I was introduced through his "Boundaries" talk [1] and had a very similar experience. I think a lot of programmers develop this methodology without really formalizing it. But practicing it consistently creates really nice code and easy unit testing. The one possible downside is long argument lists to functions.
It's true that the columnar inhibition is not the best supported part of the theory but seems a very strong deduction based on many papers that don't directly support but show the same behavior as this part of the theory. We know that minicolumns exist and cells in minicolumns share receptive fields. But they aren't always active together and HTM theory provides a hypothesis for why. This very recent preprint paper out of Michael Berry's lab at Princeton shows almost identical behavior that HTM sequence memory would predict and I'm not aware of other theories that would have predicted this behavior:
Compared with cash, digital currencies are much harder to launder. It will be obvious when you try to cash out that the money went through a tumbler. With cash, you always have plausible deniability unless you were physically seen receiving it in some illegal manner.
"We’re looking for one researcher who wants to work full-time... who are driven and talented but come from poor backgrounds." (please read the entire paragraph to understand the context of this)
The combination of hiring researchers that come from poor background with the author's stated belief that a basic income would likely be a good thing makes the entire set up seem very biased.
I like the idea of promoting research into the subject but I don't think this is the right way to do economic research.
I bought the TechMatte 2x USB-C to Micro-USB item that he gave two stars for use with my 5X. At least one of the items I got was actually defective. The other worked but charged very slowly. I wish I had seen his post before buying!
Now I am nervous because I subsequently bought a cable that he hasn't reviewed yet...
This has me wondering if there is some way to incentivize good reviewers. There is Consumer Reports and some more tech-specific sites but there is still a long tail of products that you have to buy without good reviews.
I think the right avenue is to keep the license GPL but provide a permissive as-is use license. The idea is that you want people making improvements to the algorithm to contribute those back for the public's benefit rather than charging people for their "upgraded" version. A separate unmodified-use license could be much more permissive in the use of algorithm as-is for compressing and decompressing images.
Yes same Scott. Unfortunately my first installation attempt for NuStudio was unsuccessful but I will try again soon. David has been very involved and it is great to see projects like that that make it more accessible.
I am a bit surprised to see this on HN without context. This was originally an email from Jeff to our mailing list and was a pretty good overview so we dumped it into a wiki page. For those new to the problem, the page summarizes it well:
> The basic idea of temporal pooling is patterns that occur adjacent in time probably have a common underlying cause and therefore the brain forms a stable representation for a series of input patterns.
This is a Bitcoin-like, distributed, secure payment system WITH a federated protocol for gateways/exchanges. A federated exchange network allows anyone to participate, similar to email servers. But since trust is so much more important in this case, I think that network effects will result in a relatively small set of gateways that anyone will actually interact with.
This is not a particularly informative article but I see great value in owning a Twitter botnet. As one example, imagine negative news about a major public company starts trending. That would certainly be a money-making opportunity. There are also hard-to-disprove rumors that you could blackmail high profile people with. Celebrities probably don't want to be trending on Twitter for having an affair, even if it is a fake report.
So I am not at all surprised that people are investing in building up tons of what appear to be legit, active accounts.
Those ideas would both lose valuable aspects of the buses.
Uber/Lyft/cars would have a much bigger impact on traffic on 101, which is already terrible congested during rush hour. The buses help to minimize the impact of commuters.
Public commuter options eliminate some of the convenience for two reasons. First, Google has a single destination for everyone which reduces commute times. As you increase the number of destinations, you increase the possible routes exponentially, making it much less feasible to have direct routes everywhere. Second, you can't do confidential work when on public transportation. So you would be much more limited in what you can accomplish during the commute.
I doubt Google wants to keep this Chrome-only. Just like SPDY, there isn't initially support on all platforms/browsers so you have to have a fallback. But if it catches on then you will see broader support over time.
Anyone know if NFC emulation will let me create an app to replace my Clipper card and work key fob? I would have guessed that NFC chips would have an embedded secret that you can't get directly.
I found this quite useful since he mentioned useful apps I hadn't heard of and some OS features I wasn't aware of. And I really doubt that many iPhone users are aware of many of these things.
I am not sure how the logistics work with participating stores but I would really like to see them make this an easy way for small local businesses to compete with Amazon. Basically a platform for them to list their products and the logistics to collect and deliver them.
This service seems to be mostly orthogonal to Instacart in that Instacart provides perishables but you can only order from a single shop at a time where with Express you can order products from across multiple stores but they focus more on non-grocery items.
I had the same impression. I have been working on a very early prototype of something very similar to this and was curious the know the general architecture of the system but there doesn't seem to be any documentation covering that.
I wish that more of these projects were more general like Mesos as opposed to just replicating a service across machines.
It is great to see more interest in neural networks, but the types of neural networks the author describes are missing some key aspects of what the brain is doing. I work with Jeff Hawkins at Numenta, and while our product is based on a type of neural network, it is quite different from the class of NN described in this post. For background:
It isn't an oversight by Facebook - it is by design. Facebook was a part of the decision to use Facebook login credentials to log into Spotify. Additionally, Facebook does not list access to your friend list (and your friend's email addresses) in their list of permissions. Rather, those details are implicit in using Facebook to authenticate.
As an example, using FB to authenticate with Quora does not list access to friends list in the permissions but Quora will send an email to every friend of yours already on Quora to notify them that you joined.
Another issue with this is that if you have never given any Facebook permission to Blizzard, but happen to use the same email address as listed on your Facebook account then Blizzard will attach your real name to your account without your permission.
[1] https://www.destroyallsoftware.com/talks/boundaries