And I cannot figure out why you are saying this, as nothing I have said previously either contradicts what you say here, or is contradicted by it. If you could say what you think I am saying in my posts in this thread, we can sort it out.
EDIT: I see the problem starts with the first sentence of your first post here: “Why can't we predict how big or how often those events would be?” - which is completely beside the point in my response to rgmerk, who wrote “It's not clear (yet) what a 100% clean energy powered world would use to cover the last couple of percent of demand when loads peak and/or variable generation troughs for extended periods.” My response to this and the follow-up is this: a) if we are talking about two percent, we can overbuild the renewable capacity, and b) if we are considering all eventualities, there inevitably comes a point where we say that we are not going to prepare for uninterrupted service in this event.
I'm not saying we cannot do it, just that we cannot always get it right, and there is plenty of empirical evidence for that.
The second point is that the distribution has a long tail, especially when we consider the possibility of multiple independent incidents overlapping in time, to the point where it becomes infeasible to suppose that we could be prepared to continue operating as if nothing had happened in all conceivable scenarios, regardless of how accurately we could predict their likelihood.
That is right (if rather exaggerated, and I will note that it was you who originally picked the figure of two percent), and in practice, we accept a certain risk that we will not always have all the capacity we want, even though (or because) we cannot precisely predict how big or often these events will be. There is no particular reason to think this specific case is any different.
There seems to be a tacit premise here, that anything an LLM can do is meaningless as an exercise for a student, but that is simply not true, and if it were, it would likely be the case that we would soon run out of pedagogically-‘meaningful’ (by this standard) tasks (the author has no practical suggestions for how we could avoid this situation.)
Particularly with the development of fracking, natural gas production is no longer a just a byproduct of oil production, and can be (and is) pursued independently. Nevertheless, I agree that we developing renewables should be our priority.
The donor died from rabies, and then several of his organs were donated. Furthermore: "Doctors reviewed records about the kidney donor and learned that the Idaho man’s family had disclosed the skunk scratch to doctors when his organs were being donated, the report said."
I completely agree, and I will add one more thing: when you change how the product looks or works, make sure that users - especially inexperienced users - can easily find the documentation and tutorials that correctly describes the version of the product that they are attempting to use. I would guess that out of ten people who run into a situation where something does not work as described, maybe only one or two will push through the frustration and go on to become power users.
Does that make a difference in this regard? If so, how, and is it an unavoidable penalty for PHEVs? I can see PHEVs having a complexity penalty from having an IC engine over and above the EV components, but that does not seem to be the source of the problems here.
I had not thought much about it until now, but when I did, I guessed it was a cost-cutting move, as it seems to piggyback on equipment needed for other features (principally ABS).
Now that you've mentioned it, I'm wondering if the auto manufacturers needed to get a bunch of regulators to drop secondary/emergency braking requirements in order to do this?
It seems that the author balked at a rather specific level of action: getting government agencies involved. I feel there might be more the author could say about this aspect of the event, though she is not, of course, under any obligation to do so.
Yes it is, as it is about the semantics of type hierarchies, not their syntax. If your software has type hierarchies, then it is a good idea for them conform to the principle, regardless of whether the implementation language syntax includes inheritance.
It might be argued that CLU is no better than typical OO languages in supporting the principle, but the principle is still valid - and it was particularly relevant at the time Liskov proposed it, as inheritance was frequently being abused as just a shortcut to do composition (fortunately, things are better now, right?)
Well said, and I suspect that the author's harrowing experience (in the first paragraph) was a preview of what surge pricing would look like in this market, post-Uberization.
True, but many languages now have words that were absent from their earlier vocabularies. Shakespeare did not have the option to use 'telephone', 'semiconductor' or 'entropy'.
I was struggling with it myself, but I think your reading of it is correct. The paragraph about the 1832 Prescription Act needs some work on its grammar.
Some runways have been extended with ‘engineered materials’ surfaces, often a form of porous concrete into which an airliner’s wheels will sink, absorbing a lot of energy and arresting the airplane without causing it to break up. It is very effective for landing overruns, but I don’t know about last-seconds aborted takeoffs.
Rotation does increase drag, but you need to rotate in order to achieve the necessary angle of attack. The only way to reduce the rotation angle is by going faster than the normal rotation speed for the given weight and airfield density altitude, but doing so is out of the question in this scenario.
There seems to be more to it than that - in my experience with LLMs, they are good at finding some relevant facts but then quite often present a non-sequitur for a conclusion, and the article's title alone indicates that the problem for LRMs is similar: a sudden fall-off in performance as the task gets more difficult. If the issue was just non-determinism, I would expect the errors to be more evenly distributed, though I suppose one could argue that the sensitivity to non-determinism increases non-linearly.
It is video generation, but succeeding at this task involves detailed reasoning about cause and effect to construct chains of events, and may not be something that can be readily completed by applying "intuitions" gained from "watching" lots of typical movies, where most of the events are stereotypical.
Ah, now we are making some progress on where the confusion lies (not that saying "you're confused" without justification was ever much of an argument.)
The first thing to note is that it is not necessary to dispute the notion that semantics come from a conscious observer in order to demonstrate that Searle's argument fails, as that argument is precisely about whether that conscious observer could be an entity deriving its consciousness from a digital computation. As things stand, the only thing saving you from formally begging the question is that you have still not presented a specific argument against my objection[1] to Searle's response to the simulation reply; you still seem to be trying to insinuate that it fails without being specific.
Maybe you also mistakenly think that what I am saying in this thread is supposed to be an argument for computationalism? It is just an argument that Searle failed to make his case against computationalism on account of (among other things) an unsuccessful response to the simulation reply. (I suspect that computationalism is essentially correct, but I do not claim to know that it is.)
[1] It's not just my objection; Dennett, the Churchlands, even Putnam and David "hard problem" Chalmers have raised similar objections.
It's a fair question, but would you trust them equally in an unanticipated crisis, where doing the right thing might be costly in hard-to-predict ways?
Language was spoken long before it was written (or so it seems.) This article almost suggests that sound might be a superior input medium over either digital text or images.