Niten

The new test was at slightly lower power but within an order of magnitude of the original test: 700 W versus 2.5 kW. The measured effect, on the other hand, has decreased by more than four orders of magnitude.

I haven't studied the theory behind the device, but it's hardly a distortion to characterize such a disproportionate drop as a declining effect size in the face of additional testing.

Like you, I don't claim to have any idea whether this ultimately pans out. As I said, the ultimate test is to put one in space and see what it does. Until that (or at least a more promising lab test) happens, all we have is speculation. But my point is that more skepticism is warranted than was implied by the Wired headline and by some of the comments in this thread.

It's good to see that someone finally got around to testing this in a vacuum. Here's the paper mentioned in this story:

http://arc.aiaa.org/doi/abs/10.2514/6.2015-4083

High levels of skepticism are still warranted, as much as I would love for this to pan out. Every time this gets tested—presumably more and more carefully—the measured force gets smaller. The original test at Chinese Northwestern Polytechnical University measured 720 mN; this latest test in a vacuum chamber, only 20 μN. This is so tiny that a test anomaly would seem a much more plausible explanation than a violation of the conservation of momentum.

Of course the ultimate test would be to put one of these in space and try accelerating something with it, in the same way that the ultimate test of a perpetual motion machine would be to make one perform macro-scale net work.

Neat. It would be interesting to see a comparison between this and FreeBuilder (as well as AutoValue, though I think the differences there are more obvious).

I'm all for tools that make immutability easier; the lack of immutable objects is one of my main sources of cognitive dissonance when programming in Go instead of Java.

Much of this controversy is driven by the community's loss of trust in the site's administration. Do you think the tactic you suggest would be constructive toward rebuilding that trust?

To clarify, I meant the "overhead" of defining and instantiating a container class for RAII purposes.

I hadn't considered the type of overhead you're talking about, which is admittedly more important for the kind of software that tends to get written in C.

> The addition of some orderly construct for this in C would eliminate that case, leaving no real role for goto there either.

I like the way this is handled in Go with the defer keyword: https://blog.golang.org/defer-panic-and-recover

This construct gives you most of the power of C++ RAII without the overhead. Except that you can't use it to cleanup resources after exiting an anonymous block—it strictly defers to function exit.

I think the biggest issue is that any general discussion site which simply clones reddit's conversation model (e.g., voat) might do better at transparency and free speech and community relations and so on, but is still probably doomed to end up with the same middlebrow discourse and groupthink that exists on reddit.

Does there exist good, empirical research on how different discussion models (linear vs. trees of comments, up/down-voting systems, etc.) affect quality of discourse, for some definition of quality?

Right, anyone who disagrees with Reddit's management decisions is deserving of hyperbolic abuse.

That's irrelevant, however. If systemd is going to ship with a default nameserver then it should be using the correct and public NTP pool, rather than servers from a random company who hasn't made any commitment to providing reliable UTC time to a public audience.

And if it's using a service like the NTP pool, it needs to have a vendor prefix. Not because this has anything at all to do with the notion of being a "distribution", but because the prefix provides a means for the pool to identify traffic originating from default-configured systemd instances, and cut it off if it suddenly starts to behave in a way that would degrade the rest of the network.

Or they can simply ship with no default. But you can't have it both ways, both setting a default server and claiming not to be a vendor in every sense meaningful to an NTP server operator.

No, it's a huge difference.

I'm willing to trust this one, specific company with my data because I know their privacy policy and their practices. Handing that data off to arbitrary other companies for a buck is another matter entirely, I wouldn't be comfortable with that; that's what "selling your data" means, that phrase has a specific meaning, and claiming that's what Google does is a misrepresentation.

What you're saying is that if I tell my friend Sean a secret in confidence, it's all the same whether or not he proceeds to share it with all his friends. In my mind, and in most people's minds, it isn't.

Google doesn't sell your data:

http://www.google.com/policies/privacy/#nosharing

Rather, its business model is using this data on behalf of advertisers. That is a highly significant distinction, in my personal opinion.

Here's an overview:

http://static.googleusercontent.com/media/research.google.co...

Do you mean that negative comments tend to receive downvotes, or are deleted by HN admins?

There was a great multiplayer Dreamcast game called Chu Chu Rocket which I'd put in this genre.

> When the news first broke a few years ago, given Toyota's reputation for quality and process, I thought this was an American industry lead witch-hunt of a Japanese competitor. But if this testimony is correct, what Toyota engineers have done is unforgivable.

These two possibilities aren't mutually exclusive.

Software engineering practices aside, was it ever demonstrated that unintended acceleration could actually arise from the behavior of this software?

Divide the hourly cost of a human railroad engineer by the total value of the goods they transport by train. Then divide the hourly cost of a truck driver by the total value of goods that can fit into a single truck. The latter ratio is orders greater, and that means there's a much higher economic incentive to replace the truck driver than there ever will be to replace railroad engineers. And I would wager this difference in incentives widens when the amount of damage done by inattentive truck drivers vs. inattentive engine drivers is taken under consideration.

Besides, operators will probably desire a small crew to watch over their trains anyway for the foreseeable future, regardless of whether a human is in the pilot's seat.

Agreed, of course. But I still think it's a good idea to eliminate one opportunity for bias if we can.

Well it wouldn't be that outlandish to do truly blind programming interviews.

Just use a collaborative editor and possibly wacom stylus (can't be any more cumbersome than the traditional whiteboard), plus a text chat interface to conduct the interview. The interviewer interacts with the candidate in real time but doesn't see him or her.

Where cheating is a concern you can have a different employee (who does not provide interview feedback) monitoring the candidate over a camera to ensure they're only using the approved terminal, not phoning a friend.

Any medium to large-sized software company should be able to pull this off, and it would eliminate the possibility of gender bias in the interview results.

For what it's worth:

https://docs.oracle.com/javase/8/docs/api/java/util/Optional...

> The data is certainly persisting somewhere outside of Firefox proper, as the search app is siphoning data that can only be cleared from within.

That does not follow. They are the same application.

For example, attempt to reproduce this after pressing the Clear Data and Clear Cache buttons on Firefox's Android app info screen.

This may also be a bug in the Firefox app: I've just observed that when I change my default search engine in Firefox, that setting isn't necessarily reflected in the search intent handler until after I force stop Firefox.

Either way, to reiterate, it's definitely Firefox you're seeing here. If you don't like the way it behaves, I recommend following the other suggestions on this thread to file a bug report with Mozilla.

It uses whatever search engine you have configured in Firefox. When I tried it a minute ago on my Nexus, this defaulted to Yahoo.

This is unequivocally Mozilla code you're observing. Here's what adb logcat shows when that activity starts:

    I/ActivityManager(29188): START u0 {act=android.intent.action.ASSIST flg=0x13000000 cmp=org.mozilla.firefox/org.mozilla.search.SearchActivity (has extras)}

No it does not, in fact it doesn't even work the way the author claimed. He or she mistook an activity bundled with the Firefox app for the Google Search app.

In reality that search history was never persisted outside the Firefox app.

Same here, and sorry if that seemed hyperbolic. It's just extremely frustrating to me as someone who sees (most of) the ACLU's work as extremely important, and who has donated to them in the past.

I used to work for a school system that experimented with gendered classrooms. While it would be a mistake to extrapolate too much from our experience, those became the most successful classrooms at the elementary school in question. The boys did particularly well.

Naturally the ACLU would rather fight to ban the practice outright than risk additional research possibly showing that this is a productive approach, at the expense of the preferred gender politics narrative.

> Are these kids hacking on the Linux kernel or something? Of course not.

Not most of them and not to begin with, but "educating themselves" means having a platform where their curiosity can run amok, and some kids with RPis absolutely will find the Eudyptula Challenge or some other source of inspiration and make it to tinkering with the kernel.

RPi is absolutely the perfect platform for this. Break your OS? Just throw in another SD card. Somehow damage the hardware? Whatever, it's only $35.

I'm not the only person on here who taught himself Z80 assembly language in school because it let me dig deeper into our TI calculators; if we'd been using RPis for programming assignments I would have been all over that. I think it's a little surprising that it even needs to be said on a site called "Hacker News", but let's not underestimate what curious kids with a bunch of free time will accomplish.

The Linux community uses it heavily.

Two things in this article stuck out to me:

> Here is a humanist proposition for the age of Google: The processing of information is not the highest aim to which the human spirit can aspire

That's an uncharitable paraphrasing of Google's mission statement which, taken to its logical conclusion, might as well say "To know everything there is to know in the universe, to grant agency to this knowledge, and to make it available to everyone." Maybe this just reveals a fundamental rift between my values and those of the author, but I cannot think of a higher calling for the human spirit.

> Searches for keywords will not provide contexts for keywords. Patterns that are revealed by searches will not identify their own causes and reasons. The new order will not relieve us of the old burdens, and the old pleasures, of erudition and interpretation.

The author is making an empirical claim here, and it is wrong. AI is already advanced enough that such absolutes as "searches for keywords will not provide contexts for keywords" are demonstrably false. To a limited but growing extent, software is already capable of erudition and interpretation; these are not uniquely human endeavors.

The biggest problem with this article (and so many others like it) is tying an argument for the value of the humanities with unscientific denial of materialism. That the human mind is ultimately an emergent property of matter does not make art and literature any less worthy of study.

Qt comes with a built-in event loop which can be really nice for certain types of asynchronous servers.

> Yeah, creating a safe environment for discourse is super-awesome.

You have a strange definition of discourse if all that's required to create a non-"safe environment" is to present a dissenting opinion. (Did you mean: echo chamber?)

> Filled with dishonesty, zero empathy and zero good faith.

That's a rich complaint coming from someone who called me a misogynist and awful person for the crime of mentioning an article by an actual woman in technology, with her own unconventional opinion on how to ultimately make things better for other women in technology. (Maybe she's right, maybe she's wrong, I sure as hell don't know and that's not the point—the article was censored by a mod just after I and a few other people started discussing it, so that conversation never even got to happen.)

I have every reason to believe that Susan Sons wrote her article in good faith. I sure as heck wanted to discuss it in good faith. So seriously, you should be ashamed of yourself for calling me a misogynist for caring about women in technology. People who derail respectful dialogue by going nonlinear with appeals to emotion and unfounded name-calling, like you have here, are the reason important topics like this one rarely see the light of open civil discourse.

Please carefully examine your reaction on this thread and reconsider how you have chosen to conduct yourself.