uoaei

Dialects degenerate phonemes that would otherwise occupy identity relations between different utterances of the same "word" (word/concept mutable hyperobject as is the standard in any socially-relevant spoken language) which would be a bit of irony in this thought experiment since common knowledge dictates that more data samples must be present in the dataset (not less, as in rarely-spoken languages) to associate separate pronunciations of utterances representing the same underlying concept. However very-rarely-spoken languages probably don't have distinct dialects since so much focus is put on mutual intelligibility with the few members of the group that remain fluent in that language. It's not outside the realm of possibility that small speaking communities nonetheless fractionate into dialectical specialities but that seems increasingly unlikely as the fervor for preserving/recognizing dying languages increases, and global instant communication continues to become more commonplace.

Example: Schwabisch is wild and would be phonetically transcribed very differently from Hochdeutsch which is its ostensible language progenitor (technically more a cousin than an ancestor in the lineage of language evolution), but if the goal is merely to focus the model purely on phonetic transcription then you can add additional post-processing layers which map sounds to core concepts shared across dialects for actual translation. But I like your idea of interacting with the intermediate elements to familiarize yourself at least with the phonetic patterns, we humans are still thinkers enough to infer patterns of grammar and semantics from these building blocks just as we have done for the entire history of the species/lineage before written representations of language came along (relatively late -- evidence of script cropped up only once civilization had centralized to a sufficient degree to make economics non-local and non-trivial).

tl;dr the big words: it's not til you collect enough spoken samples of the dead(ish/dying) language being spoken that the local idiosyncracies are discovered, luckily linguists are smart enough to probably anticipate and certainly post-process language snippets to grasp the common structures for this or that given language.

Approximate solutions are good enough, but approximate solutions are not part of the solution space for the kinds of problems covered by OP.

No free lunch theorem has nothing to say about approximate solutions, so I'm really not sure what you're going on about.

OR-tools is almost exclusively linear programming which according to its strict assumptions converges more or less trivially, assuming a correctly composed program.

Which means if you're paying for it "as a service" you all but deserve to lose that money.

Different algorithms are better for different problems

So... why does your rhetorical style have such oppositional tone if you're just going to reaffirm the no free lunch theorem?

*sigh* We really need to teach this new crop the term "no free lunch". Again.

The thing is, there are never only two options.

Multiple things can be bad for different qualitative reasons.

How about the broad anecdotal consensus that Duolingo, a gamified language app, famously does not do much if anything to prepare people for conversational fluency.

That's also grammar that's fit to get you harangued and chided by other 4-9 year olds on the schoolyard. Social outcomes are the primary means by which language is meaningful.

English is uniquely riddled with exceptions to rules linguistically. Teaching inflexibly only makes students flex around you.

Perfect grammar is less important than comprehensible semantics for all but the most niche linguistic applications.

Earnings represented by GDP are not distributed evenly across a given population. That 17% of extra earnings goes to relatively few people in the States (I'm putting this very mildly, the concentration in reality is insanely lopsided).

All of the optimism surrounding this industry hinges on "ifs" that never materialize.

That's what makes it a model, not a law.

You're playing party politics. That's the risk you take: that the party has goals beyond your (dareisay naive) utopian ideals for civic engagement.

Parties are not universally evil, when I malign them in this way it is in full acknowlegment that organization is the nearly singular path to "effect on target" as regards society-scale politics. What I mean is the party per se becomes a superorganism that has always as its first priority self-preservation (a la homeostasis) and it is very worth remembering this when subsuming oneself into their structure.

No, it's not a law in the same way as physical laws. Most of the failures in economics of the past 50 years can be mostly directly attributed to a fatal overcommitment to the belief that these attributed relations are incontrivertible.

What really exists, closest to the limit we will call "hard science" or "physics", is the microcosmic focus on individuals interacting in a market. Everything else -- supply and demand as a theory definitely included -- is a statistical extrapolation from micro-scale interactions. Hence the label of "dismal science". It's dismal because every hardline assumption is inevitably contravened by real life physics.

Supply and demand is and always has been a useful first-order approximation. Reality is a lot more complicated.

Dear reader, never believe anyone who says a serious societal problem is "just" anything. Especially on the internet when they use merely a pesudonym to assert their authority.

Especially when the person behind the pseudonym is a one-time washed-out author desperately clinging to their housing stock in Atherton.

It is different, but there may be some universal principles that are relevant more abstractly among both cases. Of particular interest is the empirical notion that statistical models of a certain form will always tend to "average out noise" and "learn meaningful patterns" up to the capacity that those models have for representing said patterns. A parallel notion to this is the hypothesis dubbed "thermodynamic origins of life". The universal principle binding these two seemingly disparate topics is one that seems to underlie any sense of "learning" in physical systems: that semantics of those systems depend on their representational power, and the semantics they do come to represent are the results of adding up many pushes in one "direction" (phase space / state space / etc.) encoding a pattern, and adding up many random noise jiggles will cancel out but give you a first-order sense of variance of those semantic features as expressed by the environment.

As this description is so overly abstract, an exercise for the reader is to try to work through an explanation of how, say, a river delta comes to "learn" about its environment by "reacting" to the influences at its borders, and how it "encodes" whatever it is that it learns in the substrate that it inhabits.

Tried and failed to beat Framework to market. Frankly I'm hopeful that Framework beats this offering out, though I'm happy for the competitive pressure.

Ok, sure. You want more words to say the same thing, here you are.

It got vociferous support from the highest levels of government even though the deception ("protect kids!") was so blatant and transparent, and it wasn't until a legion of privacy and in particular tech-literate advocates raised concerns in mass media together with an awareness campaign about the dangers of unchecked surveillance structures that it was finally... shot down.

You're right, mixed up the names, in UK they called it Online Safety Act.

Europe is, compared to the US, doing a lot more for protection of private data. That includes strict guardrails on what data can be collected and how it is used.

Secret courts still exist but the phenomenon of random Flock employees spying on children in locker rooms at gyms is so much harder to get away with in a system with a modicum of decency.

Chat control was actually shot down, and that was the UK not Europe (anymore).

Laws are different in different places. The world is not composed of America and other-Americas.

They don't cache model state to disk. I am proposing they do.

Your refusal to interact with subtext has me guffawing. I wonder if you even recognize what you're doing.

In the history of revolution, there is never (except in elementary school) all that much weight put on the singular act which instigated the final result. The conditions in place (Jim Crow laws, Southern pride, etc.) lead up to a final moment which our monkey brains like to point to as the cause but in reality there is a simmering cultural froth which could boil over in any number of ways: it just happens that one of the ways is what's described in the Wikipedia article, but it could have started many other ways. All of our understanding about the experience of being Black in the US during that time helps to contextualize the extreme and disproportionate outburst of violence by the White population as racially motivated, serving under an ideology best described as ur-"Great Replacement Theory".

In simpler words, the destruction of Black Wall Street is not without precedent, indeed this was merely one of the more famous and complete examples of destroying the wealth that Black people enjoyed, if only briefly due to the hate of those visiting violence upon them.

Exactly, even in the throes of today's wacky economic tides, storage is still cheap. Write the model state immediately after the N context messages in cache to disk and reload without extra inference on the context tokens themselves. If every customer did this for ~3 conversations per user you still would only need a small fraction of a typical datacenter to house the drives necessary. The bottleneck becomes architecture/topology and the speed of your buses, which are problems that have been contended with for decades now, not inference time on GPUs.

There is a moral dimension, you're just choosing not to acknowledge it.

Feyerabend speaks to things that add context and nuance to the effects, consequences, and provisions of the things you've felt comfortable discussing so far.

Hence the recommendation. Your awareness could use some expanding, if I may be so blunt.

Yours is a rather pedestrian dorm-room take on epistemology and relevance of the moral dimension to social progress whose flaws are addressed in longform by Feyerabend.

So is it likely this one merely escaped? I find it hard to believe someone who would own one of these would not be an enthusiast, and that enthusiasts wouldn't find another owner for a critically endangered species rather than merely drop it under a local bridge.

I recommend for you to read Feyerabend's Science in a Free Society.

It read like a longtime adderall addict who switched to clean meth a while ago.

you see the tax payer as a cunning and evil adversary that needs to be reigned upon, and you see that all the jokes, the water cooler talk, the general ethos is toward this vision of the tax payer

Any force employing threat of violence for control does the same. Police presence, military occupation, hell you even see it in the eyes of loss prevention folks.