ufmace

I agree with crazygringo that uBlock Origin Lite seems to work fine for me as far as blocking ads on the websites I visit.

I also agree that these discussions can be frustrating. In my opinion, that's because people claiming that Lite isn't good enough only seem to post super vague stuff, like links to the FAQ that list a bunch of technical details about what it can't do, when I don't understand the practical upshot of those things. Or vague assertions that it's not doing something which is allegedly important, where it's never actually explained what that thing it's not doing is and why it's important.

I have yet to see anybody show a specific example of a website where Lite doesn't actually work well enough. Or of any other specific thing it's not doing. I don't think I should have to read a series of 20 web pages dense with specialized technical details to understand what it's supposedly not doing. If it can't be explained simply and clearly what it's not doing that's so important, maybe it's not actually missing anything important at all.

I suppose I am a unsophisticated user of web browsers. I never got around to understanding or interacting with all the details of what "proper" uBO can do. Yet I still seem to browse the web just fine, and even build webapps sometimes, and I don't see any ads. So what's this great thing that I'm missing?

I don't think so. A moderately hard landing with an engine(s) smoldering because they were on fire but had their fuel cut off is probably survivable for most of the passengers. A moderately hard landing with the engine(s) a raging inferno pouring burning fuel all over the place because the fuel couldn't be cut off or took too long to do so is much less survivable.

Putting complex and fallible restrictions on safety-critical controls like fuel cutoff is usually a bad idea overall.

Pretty obviously a bad joke and a bad idea IMO. I did not personally downvote, but I think it deserves its current score.

Look at the timeline of the events. The switches were shut off, noticed to be shut off, and restored to the proper position within 10 seconds with the current system. Insufficient notification that the switches have been turned off was clearly not a problem in need of a solution. It would be slower and more challenging to understand an automated verbal announcement than the surely extremely obvious sudden lack of thrust and all engine dials rapidly dropping to zero.

So it wouldn't contribute at all to solving this particular case, would only be a slightly annoying distraction in the more normal case of normal aircraft shut-down after completing its flights, and would be a potentially hazardous distraction in the intended emergency case of engine is on fire and fuel must be cut off immediately, where there's probably a bunch of other extremely important and urgent things to pay attention to and do other than a silly automated warning telling you what you just did.

I appreciate this perspective, but I think building a tool like uv in Rust is a good idea because it's a tool for managing Python stuff, not a tool to be called from within Python code.

Having your python management tools also be written in python creates a chicken-and-egg situation. Now you have to have a working python install before you can start your python management tool, which you are presumably using because it's superior to managing python stuff any other way. Then you get a bunch of extra complex questions like, what python version and specific executable is this management tool using? Is the actual code you're running using the same or a different one? How about the dependency tree? What's managing the required python packages for the installation that the management tool is running in? How do you know that the code you're running is using its own completely independent package environment? What happens if it isn't, and there's a conflict between a package or version your app needs and what the management tool needs? How do you debug and fix it if any of this stuff isn't actually working quite how you expected?

Having the management tool be a compiled binary you can just download and use, regardless of what language it was written in, blows up all of those tricky questions. Now the tool actually does manage everything about python usage on your system and you don't have to worry about using some separate toolchain to manage the tool itself and whether that tool potentially has any conflicts with the tool you actually wanted to use.

IMO, this is the process for building an application image for deployment to production. If the lock file is not present, then the developer has done something wrong and the deployment should fail catastrophically because only manual intervention by the developer can fix it correctly.

I feel like it's happened to me multiple times that I've seen an ad for something I actually want, but if I click through or look up the company advertised, then do a little research on that company, I discover that it's a scam or a super crappy version, then I actually purchase the thing from a more reputable company with higher quality. So I guess they succeeded in getting me to buy something, from their competitors.

Why does this friend in particular deserve an exemption or reduced rate? Why don't they do something like take the trains into the nearest NJ suburb and leave their car at the parking lots there, which will probably be free or much cheaper, since they're doing the opposite of most commuters. Then they'd avoid this and all of the other tolls, most of which are much more expensive, and would probably be faster too.

Even that isn't enough. Much of the bias nowadays comes in which specific facts and events you report, and which you ignore.

Now that I have a little more time to search around, I easily found this study, published March 31st this year, so not quite 3 months ago:

https://arxiv.org/abs/2503.23674

I only skimmed it, but I don't see anything clearly wrong about it. According to their results, GPT-4.5 with what they term a "persona" prompt does in fact pass a standard that seems to me at least a little harder than what you said - actively picks the AI as the human, which seems stricter to me than being "unable to distinguish".

It is a little surprising to me that only that one LLM actually "passed" their test, versus several others performing somewhat worse. Though it's also not clear exactly how long ago the actual tests were done - this stuff moves super fast.

I'll admit that I was not familiar with the strong version of it. But I am still surprised that nobody has done that. Has nobody even seriously attempted to see how LLMS do at that? Now I might just have to check for myself.

I would have presumed it would be a cake walk. Depending of course on exactly how we define "average interrogator". I would think if we gave a LLM enough pre-prepping to pretend it was a human, and the interrogator was not particularly familiar with ways of "jailbreaking" LLMs, they could pass the test.

I've tended to agree with this line of argument, but on the other hand...

I expect that anybody you asked 10 years ago who was at least decently knowledgeable about tech and AI would have agreed that the Turing Test is a pretty decent way to determine if we have a "real" AI, that's actually "thinking" and is on the road to AGI etc.

Well, the current generation of LLMs blow away that Turing Test. So, what now? Were we all full of it before? Is there a new test to determine if something is "really" AI?

I did run through the whole thing. It doesn't actually say anything at all about what is or is not vulnerable, it's just a bunch of military standards and procedure about how to test for things and what to test, and some vague stuff about potential ways to protect things and what large military hardware they'd like to test.

Curiously, all of the links people have thrown out in this thread seem to prove exactly what I said - there's damn little available to the public in the way of documented experiments on real hardware for EMP susceptibility.

I don't have any really solid cites for it offhand, but it has been my understanding that small devices aren't vulnerable. I don't know EMPs specifically, but I have been involved in standard EMI testing for approval of consumer-grade electronics, so I know there's already a fair amount of testing for and shielding against EM interference with everyday consumer electronics.

Note that this testimony is a persuasive presentation, an effort to persuade congress to allocate additional spending for EMP protection. Mr. Wood may indeed be qualified to speak on the subject, but this aspect necessarily colors his speech. It's not like there's a shortage of cases of decently qualified people exaggerating the risks of something in order to get more funding and status for themselves and their pet programs.

What I'd really like is hard data on what is or is not actually vulnerable to these hazards and to what extent, based on hard proof rather than fear-mongering by interested parties, which this doesn't get us any further towards.

Okay, that's very cool! I don't mean to trash this guy at all, he may well be a real expert in some fields.

However, it is a failure mode that people who are really smart and qualified about one thing can assume they are equally smart and qualified about a bunch of other fields that require their own specific expertise. Alas, it doesn't work that way.

There's nothing wrong with not knowing about some specific technical subject. It is a red flag though when someone takes it as a threat to their identity and self-worth to acknowledge that they don't know much about some particular subject, even if they do know a lot about a different subject.

I'm not sure about most of this. The great majority of the articles and stories about this I've read trace back to layman speculation and disaster porn fiction written by people who have never claimed to actually be informed about how these things work. There's damn little stuff out there that traces back to actual experiments with real hardware. Probably most of the serious experiments are by various militaries and are highly classified. I've seen some more believable stuff suggesting that most consumer electronics and automobiles are not vulnerable at all to the much-fictionalized high-altitude nuclear EMP.

Either way, the author of this article does not cite any sources or relevant experience, and he doesn't include any biographical information about himself to judge how qualified he is to speak on such subjects. There's not much reason I see to take this any more seriously than any piece of fictional disaster porn you could buy on Amazon.

I don't know the truth for sure myself, but hopefully we all know better than to believe everything we read, especially about subjects like this where there appears to be very little hard science published.

In my opinion, never mind that stuff, it's enough of a hassle just getting good quality pictures of the thing, figuring out exactly what it is and what a wild guess of a "fair" price would be, writing up some sort of description, dealing with questions and appointments from potential customers. It makes it not worth the time and bother unless you're pretty sure you can get a few hundred bucks for it. Modulo of course how rich or poor you are right now and how much free time you have etc.

To be fair, my impression is most people have highly contradictory desires along these lines.

They say they want to be able to walk to places more. But they also want a big suburban-style house with bedrooms for everyone and storage and garage and lawn etc, easy parking for them, nice wide roads to drive everywhere on and tons of free parking when they get there. This makes it impossible for the area to be walkable unless everyone else lives in small apartments and there's actually only enough parking for just them to drive if they feel like it.

In my opinion, it doesn't work that way. Yeah everyone wants to be the special 1% like that, but only actually 1% will be. If you really want to be walkable, you personally will need to live like that too.

Did you mean to post the same link for both?

I find it very telling that there's no 2 responses to this post recommending the same thing. Confirms my belief that there is no real alternative to Wordpress for a free and open-source CMS that is straightforward to install and usable to build and edit pages by non-tech-experts.

Ditto to self-hosting wordpress works fine with standard hosting practices and not installing a bazillion random plugins.

The technical aspects of this seem concerning. I'm wondering exactly who has the authority to set a car to this mode and how the mechanism of doing so works. What happens if you sell the car or let somebody else borrow it? Or when you get another car, or rent one? What are the failure modes of it, like if the GPS glitches a little and decides you're on a feeder or surface road when you're actually on a freeway? I think we've all seen GPS guidance devices do that.

If this is actually being implemented as widely as the article suggests, I guess we'll all find out the answers to these questions pretty soon, the hard way!

I don't really see the hyperbolic language about it. I've definitely worked some process-heavy corporate jobs where I could be perfectly satisfactorily productive as far as my management was concerned while only doing actual work like 20-30% of the time. Maybe I'm super smart or maybe they're just terrible at generating and assigning tasks, I don't know. Not my problem either way.

Especially in the WFH era where it's much easier to get away with it clean, I don't see anything all that wrong with working 2 or 3 such jobs at the same time. If all of them are happy, or at least not too terribly upset, with your performance, what's the harm. There's definitely been times in my life where I could see myself doing that just for the sake of being bored.

I know the feeling. I made the switch a few months ago, and haven't regretted it at all.

It's funny in a way the reluctance people have to part with relatively small amounts of money like $5 or $10 a month for something so core. Given the amount of complaining online about being the product instead of the customer, you'd think more people would jump at the chance to establish the proper relationship with their primary source of information on the internet for the equivalent of the price of a cup of coffee or a fast-food meal.

Just that alone is easily worth the cost, not to mention the drastically better search results, and the ability to customize things intentionally, based on what you actually want, versus what Google thinks you should want, or more like what their actual customers, the advertisers, think you should want.

It's been well known for a long time that Google extensively customizes search results per individual based on dozens of unknown factors. It's entirely possible they get those particular search results right for one person and wrong for someone else, and nobody will ever know why.

Which makes things worse in a way - just like with LLM AIs, you can never be sure that what they give you is actually right unless you already know, so there's much less point in using them at all.

I'm not entirely sure what to say to someone who genuinely hasn't noticed that Google search results have gotten a lot worse in the last 5 years or so, regardless of whether any individual search result was right or wrong.

> We could have had a steam engine during the Roman Empire and thus an Industrial Revolution thousands of years earlier, but no one did it. It was just a toy. Imagine having the vision.

No, we couldn't. The Romans were missing several other critical technologies that were necessary to make steam engines more than a toy.

https://acoup.blog/2022/08/26/collections-why-no-roman-indus...

I do agree that it's probably nowhere near the top of the list of issues preventing this thing from working at least as well as conventional drilling technology.

However, anything about radar, ultrasound, or laser ToF would require electronics at the head of this waveguide and a way to communicate data to the surface. From what they're saying, the downhole environment of this thing is going to be very high temperature. Physically vaporizing 100% of the rock to make hole tends to do that. Conventional oilfield electronic tools already have trouble getting the MTBF above a few hundred hours at current downhole temperatures, which are much cooler. It seems likely that no electronics would survive at all at the temperatures they're planning on running.

That was answering silly speculation about intentionally leaving worn-out drill bits behind. There are actually workable procedures for drilling around things, but it's far more cumbersome and time-consuming than replacing a bit normally. Those procedures used for things like the drillstring getting stuck due to hole collapse; nobody would do that just for a worn-out drill bit.

The actual way to avoid the problem is by using bits whose design lifetime in the formation you're drilling is at least as long as the hole section you're planning to drill.

Oh, I just thought of another issue too. A liquid well-control incident with this thing would indeed suck for the reasons given, but there's a lot of gasses down there too. What happens if there's a gas well-control incident?

It could be flammable natural gas. It may or may not burn or explode in the wellbore, since there's not going to be much oxygen down there. How about at the surface though? Flammable gas erupting out your wellbore with this system sounds very not fun. They have megawatts of electricity flowing around, do you think all of that meets industry standards for avoiding explosions in an environment of flammable gasses? I think there's high potential for a very big boom, and maybe the whole well turning into a giant blowtorch you have no way to control.

Or it could be a poisonous gas like H2S. Poisonous gasses billowing out of your wellbore with this system also sounds like a major pain.

So, who wants to come up with a practical way for this thing to deal with that too? The oilfield has proven methods for preventing it in the first place and dealing with it if it happens anyways. Trip your annular blowout preventer, evacuate the rig, and circulate heavy kill mud until the gas stops flowing.

Maybe these guys could flood the well to stop it. Which means they also need to keep many tankers full of fluid on-hand, and after it works, they're back in the initial situation of needing to figure out how to seal the leak and evacuate the fluid again. I seriously can't think of a good way to do any of that.

Note that diamond (usually called PDC) drill bits have been in common use in the oilfield industry for decades now. Search PDC on your favorite search engine, and you'll find dozens of manufacturers actively selling them, each with a big selection.

Exactly how long you can drill with one varies widely based on a bunch of factors. You do have to pull the whole drill string to change one. It's slow, but not that slow. Most actual oil wells drilled have in the neighborhood of 10 or so trips in and out with drilling tools for the whole operation for various reasons. Varies widely of course depending on a bunch of factors, but that's usually the ballpark. Plus a few more for casing and cementing runs.

Physical support isn't actually that important - conventional wellbores are not physically supported either until they are cased and cemented, and mostly don't have too much trouble with collapsing. What they need is a seal tight against liquid and gas to prevent it from leaking into the wellbore.

Conventional wellbores accomplish this with the hydraulic pressure of the drilling fluid. These guys can't have any fluid though, so they would have to rely entirely on this condensed rock stuff to both support against the pressure and seal against any leaks. Seems very unlikely, considering that it isn't deliberately created by any kind of process, just randomly condensed from rock vapors.

Note also that they won't really start to run into trouble with this until they get at least a few hundred feet down.

Also, you definitely aren't going to drill more than 6 inches while attempting to physically support the wellbore with any part of the drillstring or waveguide or whatever they're calling this thing.

People should also understand that oil drilling is a highly competitive multi-trillion dollar industry employing tens of thousands of smart people all around the world. Absolutely everything that anyone could think of has already been tried, and adopted if it worked and abandoned if it didn't.