HNHacker News
TopNewBestAskShowJobs

YjSe2GMQ

419 karma · joined January 15, 2019

submissionscomments
YjSe2GMQ··on 2FA on the command line
The controversy of this tool vs Google Authenticator essentially boils down to how secure your laptop/workstation is vs how secure your phone is.

Without being purely "conservative", i.e. it was first done on phones therefore it should stay on phones, what are the actual arguments for one vs the other? What do phones have that laptops don't, and vice versa?

YjSe2GMQ··on What you need to know before you join a startup
This comment is probably to be downvoted as off topic, but in case anyone missed this gem.

Here Comes Another Bubble: https://youtu.be/I6IQ_FOCE6I

YjSe2GMQ··on Ask HN: How to be productive with big existing code base
Disclaimer: I don't know anything about managing a team, this advice is more for a solo developer or a developer joining a team.

#2 is what I use. After one/two hundred tiny commits (even if it's sometimes just fixing grammar in comments) you'll feel more at home. And because they're small you can achieve that within a few weeks.

This requires any way of telling that things still work. Sometimes you'll be able to do that because the transformation is an identity given the language definition, but for other cases I'd encourage #7. Having someone more senior (with respect to the codebase) review changes helps (this is what I mostly used), but I understand you don't have that option.

YjSe2GMQ··on Leading with Wikipedia: A brand proposal for 2030
This is essentially about cleaning up debt (though brand debt, rather than technical). Read atdt's comment: https://news.ycombinator.com/item?id=19259543
YjSe2GMQ··on Humans who are not concentrating are not general intelligences
Oh well there's a misunderstanding. Conscientiousness is the psychometric term for self-discipline (not 100% the same, but extremely close): https://en.m.wikipedia.org/wiki/Conscientiousness

And I didn't bother to point out particular studies validating my claims because much of this has been known to humans for close to a 100 years. In the same vein as linking to a proof of the undecidability of the halting problem would be excess references given the nature of HN community.

If you want an example study look at the "Health and longevity" section on the above linked Wikipedia page.

Or to directly support my thesis about X1 and X2 separation, first look at Wikipedia IQ page and the "Social correlations" section, it's generally around 0.5: https://en.m.wikipedia.org/wiki/Intelligence_quotient

Or look at your own link.

Second, height is correlated to about 0.29 with various measures of success: https://www.researchgate.net/profile/Daniel_Cable/publicatio...

YjSe2GMQ··on Humans who are not concentrating are not general intelligences
I think we're agreeing. I didn't spell out that I do think consciousness is not understood, it's a mystery, at least for now. The religious solution is to slam some superstition on top of that mystery to make it more digestible.
YjSe2GMQ··on Humans who are not concentrating are not general intelligences
Sure, the correlations between X1:{IQ, conscientiousness} and Y:{income, educational attainment} are stronger than between, say X2:{agreeableness, height, race} and Y.

Y are examples of what people want (wealth). X1 are examples of "valid" reasons to be recognized as useful and therefore attain wealth, X2 are examples of "less valid" or "completely invalid" reasons.

YjSe2GMQ··on Humans who are not concentrating are not general intelligences
Well that is genuinely impossible to tell apart. I mean you can always claim that a system or a person is "overfiting", and given enough time and/or data they/it will be proven wrong.
YjSe2GMQ··on Humans who are not concentrating are not general intelligences
While I agree to some degree this also sounds overly dramatic.

First, there's a ton of evidence that people also get promoted/appreciated for the right reasons, not just because they're a fancy Markov chain of buzzwords (example: serial entrepreneurs, like Musk).

Second, there's underlying reality that eventually comes crushing people that fail to meet the expectations that they had built around themselves.

YjSe2GMQ··on Humans who are not concentrating are not general intelligences
This is, I believe, one of the big reasons why people find it hard to reject religion (another big one being the difficulty of accepting death, of oneself or of loved ones). There's genuine mystery at work here, or so it seems.

You need to distinguish consciousness from ego/free will. Consciousness is the fact that there's something rather than nothing, subjectively. That you seem to experience: sights, sounds, sensations, emotions. Under that definition consciousness is something that cannot be fake (no matter what's the nature of the universe; no matter whether you're asleep or awake), simply because you experience things.

Ego/free will is a separate concept and is indeed an illusion (or an evolutionary artifact if you like The Selfish Gene). There's a lot of evidence for that, the simplest being that no mainstream physical theory allows us to have made choices any differently than we have (barring true randomness like quantum mechanics predicts; but it's also easy to see that that's not freedom, just plain randomness).

YjSe2GMQ··on Etsy sellers say their bank accounts were emptied in major billing snafu
It's how banking works nearly everywhere outside of the US, that's why most brokers handle US money transfers on separate forms. PayPal was built on top of this banking insanity, in most other countries PayPal would solve a problem that doesn't exist.

Whole thing is kind of like the imperial vs metric system.

Which makes me wonder - which other countries have banking systems that routinely allow pull-based access to accounts?

YjSe2GMQ··on Why Google needed to build a graph serving system
> Our infrastructure is somewhat ossified these days: the new and exotic are not well accepted.

I know that it does not involve storage, but Tensorflow is a purely infrastructure project. And it's certainly newer than the Freebase story.

YjSe2GMQ··on In the foundations of physics we haven't seen progress for the past four decades
If you're talking about the income tax idea by AOC it'll only boost taxes by 0.3%, assuming no tax dodging:

https://www.google.com/amp/s/amp.economist.com/leaders/2019/...

YjSe2GMQ··on In the foundations of physics we haven't seen progress for the past four decades
Well but how about a cutting edge silicon fab that gave the entire world a powerful device that fits in one's pocket to access all of human knowledge, instantaneously? For 20B? Add to this the fact that technology is powered by science and you get a very different picture.
YjSe2GMQ··on Twitter Still Can't Keep Up with Its Flood of Junk Accounts, Study Finds
This is humiliating to Twitter. The fact that someone can create a better spam filter without having access to corporate private information (such as user interaction signals, which are normally incredibly powerful) points to that they simply don't want to kill the spam:

> In fact, the paper's two researchers write that with a machine learning approach they developed themselves, they could identify abusive accounts in far greater volumes and faster than Twitter does—often flagging the accounts months before Twitter spotted and banned them.

YjSe2GMQ··on How does one appear in the Google News carousel?
That's just Google's philosophy of UI design, especially of Search. Tiny frontend (i.e. little controls), gigantic backend (that guesses what you haven't or even couldn't select). Works for >95% of population, probably >99%.
YjSe2GMQ··on Security Researcher Assaulted Following Vulnerability Disclosure
If you (like me) didn't know what a Shodan safari is, you're in for a fun ride:

https://techcrunch.com/2019/01/21/shodan-safari/

YjSe2GMQ··on Transformer-XL: Unleashing the Potential of Attention Models
How well does it do? Would you have a link?
YjSe2GMQ··on Consciousness as a State of Matter (2015)
Sorry but that is exactly what I meant: the world is a) either deterministic or b) partially deterministic and partially random (for example quantum), and in either of these cases there's no way you can construct free will agents.

Arguably the b) version is not simple physics, but that's what I meant.

YjSe2GMQ··on Chase had ads on 400K sites then on just 5K and got the same results (2017)
That should be the end of this "story" (oh yeah, Pareto/power distribution, nothing of interest here).
YjSe2GMQ··on McKinsey Advised Purdue Pharma How to ‘Turbocharge’ Opioid Sales, Lawsuit Says
It's a well known fact for anyone practicing palliative medicine that often you have to shuffle around opioids to see which one works. And there is really nothing special about oxycodone, it's just another opioid that's been known for over 100 years.

It's great that it worked for you guys (for real, it's fantastic), but the whole concept of systematically pushing a particular opioid is disturbing.

YjSe2GMQ··on Machine learning leads mathematicians to unsolvable problem
There's a very neat and often forgotten piece of math in learning, VC dimension (unrelated to venture capital):

https://en.m.wikipedia.org/wiki/Vapnik–Chervonenkis_dimensio...

Also, the reason why black box methods are such a big deal now is precisely because controlled/engineered methods turned out to be inferior (obvious example: image recognition; look no further than into the story of the dropout method and Alex Krizhevsky).

Edit: s/basically forgotten/often forgotten/ in the first sentence.

YjSe2GMQ··on Digital identity has three layers, and you can only protect one of them
If you're already a paying data subject consider ProtonMail with ProtonVPN. For example, ProtonVPN has a built in Tor mode. Works quite well for me - I can recommend.
YjSe2GMQ··on The Sorry State of the World Economy
This is crucial. The current way of governing the world depends on the pie of prosperity getting bigger and bigger each year, which allows the government to cut more and more pieces out of it. Without growth every policy that requires more spending must be countered by a policy that requires less spending.

Then there's the problem of governments debt, which historically has been reduced much more by growth than by imposing austerity.

YjSe2GMQ··on How many .com domain names are unused?
It should be fine - strong law of big numbers. They even computed some confidence intervals (99% CI ±3%), but it's not precisely explained how, just references a paper.
YjSe2GMQ··on How many .com domain names are unused?
I was hoping to see there a block with unregistered domain names.

One could do it by, for example, training a Markov chain/neural network to spew out (sample) domains, and then see how many are currently registered, to compute the general XXXX.com space utilization fraction.

This approach would be powered by a more sensible, IMHO, probability distribution than, say, taking all domain names with up to 30 {a-z0-9} characters with uniform probabilities.

YjSe2GMQ··on Consciousness as a State of Matter (2015)
> After all, basic physics requires that there be a physical world we can observe in such a way as to make true statements about it.

But that's the point of confusing free will with consciousness. You can experience something (that's the hot stove argument), and inference is perfectly possible for even today's computer algorithms to do (which are certainly not free). Even something as common as a virtual memory system deciding which pages to swap makes predictions/inference based on what it has seen in the past.

Morality is still a thing, much like there are worse and better paging algorithms. Think of it as mental models of how things work, much like the rest of science. In this view morality is a study of how things/consciousnesses feel.

Edit: another way to look at morality as a science is like this: assume most people genuinely don't want to cause suffering of others, and especially of themselves. But they can still cause harm by being simply wrong.

YjSe2GMQ··on Consciousness as a State of Matter (2015)
Randomness does not yield freedom. A robot connected to a truly random source (like a decaying radioactive substance) is no more free than a fully deterministic x86 chip.
YjSe2GMQ··on Consciousness as a State of Matter (2015)
You're confusing consciousness with free will.

There's indeed no free will (basic physics tells us that, can be also experienced directly via LSD or meditation), but to see if there's consciousness just inflict pain onto yourself.

All this being said it's pretty clear why would humans, or any other apes, or any other species, have morals - to reduce the pain/suffering that them and others receive.

YjSe2GMQ··on Consciousness as a State of Matter (2015)
I think he meant that consciousness is currently irreducible. Consider this:

1. Someone figures out in a lab all possible causes of fear via the surveying method. Fear seems solved.

2. New people appear that claim to experience fear when someone taps their left feet.

3. You have to redo 1. Therefore fear wasn't solved in the first place.

Replace "tapping left feet" with something more obscure, or fear with pain, to continue.

Underlying problem: you cannot compute the "preimage" of fear.

← PreviousPage 4 of 5Next →