@VoxelVexillologist's banner p

VoxelVexillologist

πŸ‡ΊπŸ‡Έ Multidimensional Radical Centrist

1 follower   follows 0 users  
joined 2022 September 04 18:24:54 UTC

				

User ID: 64

VoxelVexillologist

πŸ‡ΊπŸ‡Έ Multidimensional Radical Centrist

1 follower   follows 0 users   joined 2022 September 04 18:24:54 UTC

					

No bio...


					

User ID: 64

Thanks. I suppose publishing a white paper at least opens them up to more serious scrutiny (which I've seen in serious-ish forums like HN). But my initial response is rather skeptical still, though. The encryption methods seem like they should be far more expensive than Apple is letting on [1], and they say they're using this for querying remote databases with encrypted queries. They're less clear on how these databases (for photographic landmarks, URLs, and such) are encrypted in a manner that actually hides the query from Apple. Are the encryption keys different per-device? If so, how do they avoid needing a separate database per device? And if not, it seems there's a lot of trust that they would be unable to figure out which rows matched.

I know Google's approach to similar issues has been focused more on device side ML models. Pixel phones support offline song recognition (I've noticed it's fairly limited to popular songs), and Google Translate can work (in a limited fashion) offline. Why does Apple need to do cloud-based POI recognition in photos? The whitepaper only shows 6 very well known landmarks, but it seems like it'd be easy enough (and secure!) to do this on-device. Given the known computational costs of FHE, it might even be better for battery life.

  1. Quoting the linked FV paper: "As to whether any of these [previous] proposals is really practical, the answer is simply β€œno”.... The most recent paper manages to execute one AES encryption homomorphically in eight days using a massive amount (tens of GBs) of RAM memory." That's before their proposed algorithm, but they don't directly seem to claim many-orders-of-magnitude improved tractability.

There was a fairly infamous case in Cleveland, TX in 2010. Some might debate "Western men" here, but those charged were American citizens as far as I know, not MENA immigrants. Maybe some of them were Muslim, but probably not all.

pathological need to double down at every opportunity.

Isn't there a classic stock trading scam where you send 2^N Boolean tips ("will go up or down today") for N days, then focus on the small-but-nonempty group that got N correct tips in a row to try to convince them to trust your judgement? It seems like given enough VCs (monkeys) and opportunities (typewriters), even random chance will give you a few coherent words of Shakespeare and a fawning audience expecting great wisdom.

Of course, I do think your average VC billionaire is much brighter than random chance, but there is a huge selection bias when you find the one (of an unknown size set) that has consistently doubled down and won and assume this to be representative generally.

homomorphic encryption

Are they currently claiming they do this today? My understanding of homomorphic encryption (admittedly a bit outside my wheelhouse) is that it's nowhere near as well-trod a space as, say, RSA. When I last looked, it was possible -- with a bunch of caveats -- to do simple things with a whole bunch of overhead, and certainly no equivalent of a NIST standard (if you trust those: say "Dual EC" with me) for it.

I didn't think the technology to do this well was ready for prime time today, but maybe I'm just a bit out of date. Do they have a white paper, or, better, a bunch of academic papers?

Viable quantum computing dropping today (or even in the next decade or two) would also break almost all extant (asymmetric) cryptography. Yeah, NIST just recently published specs for post-quantum crypto, but I expect it'll be a decade before those are universally supported. Maybe less if it happens: SSL everywhere happened fairly fast, but became a real priority almost overnight. But if quantum were something any well-founded startup could do, nation-state actors could throw some impressive wrenches into any secure networks for a while.

As others have mentioned, good contractors have enough customers lined up that they don't care.

I wonder if the right disruptive model here almost goes the other direction: not a clearinghouse for finding tradespeople, but a trusted service you can have "on retainer" effectively to subcontract the work. Ideally, you call 'em up, ask for [task], and they shop various provider options (some perhaps that they use often) with a reasonable expectation of the marketplace, and maybe even for an extra fee can manage the "will arrive sometime between 9:00 and 3:00" part where it's pretty disruptive for folks with full-time not-from-home jobs to let them in and get things done without the usual concerns. Bonus points for being able to make sure the job is done correctly the first time.

Although what I've described sounds a bit like a combination of the network of a general contractor (for larger tasks) and a rental property management company. But I've never heard a sufficiently-glowing review of the latter from a renter to want to consider asking "Hey, I own this house and live here: would you be willing to handle when something breaks?" Does anyone actually do that?

In my experience, hospitals are more than happy to screw over patients in billing as long as they don't complain too much after the fact. Surprise out-of-network anaesthesiologists used to be common (now prohibited), and I've seen hospitals try to tack on not-covered-by-insurance fees that show up much later and weren't disclosed in advance (not that they ever give straightforward billing answers in advance). Yeah, they'll "kindly" remove or waive those if you call and complain a bunch (probably marking it down as "charity"), but it's really annoying and not always worth my billing rate.

What I find interesting about this is that it wasn't exactly quiet at the time: The Kite Runner was a pretty popular book at the time and featured the practice pretty prominently, albeit set at a time other than the US occupation.

Arguably, there have been several instances where acts that were themselves considered terrorism have at least swayed the public consciousness about other events. I find the Oklahoma City bombing pretty abhorrent, but it demonstrably caused reconsideration of the narrative behind the Waco siege. Or the Tunisian street vendor whose self-immolation kicked off the entire Arab Spring. Or why we're all talking so much about health insurance denials suddenly.

Of course, plenty of actions meant this way are not successful, but IMO something dramatic and well-documented in this instance might have caused public outcry.

Earnestly, it'd be nice to see the US improve vehicle-pedestrian safety standards more like those of Europe.

More tongue-in-cheek: we need to ban assault cars! Anything with a higher top speed than 85mph (fastest posted speed limit in the US), better zero-to-60 speed than a Honda Civic (7.5 seconds), or dangerous cosmetic features (spoilers, racing stripes, red paint). Or anything over 3000 pounds. Whoops, most vehicles fit at least one of those.

The last few times I've gotten a new Android phone (Pixel), it's come with a USB-C to USB-C cable (with a C-to-A adapter for chargers) and simple instructions to connect 'em directly, unlock the old one, and agree to transfer everything. Admittedly it missed quite a few apps the first time, but the most recent time things almost entirely Just Worked. I think they may have wireless or cloud options there too, but the cable was simple.

Although the security-minded have complained about some of the changes in, specifically, Google Authenticator over the last few years to support this.

Defending hadn't worked for anyone yet

Thailand was never colonized. It was also, weakly, a member of the Axis during WWII.

Are you limiting "tech" to software companies? Otherwise I'd point to a few European examples like Nokia (pre-acquisition) or Ericsson or AMSL. Acquisitions are also probably skewing your count.

Are similar warping effects (e.g., make-work projects or services that are created to compensate for a societal failure) a major contributor to variations in GDP values?

As a thought experiment, if nobody got sick, we could eliminate healthcare from GDP, which would be a huge paper loss (and lots of jobs) but doesn't seem to be a loss for what I'd consider "forward progress". I'm not really an RFK-stan or anything, but we presumably have a net GDP gain from incentivising unhealthy lifestyles and then paying for expensive mitigation treatments.

If, despite her skinniness and near-veganism, she's strong as an ox and endures like a camel, then there is no problem with her diet.

I think this is probably a good starting point, but I will observe that there are plenty of professional athletes, even successful ones, who have eating disorders. The term of art there seems to be RED-S these days. For those, the concerns are more "are they getting injured more often than they should" (energy deficiency can cause bone density issues), and, for women, if they are menstruating regularly (hormonal birth control can mask this issue).

I have lately wondered if what we're missing is a bunch of, I suppose, nuns with rulers or the equivalent to punish modest anti-social behavior (defecting, in this conversation) promptly with a transient painful stimulus that neither leaves lasting marks nor a permanent legal record.

But I don't think it would work in all cases, or maybe even at all. And it's almost certainly disallowed by the Constitution. And suffers from a lot of ambiguity as to where to draw the lines.

There are some fairly highly ranked red state public schools that are probably less-left-leaning than most given the leverage state elected officials have over the institutions: Texas and Florida have both explicitly banned school-sponsored DEI initiatives. At least Texas A&M has a bit of a reputation for conservativism, and at least in the past when I've talked to lefties from Colorado they treat Colorado Springs as a very red part of the state because of the Air Force Academy there, but that may be a bit out of date at that point.

I'd bet any heavily-Greek school is "conservative" in at least a change-averse, slightly-social-conservative fashion that might not map to politica, but I don't have direct experience there. And this isn't to say that these schools are necessarily right-of-center, but more right-leaning than most universities.

In my experience, the few folks that care about "noon is noon" (say, those aligning telescopes) already aren't using UTC because the second granularity of leap seconds isn't good enough, and the statement is only true for a single meridian within your time zone anyway.

For example, almost all satellite navigation (also used for time synchronization of local clocks) uses monotonic time on the backend.

The fact that Unix time is based on UTC, which observes leap seconds, and not TAI, which is monotonic, is IMO a pretty poor decision. It's not "the number of seconds since the epoch", it has weird jumps and skips.

IMO Mars doesn't necessarily seem like it will be the most interesting destination: It seems likely that once we have the technology for extended in-space habitation to get there, the bottom of a large gravity well seems a relatively boring place to hang out. What does the planet get you? Gravity? Spinning habitations seem easy enough. Meteorite protection? We'll need to have figured that out anyway. Land? Is it really easier for farming than in-space?

The asteroid belt looks a lot more tempting to me because even if resources are scarcer (unclear), they are easier to move elsewhere.

I feel like I've seen plenty of Christians do this too: see gosh or geez for long-standing examples of the practice.

Thanks! I guess for context I considered you as "part of the Anglosphere", although there are different degrees of centrality to that concept. I remember you had a couple (interesting, IMHO) posts a while back about how difficult the US regime would have made transferring your education credentials.

Do these people want to come? I'm not sure they do.

I would be interested to hear @self_made_human respond to this, if willing.

Heh, I had a (mostly-indirect) experience with this as a patient once: I got the letters from my insurance denying a claim (twice!) because the provider, according to the letters, neglected to put their tax ID on the paperwork. It got sorted out mostly without my involvement, but I did spend time worrying about it and even calling the provider to inquire about it.

I assume they filled out the form correctly on the third attempt, but it presumably behooves the insurance companies to have increasingly Byzantine processes for paying providers.

Wasn't that roughly the idea behind HMOs? Those seem to have fallen out of favor because they always require referrals and heavily limit providers in practice. But I'll admit I've never had (or even been offered) such a plan.