SomeStupidPoint

I'm terrified of people who use phrases like "inciting hatred" as an excuse for why we should start ostracizing entire swatches of people from social discourse. Hilariously, I'd say that if anyone is "inciting hatred", it's probably you at your strawmen.

In my experience, they're much more dangerous to the well-being of society than the people they're trying to silence, as they're generally authoritarians of the worst sort without a shred of remorse because they're doing it for The Right Reasons(TM) and are sure of it.

But if I tolerate you speaking, someone I view as an immediate and visceral threat to my society, I suppose I can tolerate just about anyone using words.

It's a shame we'll likely come to violence because you can't show the same tolerance, and use words not force on people you disagree with.

> This is also naive. In that case you won't be able to publish your results in prestigious journals and your career will die out.

Charlatan fortunetellers aren't owed careers as scientists -- if they can't produce reliable, predictive models because they utterly fail to control experiments,then they should have their careers destroyed.

We've seen enough damage from corporate leaders and politicians relying on guides to reading chicken entrails masquerading as science.

It's time to stop the fraud.

Bitcoin radically gimped their VM at the start, destroying the inherent value of the coin.

BTC is going to die to a gen 3 VM that learns from the ETH mess of gen 2 and just ungimps the BTC VM.

I don't believe BTC has the political leadership to fix that poor decision, given that a) they've demonstrated poor leadership until now and b) some of the people involved in the gimping are still in positions of influence.

Then it's outright fraud to present your results as reliable when there's radically uncontrolled variables that dominate the results.

Your experience of it being hard to even get started because of the insular hardware community mirrors mine. (Well, I guess my employer does use FPGAs somewhere, but in a department far from me.)

Are you just using a Cyclone V demo board? To be honest, it's actually a little hard to even find FPGA accelerators that aren't bespoke.

Tangent, but I suspect you've researched this topic more than I have -- do you know of any good source on the subject of FPGA NNs?

It dawns on me that FPGAs might actually be better than GPUs for a project I've been thinking about, but I know relatively little about state-of-the-art design/synthesis techniques.

Do you even want to use floats when talking about FPGA NNs?

You can talk directly about circuits on bitvectors, of which a subset looks like floats doing math, but your network might have a better encoding than that.

It's mostly just pointing out that in a particular extension of the integers, if we attach the square root of -5, then (1 + sqrt-5) * (1 - sqrt-5) = 2 * 3.

3 divides the left hand side, but it is not a divisor of either term. This is different than in the usual integers, since there are irreducible terms (not divisble by anything) that aren't prime (if they divide a product, they divide one of the factors).

Fun fact: many desktop printers can hit 3kbps in V40L QR codes (biggest + lowest ECC).

(Estimated at approx 10ppm by 20KiB per code.)

For the security conscious on a budget, printing QR codes to scan to a networked machine can be a way to get information off an otherwise air-gapped machine, with easy introspection and auditing. (This came up in the context of offline signing of Bitcoin transactions with a "warm" wallet -- digital, but airgapped from the "hot" wallets.)

> core functions

We're not talking about outsourcing tangential features, we're talking about the outsourcing of waging war and core abilities of the military, eg the hired hackers at the NSA or outfits like Blackwater.

Privatization isn't about purchasing supplies through corporations outside of the government, but having corporations perform key features of the government.

Your example isn't really about that, so you're right -- you're not an expert, you're talking about the wrong thing.

How -- conceptually -- does the military being unable to afford the salary of person A get fixed by subcontracting to hire person A through corporation B?

That literally just makes person A cost more money.

Aren't muons terrestrial particles created by cosmic ray impacts on the atmosphere, though?

This is what happens when you force the military to outsource core functions to fleece taxpayer money for your buddies.

For a long time, there's been a push to not let the military develop capabilities in-house, despite repeated contractor fuckups and huge overspends.

Can we just admit privatization is a failure?

It's literally destroying national security.

Everyone should appropriately consider the source (and their security concerns), but this also exists:

https://github.com/iadgov

It provides some advice and references a number of other government sources once you dig into it.

The question that you're circling, but can't seem to put your finger on is "if they weren't concerned with oil's future, why would they diversify?"

Historically, oil has gone gangbusters. I doubt any of the sovereign wealth funds lost money buying oil stocks. Yet, within a decade, we've seen a few of them announce that they're going to diversify. Ostensibly, this is a move to stabilize their long-term value against market fluctuations, but given the criticality of energy and how well historically these companies have done, it seems more likely that they're worried about a major market shift on the 30-50 year horizon and are looking to reposition themselves ahead of when the market itself moves.

So to me, it looks like major oil producers are preparing to leave oil, when viewed through a generational lens -- the wealth funds are clearing out, educational training programs are shifting, etc.

In 10-15 years, the flow of new workers will hit whatever low-level long-term carrying will be; in 30-40 years, oil will have settled into some kind of "post primary energy" role. (It's not like petrol products are suddenly not useful; just the bulk of traffic will shift, and so there won't be enough business to support several nations.)

Of course, 2050-2060 was when people were guessing that was going to happen for quite a while. But this seems like a sign we're on track for that, broadly speaking.

Also, what a great chance to have a sanctioned hunt of an Argentinian sub/vessel.

Nothing breeds experience at hunting subs like going sub hunting -- and now several navies get to, in an area they know one exists, without it offending anyone.

OP is dumping the change at home, then converting it to bills all at once.

That works even when coins have non-trivial value.

Also, healthcare isn't the only industry with that type of problem.

That makes the issue more resonant, because the deep corruption of the legal system and routine criminality of corporations cuts across pretty much every facet of life.

Intentional criminality by corporations should be treated exactly like intentional criminality by gangs, with the leaders rounded up and served sentences according to the scale they operated on. Anything in excess of a billion dollars of criminal activity should be a life sentence.

I somehow think compliance would be much better if executives were held responsible for it. But... That's how corruption works -- they're not.

That argument is based on the premise that the best thing to do is be cowardly if it benefits you, even if it costs a lot of people who you have a duty towards substantial amounts. The long term prospect of repeatedly caving to corruption taking relatively small portions of the population hostage is eventual corruption of the entire system -- death of the system -- which costs the futures of the very people who would be angry at you.

I'm not sure I agree with that premise.

....Default 'featured' sort?

When I go to Amazon.com and search, I see 'relevance' as my default, with 'featured', some price related ones, 'average', and 'new' as options. ('Featured' only seems to exist on some products, and be related to ads.)

Is it not the same for you?

-----

As for your main point (because I think that your complaint is still valid even with 'relevance' as the default), it sounds like what you want is a way to choose what factors are applied to your sort.

I'm not sure, but it seems likely that 'relevance' is doing more than just averaging, and so being able to select which parts you apply (eg, only use a statistical notion of best, don't consider availability or shipping times) would cover your usecase, right?

Well, you might want to be able to choose between a few models of 'best', but the real issue, the core need, is that you want control over the model that Amazon is using to sort what you see and to have some input on what that looks like. (And not just have 'lolsux' or 'Amznsort', to be a little glib.)

Gotta say, that actually sounds like a pretty reasonable ask. I'm not sure why it doesn't work that way, either.

They invented a whole language to paper over the fact that their entire field is weaponized psychology employed against their customers. Do a noun for noun substitution to unpack a marketing meeting, and you'll want to puke.

Frankly, it's shameful for any company that says they care about their customers to employ marketing psychology tactics. Marketing and advertising are (in their modern, virulent form) simply an attempt to hack viewers to shift the locus of control from the customer to the business. (Other objectionable things about tech have this kind of behavior, too.)

Frankly, it's just disgusting and I can't blame customers for really viewing companies negatively based on their advertising practices.

(If all that wasn't bad enough, online advertising often comes with literal malware ontop.)

What's weird is they don't even test knowledge of the theory or operation of data structures, just if you can puke a few classic ones on a whiteboard.

Honestly, I think they've optimized on simply hosting an ordeal and selecting for any employee capable of performing the feats, even if they're unrelated to the job. It selects for employees willing to put in work and take a lot of shit, which is what big corps want.

I'm not sure what you mean -- could you elaborate?

Amazon probably doesn't use straigt score averaging to decide "best" items sort, and this is just proposals of how to change that to be better by not just using averages. So what is it you're looking for Amazon to add?

Disclaimer: work at Amazon, not on anything search related.

....Dosing has everything to do with how harmful something is.

Otherwise, the term overdosing wouldn't exist.

I'll be honest: I suspect that this is actually quite common and I have absolutely no idea how to fix it.

But I suspect that much like homelessness, step 0 is gathering an accurate, systemic level picture so we can even understand what we're trying to fix. Targeted treatments and good intentions aren't working -- we need a system level mechanism.

Ed:

I replied to a cached copy; leaving in case anyone raises the same point.

> Child endangerment laws are enough.

Are they, for small doses?

Sincere question -- how do we spot things like this?

Let's pretend for a moment that I'm a relatively competent programmer with small amounts of disposable income to spend on hobbies, such as running algorithms over government datasets. (Okay, that's a weird one -- but I think it's good to apply my craft towards politics.) I know a lot about making computers scan through large sets of data looking for patterns, and some about making websites and whatnot.

I normally don't think technology does much for political problems, but this one legitimately sounds like we should be applying fraud-detection style algorithms to the names that appear in court records, looking for abnormalities. (Heck, we should probably look for all kinds of irregularities, not just in the pattern of names.)

Identifying the problem doesn't magically solve it, but it sounds like being able to assemble subtle cases out of thousands of documents across hundreds of jurisdictions actually would provide utility towards addressing this.

It doesn't stop the corruption directly, but it shines a pretty bright light on it. And it allows groups looking to tackle it head on to focus their efforts on litigating, instead of document finding or figuring out who the culprits are.

What really changed laptop buying for me was the cloud.

No laptop is going to have 8x discrete GPUs, but all of them can manage to SSH using my cellphone as a hotspot.

So I started buying laptops that were easy to carry, nice to type on, etc.

Tangent, but...

I'm actually curious if the wealth gap is part of what caused the slowdown -- the super rich got greedy, ate into their seed corn by taking too much wealth from the productive parts of the system, and now won't make the massive investments and restructuring of intake from them necessary to restore the system.

They could directly invest in infrastructure (coupled with suitably sized bribes, which can be kept to a small portion when done directly), which is what China is doing.

But they'd get accused of colonialism if they actually tried to help directly like that (and perhaps it would be true). So they're left in the untenable position of dealing with mass migration or funding warlords.