Good question. I'm not sure, but I suspect it shouldn't do anything interesting; mercury allows oxygen to move through the otherwise impermeable aluminium oxide crust and "rust" the metal.
Since sapphire is already 100% aluminium oxide, but arranged in a crystal lattice I think it should be stable in contact with mercury unless the mercury disrupts the lattice (a possibility--my physical chemistry is weak).
Often data contains structural features that initially aren't observed and without them appear to support a hypothesis only weakly. By exploiting the structure, we can see things much more clearly.
An example: say we want to know how a drug affects cognition. We give a simple test to a bunch of people on and off it, blinded, etc. The control group's average score is 74, and the test group's average score is 72. We can use a t test to see if there's a statistical difference, and find there isn't. We can't conclude anything about the drug.
Now imagine we have exactly that same data, but we were careful to give two tests to each person (in a random order, and different tests, of course). We take another look at the data and find out that every single participant scored lower when they were on the drug. With even a fairly small sample size this provides strong evidence that the drug impairs cognition, and probably tells us quite a bit about how much it does.
The article is probably talking about multivariate regression; the more important number comes a few sentences later---"retention was related more strongly to manager quality than to seniority, performance, tenure, or promotions". So presumably, they did the same sort of analysis, carefully pairing people who were similar in as many ways as possible, and found out that good managers are more important than seniority in terms of employee retention. The more variables you have, the more even large differences can hide in raw group averages.
Your last sentence answers the question you pose: the new UI is attempting to train users to use the browser search bar.
A huge number of Chrome users go to google.com to search for things; if the new page looks like the Google search page, and it moves your cursor to the address bar, Google hopes these users will learn that you can just search from there. Maybe it's an experiment that will fail, but it seems like it might work.
> they could even make ad targeting work by moving their mail-reading bots to the client
Gmail ad targeting isn't a bit of JavaScript that can run in real time in the background. It's a series of huge map reduces touching data sets larger than any client computer could store. Just indexing a pre-prepared list of mothership ads would result in a horrible user experience (far poorer ad targeting).
Google has a vested interest in only showing you ads you might be interested in; it doesn't work as well as it should, but it works magnitudes better than the state of the art 10 years ago.
I can easily imagine an authority issuing a warrant to rsync.net without knowing that there is a canary in place. It seems less likely for an NSL with an attached gag order, but still possible.
Nothing in the names policy prohibits pseudonyms. If your given name is "Jason Ramirez", you're welcome to have a separate Google+ account with the name "Nancy Young".
You only have to prove people know you by a name if you want to go by "#RS", "Albert Einstein", or "GreenLife Rx"; all three of those are likely name violations for reasons that have absolutely nothing to do with knowing who you are.
This would be much more interesting if it asked you about symptoms of sleep deficit. E.g., "if you go into a dark room and lie down in the middle of the day, how long will it take you to fall asleep?"
Someone who habitually sleeps 6 hours a night but never feels tired is probably fine. Someone who habitually sleeps 8 hours a night but always feels tired should try sleeping more, and if that doesn't work they should seek advice from a doctor.
I expect there is a large segment of people who suffer from sleep deficit and would be more effective if they slept more, but don't know that it would help them. I don't think this is the correct way to reach such people, or the correct message to give them.
A few kg of blueberries a day? All my food for one day together weighs less than that. Did you mean grams? That seems low, but kg/week is still way too high.
How on earth can we usefully generalize from the experiences of white men born ~1918 who attended Harvard? I think longitudinal studies are massively important, and I'm all in favor of them, things like:
"Alcoholism was the main cause of divorce between the Grant Study men and their wives"
Don't really give us that much of a clue about how alcoholism (for instance) affects everyone else. I suspect it's similar; but this doesn't give us any data on anyone who isn't in a very very specific group.
This is after his first vocal cord was paralyzed, but before the second one was also damaged. You can hear what he called "a slightly weaker voice than normal which some people think sounded a little funny"--it quavers a bit. I was hoping to hear his voice before the first injury, but that was before Google was a phenomenon so there might not be any public recordings of him.
The objective is same-day arrival, and it works pretty reliably as long as you order by 3pm or so (in my area, at least). More importantly, you're given a clear indication when you order when you will actually get it (you can choose a delivery window, and see which ones are greyed out).
The delivery windows go as late as 9pm, so I'm sure they will try to push back the last time at which you can order and still get it delivered same-day.
Without testing, I'm not sure, but I think Mini Kites is actually likely to be a flagged name. The last vestiges of the "real names" policy are really just to keep businesses from getting Google+ profiles (Google wants businesses to keep Google+ pages instead). So there's an ML algorithm that determines how name-y or business-y a given name is, and flags names that it thinks belong to companies rather than people. I suspect Mini Kites (and almost certainly MiniKites or Mini-Kites) would trigger the review process.
Except for #3, I don't think these measures could be considered intentional. However, I imagine they have a large effect and go a long way to explain the difference between SF (or NY) and most of the midwest/mountain/central US.
No, but I work at Google, and I'm interested, so I've had lots of chances to play with it and talk with people who are on the team or are in the dogfood.
ETA response to the second question: I'm very happy with the display. That's probably as much as I should say, and I don't know the actual details anyway.
The display in Glass is incredibly unobtrusive. Glancing at a direction arrow (for maps) is no more distracting than glancing at the speedometer--it's less distance for the eye to cover, and the displayed images are ruthlessly simplified so that absorbing them is almost subconscious.
I think the danger comes from people who get text messages while they're driving and want to look at a shared photo. That should be disabled while driving, which glass can make a pretty good guess at.
Every single person I have talked to who has used glass for more than a few days has found the instant camera to be the most compelling feature of Glass. The author of this article did as well! In that context, I can't see how it makes sense to offer a camera-less version, at least not at launch.
I'm frustrated at how frequently gwern is assumed to be male. If you aren't sure of someone's gender, please use gender-neutral pronouns. So many of her articles have been posted to HN lately, yet I've only seen masculine pronouns used to refer to her.
When I was hired, I had a list of five managers to interview with. During the first two weeks after I started, I talked with each of them and decided where I wanted to be placed. It was made clear to me that if I didn't like any of those five, I would get another list.
Just my experience; I started at the beginning of 2012.
Since sapphire is already 100% aluminium oxide, but arranged in a crystal lattice I think it should be stable in contact with mercury unless the mercury disrupts the lattice (a possibility--my physical chemistry is weak).