ramidarigaz

In some of his recent keynotes (especially the later quakecon ones) he focuses a lot on how he's changed his mind and that he now believes that software development has a lot of "social science" aspects to it. I think he's given a lot of consideration to how one manages a software team, and even though he sometimes disappears into his office to write something like this Netflix app, he still does manage people.

I'm in the same boat, mostly from weapon repairs. My income is pretty good, but I usually need to repair things after a quest, and the cost of repairing really puts a dent in my accumulated cash.

That seems to be slooowwwwwwly improving, but honestly the biggest improvement for me will be the light rail from downtown.

It was going much too fast to stop (given the available propellant). Also, that would have extended the transit time significantly. On the upside, it's sounding like the Kuiper belt is a really interesting place, so we'll get to see much more!

Getting under the surface of the ice might not be too bad, but getting a signal back up would definitely be.

This was linked elsewhere in the comments. I think it answers some of your questions:

http://lwn.net/SubscriberLink/644675/5be656c24083e53b/

Yeah, it definitely wasn't as impressive as it seemed on the surface. A much better test of Watson's performance would have been QuizBowl, where contestants can buzz in while the question is still being read.

Yeah. The coverage area of the transmitter must be massive.

My understanding is that it is going to burn up naturally unless they do something about it. Its orbit is low enough that it feels a fair amount of drag.

You can actually see the video feed blip out once per revolution. I assume it's because the transmitter is briefly pointing away from the receiver.

The lab I work for (http://lasp.colorado.edu/) has an instrument on board (MASCS). I worked on a 3D visualization tool for the data it produced.

As far as I know, it accomplished all its goals (and more!). Many of the findings, especially the ice in the polar craters, were completely unexpected, and that's the most exciting part of doing science!

I was actually fortunate enough to work on the MESSENGER mission for a while (in a small way). Sad to see it go, but it had a really good run.

This article has a nice overview of the things it accomplished.

http://www.colorado.edu/news/releases/2015/04/16/after-succe...

Good to know their SQL sanitization is working...

There are forks of Tomato that have been updated much more recently. I'm running the "Toastman" build of Tomato.

That metallurgy book looks awesome! I just picked up my blacksmithing hobby again recently, so this is very timely. Thanks!

The "sports or business section of the newspaper" is the internet these days...

Yep! Currently it has some graphics performance issues on Linux, but it's perfectly playable and they're actively trying to improve it.

Also, for anyone who has an Oculus Rift (or has access to one), Titans of Space does a really good job of depicting things like this.

Looks like they diverged. Lots of duplicate comments.

Edit: Nope. All comments show on both.

Worst case, I bet someone will make a chrome plugin to purge twitter results from search.

That sounds like a seriously awesome course. I wish my school would do that...

This discussion is happening quite often in the VR world. Essentially the accepted opinion is that 1) AR is harder than VR and 2) solving the problems that VR has will get us closer to usable AR

These are going to be LEO satellites (no higher than a few hundred km), instead of GEO satellites, which would make the lower bound of the latency pretty low (comparable to cable or DSL).

From this tweet, it looks like it was the residual fuel:

https://twitter.com/elonmusk/status/555980855537975296

Gosh yeah. I use Linux as my primary OS now on a cheap-ish laptop (student), and even with the low-DPI screen, so-so keyboard and odd hardware compatibility issues, I love it. I'd pay quite a bit for some high quality hardware.

It's not that they've been sitting in a hangar for decades so much as that we've gotten a lot better at building engines. Presumably modern designs have a lot lower catastrophic fault probability.

I haven't actually looked at the demographic information we've collected. Most of the work the group does has to do with memory and engagement, which varies enough person-to-person that their demographic doesn't matter so much.

I'm doing a Master's thesis in the intersection between Cognitive Science and Machine Learning. The group I work with looooves Mechanical Turk. It's such a cheap way to do small validation studies.

There's a bootleg stream of the event at: http://new.livestream.com/accounts/10419610/events/3475766

Except when it doesn't. I'm facing this problem with an embedded-ish project I'm working on where Python initially seemed like a good choice, but now we're regretting it for both performance and power usage reasons. If we'd gone with C or C++ initially we'd be having a much easier time optimizing.

Not that I'm implying Python is bad, but if your target use case doesn't allow for a desktop or server, think carefully about what tech you use.