joshuahhh

5 ms·

1.
by joshuahhh 3y ago
Impressive work! I wonder: is the inertial tracking sufficient to cover for occasional occlusion of the markers?
2.
by joshuahhh 8y ago
More details: https://books.google.com/books?id=fRqKreZFVTYC&pg=PA22&lpg=P...
3.
by joshuahhh 9y ago
I am delighted to have the chance to learn what the Hacker News community thinks about capitalism.
4.
by joshuahhh 9y ago
Pushes are failing for me. (To one repo, at least.)
5.
by joshuahhh 10y ago
I am no expert on this, but Wikipedia cites four sources for the claim that "further examination of the Perot vote in the Election Night exit polls ... showed that Perot siphoned votes nearly equally among Bush and Clinton" ( http
6.
by joshuahhh 10y ago
I second the criticism of using violin plots here. A violin plot is a kernel density plot. It is designed to show a distribution at a scale much larger than variations in the data. Its raison d'etre is to smooth out and aggregate these
7.
by joshuahhh 11y ago
http://comicsalliance.com/scott-adams-plannedchaos-sockpuppe... http://blog.dilbert.com/post/102881515756/planned-chaos
8.
by joshuahhh 11y ago
Pulling data from prediction markets into Guesstimate is an exciting idea. A few thoughts: * Prediction markets are usually for binary outcomes. I imagine the most useful role of binary variables in Guesstimate would be to mix two different
9.
by joshuahhh 11y ago
Thanks for the link! FWIW, part of what I am trying to do with this gallery is show the power of images to illustrate /concepts/ (or ideas, or systems, or models, ...), rather than just /data/. There is a lot of overlap
10.
by joshuahhh 12y ago
I am impressed by how quickly the notebook goes from raw data to interesting patterns. Nice. If I wanted to be picky about statistics (I generally do), I'd note: 1. Equating enforcement-related statistics with usage-related statistics
11.
by joshuahhh 12y ago
I'm surprised (and disheartened) to hear that psychedelics served as your gateway into harmful drug use. Can you tell a bit more about the role psychedelics played?
12.
Show HN: pandas-ply – cleaner data manipulation with pandas(github.com)
3 points by joshuahhh 12y ago|0 comments
13.
by joshuahhh 13y ago
That article says nothing about preloading pages. The speed gains are purely a UI thing -- if you're clever about keeping submenus open even when the mouse leaves the submenu's entry in the main menu, you don't need to provid
14.
by joshuahhh 13y ago
Wow -- you've described my situation to a T. Except that you decided to do something about it. I'm looking forward to trying this site and seeing if it works for me. Thanks!
15.
by joshuahhh 14y ago
I think you're trying to say something like "as long as developers are primarily men, marketing to developers means marketing to men". I see logic in that. But what you're leaving out is that the lack of women in tech fields is a serious pr
16.
by joshuahhh 14y ago
As much as I personally enjoy the demo the creator chose for the main page, I wonder whether said creator could have found a "flo-mo" view of something captivating for reasons other than "this chick is hot". (There's a really unpleasant his
17.
by joshuahhh 14y ago
It's actually quite useful to have talks like this spreading across the Net. They act like chemical tracers, flowing through the memetic channels of crackpottery -- once a blog picks one up, you know immediately who you can ignore.
18.
by joshuahhh 14y ago
Not gonna lie; the MindMaze theme brought me back in a powerful way.
19.
by joshuahhh 15y ago
I am sympathetic both to your comment, and those replying to it. I think what this suggests is that modern psychiatry does not have a rich enough understanding of the enormous variety of mood disorders involving depressed states. If it did,
20.
by joshuahhh 15y ago
This appears to be totally broken in Firefox (9.0.1, on Linux)... don't know what's up with that.