CommieBobDole

Looks like Dekton is primarily a countertop material and the reason the cooktop doesn't come with the normal top is it's designed to be hidden under a countertop for trendy minimalist kitchen design reasons. Which also goes against the "kitchen equipment nerd" framing since the hidden induction cooktops seem to be worse in several ways than the normal visible ones.

I wonder why it's still here after all these years? It seems both useless and harmful, one of those features whose lifecycle consists of someone rolling it out because of some idea they had and then removing it three hours later after having to fix multiple headlines because they didn't think it through.

What is it even supposed to do?

So this is a really neat technical demo, using spacecraft cold-gas thrusters and guidance hardware/software to do crazy stunt stuff in a car, but what's the relevance to an actual product that can be sold, even as a small-volume exotic? These are pressurized gas thrusters, so either the tanks, valves, and piping take up most of the interior of the car, or it has a low-single-digit number of seconds of thrust before recharge, probably both. Also the tanks for these things tend to use incredibly high pressures and are apt to explode spectacularly in any sort of accident.

This is sort of like demonstrating a car that softens the asphalt in front of it with lasers to get a better grip; difficult, technically impressive, and zero chance that it will ever be allowed on a public road.

Edit: According to this Motor Trend article, it uses a big carbon-fiber pressure vessel at 5500psi, so basically an enormous fragile bomb. And the one that did all the stunts had two of the pressure vessels which would have filled most of the interior.

https://www.motortrend.com/news/tesla-roadster-thrust-system...

Looking at the PAGER info, it looks like it was pretty far from the largest cities and only affected them, and Panama City in particular, minimally, which will limit the impact quite a bit.

https://earthquake.usgs.gov/earthquakes/eventpage/us6000u18k...

There has apparently been no confirmation of his death - Wikipedia still lists him as alive and editors there say there's insufficient evidence. If the linked article is true, maybe he faked his death.

This is nice, but how are you determining the location of a camera for map purposes? Some of them are correct, but there are others that have very specific map locations that are off by as much as a hemisphere - an Ohio one in Chicago, a Venice, Italy one in San Bernardino, Pennsylvania and Ohio in Phoenix, etc.

"No, we don't know why the two positions of the switch are labeled "magic" and "more magic".

I have a bin of cables marked "Obsolete". A cable only goes in there if there's absolutely no possibility of it being used for any reason ever again.

I use cables from it once every few years.

What do you mean? I assume the "we" and "our" refers to the author and his wife.

It does seem like a lot of the valuations and infrastructure investments for these companies only make sense if each one assumes they will be the first and only one to invent superintelligence and that it will largely replace all knowledge work.

The article touches on something that I've been thinking about with regards to Google's AI strategy; the automatically-generated AI search summaries are not great. They very frequently confidently misinterpret what the user is searching for and generate half a page of useless information that pushes actual results down the page, and they are occasionally hilariously incorrect, with hallucinated facts.

This is probably a difficult-to-solve problem; given that they generate billions of these a day, not even Google can afford to devote enough compute to each query to reliably generate quality results. You can see this by selecting the "AI mode" from the search interface after getting the mediocre summary - the results are much better and generally perfectly usable. Though even that is probably a special minimal-compute version of the lowest tier of Gemini, it's still maybe an order of magnitude more capable than whatever generates the search summaries.

The bigger problem is that these search summaries are the default and by far the most common interaction that the general public has with "AI", and because this experience sucks, they just assume that all LLMs are similarly stupid and mostly useless. In non-technical spaces I frequently see the argument that "AI" is not useful for anything, all it generates is garbage hallucinations, and almost invariably they cite some actual terrible experience with the Google AI search summary. I would argue that the strategy of adding LLM summaries to every search is the worst of both worlds - it makes classic search worse while poisoning users against the idea of actual LLM-assisted search.

Relevant XKCD:

https://xkcd.com/2501/

LLMs have solved writing in the same way image generators have solved stock photos - if you need an image that doesn't need to convey any information other than "I have added a somewhat-related image that doesn't look obviously AI-generated to this", you can now generate such an image pretty easily. That's a low bar, but almost the entire genre of stock photos exists in service to this goal.

Likewise, there's a whole lot of day-to-day writing that mostly exists to communicate "I have written some text about this subject because I am expected or required to" and at most needs to communicate some simple information. LLMs will probably replace a lot of that writing, at least temporarily. As people increasingly use LLMs to consume that information as well, it will probably eventually be replaced by some sort of inter-agent communications protocol.

Also, the dark pattern where it shows the interface and lets you enter a prompt/set settings, but then pops up the 'create account' dialog when you press submit is pretty annoying.

There's a blurb in the article:

>"Massgrave, the group behind Microsoft Activation Scripts, has noted that Windows setup sends hardware info to Microsoft and receives identifiers back that are later used for Store access and licensing. Blocking GDID assignment breaks both activation and UWP apps."

I think probably it sends the gdid back to Microsoft as part of telemetry/updates/MS account periodic re-auth, which lets them know the current IP address, and then once the government has Microsoft's logs and the various target websites' logs, they can correlate based on IP address. I don't think it's actually sending the gdid to the web sites. Maybe.

This serves to further illustrate that nobody should be using Windows for anything that involves the need for privacy. And doubly, triply, and morefold so, nobody should ever sign a Windows machine into a MS account for any reason.

Paywall-free link:

https://archive.is/Ipm3J

For many years, the title of the leadership role over the various precise time products at the USNO was "Director of the Directorate of Time"

The article is not about "should people be allowed to buy this product because it's potentially dangerous/addictive/etc" but "Should the company be allowed to sell this product because it consists of acetaminophen plus two useless ingredients and is basically a scam".

This is the Platonic ideal of an LLM-generated blog post; it's pages long, it uses a lot of flowery language, and communicates almost zero actual information. I can't disagree with the premise, because the premise melts away when you try to pin it down, like cotton candy washed in a stream by a raccoon.

Congratulations, person who wrote the prompt that caused a machine to generate this article, you have truly achieved some sort of perfect something.

I know I'm replying to comment from yesterday that nobody will see, but here's my setup:

I run a local Qwen 3.6 27B at a 5-bit quantization on a 3090ti - it's a dedicated server that used to be my old gaming PC/workstation. I mostly use it with a chat interface for deep web searching and simple script generation. Here's what my setup looks like:

LLM host - the aforementioned physical box that contains the GPU and hosts Qwen 3.6 27B and also a CPU-only Qwen 2.5 1.5B that I use for generating chat titles in Librechat.

Librechat host - a VM hosting librechat; this provides the chat interface and interfaces with the search provider.

Search host - a VM with SearXing and Firecrawl.

The flow is:

- Qwen decides it needs to make a search tool call and does so - Librechat receives the call, and feeds it to the SearXing API. It receives search results in JSON format.

- Librechat feeds the result URLS to firecrawl, which scrapes the pages and returns the contents to Librechat.

- Librechat prompts Qwen with the results of the search, and it either formulates an answer or performs additional searches based on the results.

What I don't have yet is a reranker; this is a specialized model that analyzes the scraped documents and returns only the ones (or even just snippets) most relevant to the query. This would be useful, because sometimes the LLM ends up semi-randomly with very irrelevant results and spends multiple turns trying to get better ones and/or gives a bad answer. No specific reason not, I just haven't implemented it yet.

I agree; the situation "young people know all about computers" was maybe a twenty-five year phenomenon; before that, there weren't enough computers for it to be generally true, and at some point you no longer needed to know all about computers to successfully use a computer for the sorts of things the general population does.

The issue with this is that we don't know how it works. Generally speaking, we know how the level of abstraction that we were born with works. We might have some understanding of one or two previous levels, but that decreases the farther down you go. We might understand the next level, and some of the next after that, but eventually people will be making things that we don't have the context to understand without having to unlearn a lot of what we know now.

I'm old enough to see this process in action; I used to be young and in possession of esoteric knowledge that made me infinitely in demand and now most of the things that young people have esoteric knowledge about is things that I don't particularly care about, and I'm left with a lot of finely honed skills to solve problems that have mostly been abstracted away.

I've experienced the same thing except sed 's/HEB/Publix/g'

This is probably just a 'Kroger vs nice supermarket' thing.

You're right for the link as provided, but the (apparently) correct link provided upthread is https://wizardzines.com/comics/no-feigning-surprise/, which is pretty similar to the XKCD.

Isn't most of our economic growth based off inputs of energy (in the form of labor, electricity, etc) that ultimately derive from the sun, though, rather than primarily non-renewable non-recyclable resource extraction? The sun is technically a non-renewable resource, I guess, but only on cosmic timescales.

This has always been a thing with IT advice, though - the more complex a system and the outcome, the harder it is to clearly define "better" or "worse". Add in the fact that LLMs are intensely and emphatically non-deterministic and LLM guidance basically becomes gardening advice.

Heck, even the 'benchmarks' are mostly somebody's attempt to crystallize their vibes with varying amounts of success.

Right, but that's still external to the LLM, it's just a KV cache that's stored on the provider side for performance reasons, so that the client doesn't have to re-send the whole chat history with every subsequent call in the conversation.

It still generates every response using the model's pristine state with every new API call; whether the context is provided from the client or from a colocated cache server doesn't really change that.

The fact that a LLM is essentially immutable would be my biggest argument against consciousness or self-awareness.

It's a big file with a bunch of coordinates describing spatial relationships between tokens. When you give it a prompt, it uses those relationships to generate a string of tokens that is a statistically likely response to that prompt, then it stops. It's not changed by the experience. It doesn't remember anything. It doesn't sit around thinking on its own.

Even if the model itself were extremely complex, it's hard to imagine a definition of consciousness that includes something that doesn't remember and can't change.

This article suffers from two things:

First and most importantly, it's not really about LLMs, it's about AGI, and the second does not necessarily follow from the first; LLMs in their current state are pretty clearly not AGI, and most of the LLM-world progression in the last few years has been about better tooling/interfaces, refinements in training data and techniques and people learning how to use LLMs effectively rather than the huge leaps in fundamental capability that we saw in earlier years. It seems more likely that at this point, when AGI comes, it will be something entirely new or something that LLMs are only a component of, rather than "we built an LLM with ten trillion parameters and suddenly it became God".

Second, it's not even really about AGI, it's about AGI superintelligence. And more than that, it's about affordable AGI superintelligence, assuming that such a thing won't cost billions a year to operate.

I've found that the AI overview is usually right but confidently wrong enough of the time that I don't trust it. The interface that you get with the 'AI Mode' button (which I assume is just Gemini with very low compute settings), however, is usually pretty solid for well-documented queries.