dspeyer

Eliezer does not curate any web forum.

He says he is not planning to bomb any data centers, but he's also said that if he were planning to bomb data centers, he'd lie about it in public, so...

The author and most of his associates generally qualify as techno-utopians, and speak routinely of the "glorious transhumanist future".

Smarter-than-human AGI really is different from all previous technologies, in much the same way that homo sapiens is different from all preceding life on earth.

Off the top of my head...

    * A plan exists
    * The plan itself predicts non-obvious results in smaller systems, and those tests have passed (prediction written *before* running test)
    * A bunch of smart people have looked at it and said, "Yes, this looks plausible"

The closest thing to a plan is RLHF, which has failed every toy problem its been thrown at and made everyone in the field say "even if this worked in toy problems, it wouldn't generalize".

I have yet to see a large static typed program that didn't -- somewhere -- run into the limits of static typing and contain a set of workarounds, using void* or linguistic equivalent. That's code a dynamic language doesn't need.

The only code you can be sure isn't buggy is code that doesn't exist.

This isn't a question a non-engineer phone screener can ask. Coming up with first pass filters that don't require an engineer to interpret is harder.

I zoomed the image in and out a bunch -- no change. When I zoomed in enough, I could see the checkerboarding in A, but the overall color stayed the same as B.

I also had trouble calling either "more even". The first one has greater division at the darker end, and the second at the lighter, but they about balance out.

The extreme for me was figure 12. A and B are so similar I can't see the line between them, but C (the "corrected" square) is a completely different shade.

I'm viewing on a data projector. That's probably the reason. Still, it makes me skeptical that there's anything display-agnostic you can do for gamma.

Last time I manually packed structures I was doing GPU programming. I forget the details, but the CPU and GPU had different alignment requirements so anything other than a manually packed structure broke in weird ways.

Is there a good reason for making day 42 (out of 84) the primary endpoint? Did they at least decide to do that before looking at the data?

Because passwords are terrible.

Lots of users pick guessable ones unless you have extensive password rules to stop them, and those rules are a huge pain to your users.

Lots of users have terrible password hygiene, leaving them all sorts of places they shouldn't.

Enough users are going to forget their passwords that you need a recovery mechanism. Which means an attacker needs to break either the password or the recovery. The most common recovery method is email, but Google often is the email provider.

I suspect most people who don't use a password manager reuse passwords. A lot.

If you get the password to one of my unimportant accounts, you get the password to most of them. They're unimportant, but still...

Not so easily. If you're trying to learn really difficult things, classes are a huge help.

Often they don't. Maybe by nominative tuition, but not by median.

They don't mention Researcher Allegiance Effect. Therefore I suspect they did nothing about it. Given the nature of the study, I feel comfortable ignoring the results on that alone, without bothering to find specific methodological flaws.

Rename doesn't always do what we want. The one serious investigation I've seen [1] unhelpfully states that "[W]hen a file is appended, and then renamed ... many file systems recognize it and allocate blocks immediately." Suggesting that some file systems don't, but failing to name them. In general, appends followed by directory operations are not order-preserved.

[1] https://www.usenix.org/system/files/conference/osdi14/osdi14...

The timeline:

    1934: This is predicted
    1997: This is done http://www.nytimes.com/1997/09/16/science/scientists-use-light-to-create-particles.html
    2014: We will "demonstrate the feat within the next 12 months"
    2015: Didn't bother

First thought:

The usual pattern is that only one X chromosome gets transcribed in any given cell.[1] This keeps the dosage of all those proteins correct. Otherwise there would need to be a separate set of dosage controls for XX and XY people.

The inactivation occurs pretty early in life, and when cells replicate they keep the inactivation. This results in macroscopic regions of the body with consistent inactivation. Usually this isn't noticeable, but in cats coat pigment is on the X chromosome, and heterozygous cats often show "tortoiseshell" or "calico" coloring. The patches of contiguous color there are larger than a retina.

So it seems entirely likely that all the cells in a human woman's retina would use the same X chromosome.

On the other hand, that predicts half of het women would be colorblind, which isn't observed. So maybe not so simple...

I have a second thought about downstream neural hookups and the mechanisms for those, but it'll have to wait for later.

[1] https://en.wikipedia.org/wiki/X-inactivation

When those senior doctors are out of town, are risky-but-necessary procedures performed by less senior doctors or delayed until the senior ones return?

Actual descriptions of medical tests routinely give both rates. They often call them "sensitivity" and "specificity". Good luck remembering which is which.

But if only one rate is given, that indicates they're equal. If they're not, then it's reasonable to describe the documentation as incorrect.

The question is can disrupted comets or oorts block over 20% of the light from the star. A comet nucleus is a lot smaller than a star.

From the article:

> I’m angry, and I’m a little upset with myself because I want to be useful.

I'm thinking the same thing. What I'm wishing is that I held a position of influence at a private high school, so I could offer him immediate entrance and a full scholarship. Mostly to correct the injustice, but also because he's clearly a top student: unassisted home-made clock trumps perfect SSATs.

So if anybody reading this has such a position -- even a little bit -- please try to make it happen.

The case in Boston a while back involved a white victim. Granted, that was TSA, not the school, but it seems like a pretty similar group of idiots.

https://boingboing.net/2007/09/21/mit-student-arrested.html

As I understand it, eigenfaces require preprocessing to orient and align the images. That doesn't seem to have come up here. Maybe Amazon is really consistent about centering models in dress photos?

That's there. Search for:

> The misclassifications are interesting too

One problem seems to be that it concluded she'd dislike anything the exact opposite color from her favorite shade of red. A common flaw in linear models.

What do we know about Sundar? It sounds like he's really taking over Google. Is he likely to be up to it?

Bored and overwhelmed.

At least a few years ago Larry really wanted to understand everything Google did. It wasn't possible, and the attempt was distorting things.

This is a better solution.

Or so that once there's good news from a spinoff, they'll be able to announce it. A small profit there would get lost in the bigger revenue streams under the old model. Remember how the media kept running "How long can Google afford to subsidize Youtube?" stories well after Youtube turned profitable?

It is weird that this exact phrase kept getting used against the Irish and not other discriminated-against groups. Nothing for Poles, Slavs, Italians, Russians... Or at least that's what I'm finding on the LoC search. A few "No Jews Need Apply", but only a few, and those aren't job listings.

Probably not a particularly important mystery, but weird.

So we want to insert a value. We find an empty slot. We find the bucket it's in. Do we have to realloc the bucket and move all downstream data down by one? That sounds slow.

I notice that once you go down the "is it for a legitimate business" branch, the answers are all "NO". They could have omitted all the follow-ups.