensive regardless of any alignment with animal protection groups.

I haven't used the checkout button yet, but here's what I learned from rolling my own: the form comes in with a token which you then charge with the stripe Python API, and on successful payment, you receive a callback at your webhooks url (which you set in the stripe settings). Take a look at the Django-Zebra app (https://github.com/GoodCloud/django-zebra).

Windows Azure Websites (http://www.windowsazure.com/en-us/home/scenarios/web-sites/) is similar to what you're describing, you can install a variety of popular platforms in one click (ie Wordpress, Drupal, Joomla, DotNetNuke,...) and get up and running right away. You own your data, the apps (ie Drupal, not Azure itself) are open source, and most things are easily managed for you.

Does anyone know if the same thing is possible for Azure Blob Storage? I don't believe it is now, due to the Cross Origin Resource Sharing (CORS) limitations of Azure (http://social.msdn.microsoft.com/Forums/en-US/windowsazureda...), but I'd like to double check.

Matt, thanks for hosting the event, it was a great way to meet and talk to like-minded individuals. Also, my compliments to the cook, the meal was delicious! Best of luck with http://SwigMe.com, Adrian (co-founder of http://audionotch.com)

I haven't tried garlic, however I have had very good results supplementing with Magnesium (I use the Citrate form). New research has verified the efficacy of Mg as well, see http://www.sciencedaily.com/releases/2012/03/120313230354.ht..., for example. It also has a positive effect on testosterone!

They're also building up their human capital. They've recently been hiring quite a lot of engineers, which will certainly yield later returns. I'm basing this on my own experience with them, as I got a job offer, spoke with several recruiters, and toured their campus, and now live in Seattle.

How the device fits over the lock handle reminds me of another device I made to unlock a door. I had to get into a room to reset a router, however the lock was a deadbolt, and I didn't know how to pick it. Using a (heavily bent) coat hanger, I squeezed it through a crack in the top of the door, lowered it with twine until it grasped the lock handle, and then turned the device with some more twine (from under the door) and I was in.

Locks that open when the door handle (not a knob) is turned can also be opened relatively easily with coathangers from under the door. Make sure to consider all aspects when making secure systems!

You'd be much more motivated if you have thousands of dollars in your account! A similar thing happened with my business's PayPal. I tried to transfer funds, and they decided that it was a good time to re-verify our location. Since we missed the phone call, we now have to wait for them to mail a letter.

The no control for medication is primarily based on the conclusions of Boutros 2008, a summary of spectral abnormality studies of schizophrenia. I reproduced a table from the paper in the appendix. What type of psych drugs were you working with?

As an aside drugs used to treat schizophrenia have especially nasty side-effects, such as weight-gain, reduction of white blood cells, and--ironically--some of the "negative" symptoms of schizophrenia when given to healthy people. In fact the Soviets used to use antipsychotics to torture political prisoners.

I concur that active EEG could lead to a much more accurate diagnosis, however we had access to quite a bit more resting data, and it does make the math easier (stationarity for one). Standardization of other factors would also probably increase accuracy, but once again, I was limited by my data. Since I'm no longer in academia, I won't be continuing the research, but hopefully further studies address these questions in more detail.

An interesting study highlighting some of the problems with diagnosis of schizophrenia was the Rosenhan experiment (http://en.wikipedia.org/wiki/Rosenhan_experiment) essentially it showed that doctors working in mental hospitals were unable to reliably tell the difference between those with schizophrenia, and those without it who reported a single hallucination. Albeit, the study is now close to 40 years old, and the diagnosis techniques have been improved, with for example, further editions of the DSM, major problems still exist.

In my Master's thesis I created an algorithm to diagnose schizophrenia from EEG recordings. It's obviously very preliminary research, but it would be amazing to see computers revolutionizing the way that we do neurological diagnoses.

My thesis is here for those interested: http://www.adriangreen.ca/Green_Adrian_CA_201211_MASc_thesis..., and hopefully soon to be condensed and published as a paper.

I just tried it out with a .psd, the downloaded zip contained what appeared to be other users projects. This may be disconcerting to some, and will hopefully be quickly addressed.

The tool, however, is very impressive. I'll try to respond with a list of bugs I notice once I figure out how to fix them through hand-coding. Good work!

From an SEO perspective, does the generation of fresh content increase your rankings in SERPs by itself? Or is the idea to generate interesting content that others will link to?

It's also worth mentioning that blogging gives you a great way to go after the long tail.

Personally, my aptitude was unexpressed until college. It was first-year computer programming (standard in Canadian engineering schools) that opened my eyes. Prior to this, I never really had the desire to program anything because no one had told me how fun it was. It was not encouraged by my parents (who didn't even own a computer for the longest time) or my teachers both of which did not know the first about programming or why you would ever want to do it.

I certainly think that the "talent shortage" could be reduced by making computer science mandatory in high school, or at least on equal footing with other sciences such as biology or physics.

I wonder how a simple pre-processing operation would affect the accuracies? In my experience, standardization (z-scoring) can really improve SVM accuracy.

Biomedical applications of machine learning have great potential, but they suffer from a lack of data problem. It's great to see Little championing for a more open database, hopefully it has a positive effect.

For more traditional machine learning research, there are common sets of data (i.e. MNIST for handwriting recognition) which serve to benchmark new algorithms.

The main problem with biomedical data is the difficulty of acquisition, and the fact that many researchers are afraid of discovering findings that they may have missed.

I'm impressed by his business savvy considering his young age, but he owes at least part of his fame to technology and social media, due to his YouTube start. And he's very rich. That helps.

Bring your money to Toronto! I'm currently living in Toronto, but looking to move to Seattle once I start at Microsoft. That being said, I'll be making about 70% more in Seattle than I could hope to make in Toronto.

Toronto has favourable taxes for small businesses, and is close to some top engineering / computer science schools in Canada and the world (University of Toronto, Waterloo, Queen's University, McMaster). All of my friends currently working on startups in Toronto (myself included) are all bootstrapped, as there's not a lot of VC money around here.