748 karma · joined November 2, 2014
Even if you don't retain the copyright to the camera-ready version of your paper, you can just redistribute a preprint version that is almost identical (on arXiv and/or your website).
A "global language" is by definition a language learned by significant numbers of people as a second language [1].
By most people's accounting, only English, French, and Spanish are global languages. "Minor" global languages on the rise include Mandarin, Arabic, and Hindi, but these do not have nearly the geographical distribution nor the populations of second language learners that the main global languages have.
That said, unless you have a partner or other significant connections to your target language community, consume a large amount of target language media, or are living in a country that speaks your target language, speaking to other people is not the main motivation to learn a language. You learn a foreign language because it's fun and because there are numerous studies indicating knowing a foreign language benefits your mental abilities and health.
Why focus money on developing a hyperloop++, when we could have something almost as good with technology that currently exists?
Don't forget that Scotland very nearly did leave the UK, with the government's consent, in 2014 (but the referendum didn't pass).
The Catalonia case is a bit murkier, since the Catalan government did not actually reach an agreement with the Spanish government before they went ahead with the referendum. Also because of irregularities seen by international observers during the referendum, the Catalonia referendum result was generally not recognized.
That said, the architecture was actually quite nice, with lots of natural light and open space.
Of course, there are immense problems with this. For one thing, it's usually crippled by latency. Even if the client X application is in a Docker container on the same machine, applications are only halfway usable. Use a local Ethernet network, and things start to fall apart. Going over the Internet isn't really worth trying.
The other major problem is that there's no security. If you have an open X server, anyone can connect to it and run random applications on your display. But in addition to opening you up to annoyance from cyber-hooligans, allowing an X application to run gives it access to all your keystrokes, which may not be something you want. Beyond that, all the X data is going out over the network unencrypted. There's just no security model here.
That said, it would be cool if there were some kind of plugin for Wayland that would let you use this feature (or even something better, if someone wanted to implement that). I've wondered---can you set up XWayland to be an open X server and just let random X applications connect to it? It would be interesting to hear from Wayland experts on this point.
The opinion and longform side of the paper are another matter, though. Almost all the columnists are pro-Beijing hacks at this point. It's kind of like how there's a divide at Fox between the "news" side, which is relatively reasonable (e.g. with Chris Wallace), and the "opinion" side (e.g. Bill O'Reilly an his ilk). While the news folk at these kinds of places might be principled journalists, their ultimate purpose is to legitimize the propagandists on the opinion side.
To figure that out, you'd need to do a speed test to the target host, then do another one while simultaneously running Speedtest.
Your Internet speed is also affected by the peering agreements your ISP has, i.e. how "well connected" it is to the ISP of your target host. Speedtest targets (and Fast.com) are pretty well-connected; i.e. they have a lot of peers, so you can get to them in relatively few hops. Many times, they're even run by your ISP, so the number of hops is minimal.
I've noticed there's a big difference in both bandwidth and latency based on the network my requests actually go out on (from my university Internet connection). For some reason, the Univ. of Cal. Network > Commercial ISP > Internet2. (Not sure why Internet2 sucks so much, but it seems really slow for some reason.)
Keeping one in your house would be ridiculous and loud, though. Mine actually required two power cables! I ended up pawning it off to a friend for $20, who pawned it off to one of his friends for $20, and so forth. I wonder where it is now.
I unfortunately decided to install RW_LEGACY years ago when I set it up. It's now in a weird state where it has SeaBIOS installed but legacy boot is disabled. From what I've read, I need to boot into a Chrome OS recovery to enable legacy boot, which should allow everything to work. But I haven't been able to find a simple Chrome OS recovery shell image for me to boot into, so I'm kind of stuck rn.
I used Arch Linux on a Toshiba Chromebook 2 as a primary laptop for about two years, but eventually moved away from it because my bootloader got un-blessed due to the battery zero'ing out, and I didn't want to deal with it.
I guess Chromebooks are back on my radar!
I never seem to be able to get hardware acceleration to work properly in Linux on my MacBook---whenever I open up a webpage, CPU usage goes through the roof, unlike in Mac OS. Wayland seems to help a bit, but there seem to be lots of bugs than can only be fixed by the window manager.
1) Tor exit node operators will do all kinds of nasty things: e.g. MITM-ing your sessions, refusing to handle certain kinds of traffic, and recording everything.
2) If you use your regular SSH key (which you will, because you needed to use this script to do this), your SSH key fingerprint and thus your identity is still recorded by the remote host. Doh!
3) Everyone at the remote host will think you got haxed. Because you're logging in through Tor. Or else they'll think you're super shady.
That said, Tor is super fun to play around with. Just don't assume anything gives you real protection unless you're careful and really know the ins and outs.
Damn, poor guy.
> "When training, a screw nut approximately 3% of their mean body weight was tied to the top end of the tail of each rat, and all rats were kept in the special pools with water of 60 cm depth (water temperature 25–30°C, pH 6.5–7.0) until fatigued (submerged below the surface for five seconds twice). The fatigued rats ceased training immediately, were removed from the water for a short break, showered with running water and then dried with hair dryers."
So, systematically drowning rats until they aspirate water, every day for twelve weeks. And they showed negative respiratory effects. Huh.
If you swim in a well-maintained pool (e.g. one of my university's competition swimming pools), there's no "pool smell."
That said, I don't really know anything firsthand about the architecture.
Any references? I'm curious.
Even if you were going to be executed anyway, you obviously have a right to determine what happens to your organs, particularly in cultures like China's, where it's seen as very important that the body remain whole.
Chromium disables hardware acceleration by default for a wide swatch of graphics drivers too (including all Nvidia GPUs).
However, there's a more important medical reason: acetaminophen accentuates the analgesic affects of opioids, although the mechanism for this is not clearly understood. This makes acetaminophen-enhanced opioids more effective drugs.
Opioid abuse has become so rampant, though, that we're moving away from prescribing acetaminophen-enhanced opioids like Vicodin and Percocet to try to stop them entering the recreational drug market. As opioid use has increased, so has the number of liver toxicity deaths due to the acetamiophen in some prescription opioid drugs.
There is language in the deal that would give China control over the port if Kenya was unable to pay back the loan. While concerning, this is very different from that actually happening.
Presumably they could handle the radio hand-off very precisely. ;)
But actually--the precision of an atomic clock (nanoseconds) is orders of magnitude higher than lag due to radio wave travel (many microseconds). And even if you account for this, you probably only know your distance from the atomic clock to within 10 meters, which is tens of nanoseconds of uncertainty.
Matt Pharr's posts on ispc are an extremely interesting read, both from a technical perspective (programming abstraction and hardware design) and a human perspective (Intel politics, managing style, and career decisions).
This doesn't require a parser--you just have to know what the enclosers are (one of the things that goes in a reple config file) and then count them to see if they're matched or not.
For languages without Python-like syntax or with enclosers that can be ambiguous (some dialects of Fortran), this would be pretty difficult.