imh

66 ms·

661.
by imh 11y ago
Most languages have a single implementation that everyone uses. Unless I'm interested in writing a new implementation (never), then I'm going to be interested in comparisons of languages based on their well supported implementatio
662.
by imh 11y ago
I'm curious how they arrived at affordability as a function of income. I'm surprised they weren't transparent about that.
663.
by imh 11y ago
For other people like me, who didn't know what this was referring to, I think this is the post brudgers is talking about http://www.hanselman.com/blog/DarkMatterDevelopersTheUnseen9...
664.
by imh 11y ago
That's not the point. The more ads there are at the top, the more google is effectively allowing you to pay for good ranking. This violates a lot of the search neutrality that makes google so useful. I think of it like Yelp. I want the
665.
by imh 11y ago
Jesus, a search for business cards returns literally nothing but ads until I scroll down. Time turn ad block on for google.
666.
by imh 11y ago
Delete all the cookies.
667.
by imh 11y ago
> ...it does highlight that we have awesome new technology that creates a serious tension between two values we all treasure — privacy and safety. I hate this narrative so much. Just because we happen to be storing everything we do now,
668.
by imh 11y ago
Note that the video is from 2014 and the article from about a year ago. That said, I want this so badly.
669.
by imh 11y ago
It's in a sweet spot. It's good enough for unstructured exploratory analysis, for production code, it's super easy to get started in, and there are some good libraries. There's really low overhead to poking around data i
670.
by imh 11y ago
There are some really interesting dynamics to that. Especially owner occupied deductions that effectively make it more expensive for pure investors and landlords.
671.
by imh 11y ago
Nope, it's just 80ms multiplied by the number of possibilities.
672.
by imh 11y ago
It's not too high risk. Worst case, it becomes a pretty portfolio piece, which is useful enough.
673.
by imh 11y ago
They address this at the end of the letter. They say it's "an unprecedented use of the All Writs Act of 1789 to justify an expansion of its authority." They go on to talk about what that precedent would mean. It's at the
674.
by imh 11y ago
It's so weird how hard it is for the brain to handle exponential growth. I was amazed that a 4-digit password can be cracked so quickly at 80ms a pop, but you're right. Just for the hell of it, here's how long it would take f
675.
by imh 11y ago
Order here roughly refers to the idea that for any two suitors, you can say which is better (or that they're equal).
676.
by imh 11y ago
You say that's how other markets work, but I can't think of any other market where an increase in the price of something you already own can make owning it unaffordable. If I buy stocks, or gold, or a car, and it gets more valua
677.
by imh 11y ago
I thought Rust left the optimization to LLVM. Is that not the case?
678.
by imh 11y ago
>Here’s the thing: if I were the type of person who would be good at coding ... I don't like this sentiment. Programming (at least the basics, no monads here) is really simple, and the perception that it's difficult seems damag
679.
by imh 11y ago
I've only just started down this path, but http://www.theodinproject.com/courses seems pretty good so far. It's more a curated, guided collection of other resources than a resource itself, but that works for me.
680.
by imh 11y ago
Since the google thing doesn't work anymore, I deleted my cookies and it worked like a charm.
681.
by imh 11y ago
How does this work if the graph isn't connected? Like if there are people who are only friends with each other and no one outside their group. Surely out of a billion people, there must be some people like that on facebook.
682.
by imh 11y ago
What kind of info are the Teslas beaming back to HQ? It can't be the ultra high volume data google's collect, can it? This [0] says google's collect almost 1GB/s. Presumably they keep a good chunk of that. Does anyone kn
683.
by imh 11y ago
The counter that you made relies on transitions in the job market. That requires people learning new skills on large scales. But what if most people can't learn to do the new tasks as quickly as the top few can learn to automate them?
684.
by imh 11y ago
If this trounces the leading human this spring, I'll be really excited if the AI experts start working on physical games. The RoboCup (robot soccer) hasn't had nearly the the improvements that more logical games have (e.g. chess a
685.
by imh 11y ago
>We are living in a time of technological marvels, with advances like CRISPR gene-editing being used to bring back extinct mammoths; lithium-air batteries that store 10 times more energy than conventional lithium-ion batteries; mitochond
686.
by imh 11y ago
If you want more than a 4 lecture course, I recommend Nando de Freitas's course. It's very high quality and free. https://www.cs.ox.ac.uk/people/nando.defreitas/machinelearni...
687.
by imh 11y ago
I'd recommend really mastering basic statistics if you aren't going to go all the way with learning data analysis. It's surprisingly subtle and more widely applicable to a broad range of careers.
688.
by imh 11y ago
Because of things like liquidity preference, a valuation going down can wipe out the employee stock.
689.
by imh 11y ago
>Writing out monadic code explicitly gets to be a huge pain when using monads in another language. Couldn't any other language have a syntax like thing.do(\x -> other thing(x)).do( \x -> yet another thing(x)).do(
690.
by imh 11y ago
Raw probabilities are terrible for talking about this kind of thing. You are also probably less likely to be squished by a falling piano, but if you see one overhead, you're rational telling people to get out of the way.

More ›