mokus

I think the bigger point is that it’s like a guy in the room. ChatGPT shows that the industry is graduating from Turing test territory to IQ test territory, and a lot of people are really hung up on the fact it’s failing the latter.

I’ve been in car crashes before but that doesn’t mean I’m ok with increasing their frequency and severity.

That article only says it’s the first this century.

Perhaps if they try to force you to arbitration you can push back and demand proof that you agreed to their terms, which is conveniently achieved by creating an account.

I’ve been using iPhone 6S since it was the flagship and I still don’t know any 3D Touch operations or shortcuts. I actually had to look up just now to double-check that this phone even has it and it isn’t disabled. I’m sure I could find a cheat sheet if I cared but generally I think of “intuitive” as being something you’d discover naturally at least once in nearly 7 years of daily use.

It doesn’t matter what it’s about though - if they were first at something then barring time travel they will always be the first, no matter how many more do it.

This makes me wonder if there might be a project out there that can compile Verilog to a schematic or netlist for 74-series logic.

Edit: unsurprisingly, the answer is yes - http://pepijndevos.nl/2019/07/18/vhdl-to-pcb.html

I hear this a lot, but what jurisdictions and why? On the face of it, it seems batshit insane to me that there would be a place where I can’t waive my property rights on something.

They have way more patience than I do. I’d have pointed out pretty quickly that this “customer” is not a customer of the gfortran team and is therefore irrelevant to the discussion.

I feel it’s even more accurate to say they are human behaviors that take advantage of religion (or any other available power structure). In a world dominated by religion they take religious form. If we remove religion from power or power from religion, they will take whatever form allows them to thrive in the new system.

This really, really angers me. Even with people i kind of agree with.

When lazy or dishonest or weak arguments are used, I get more upset if it’s in support of things I agree with than things I don’t.

It undermines the position and makes it that much harder to discuss it in the future because I now have to disentangle from that on top of actually making a proper argument in favor of the position.

This article repeats “安谋科技” so many times but as far as I can tell, doesn’t mention even once how I might attempt to pronounce it or recognize when anyone else mentions it in the future. Can anyone at least provide a transliteration?

The “nudge” can also come up even with exact starting information. When you calculate the next step and round it to the nearest nanometer, you’ve just nudged the system by enough to eventually make your prediction worthless in the 3-body case.

How do we know we aren’t the colonists?

Personally, I just want to know the cost of services before committing to pay. I have been unable to get that in many situations, even for non-emergency care.

The “exact” version they wanted was full of approximations too. They just didn’t have enough numerical literacy to understand how to say how much approximation they are ok with.

I guarantee nothing in anyone’s time accounting system is measured to double-precision accuracy. Or at least, I’ve never quite figured out the knack myself for stopping work within a particular 6 picosecond window.

This is a bit of a tangent, but that makes me wonder… are there any languages where this particular collation would be correct?

Thinking you can buy a business relationship with millions of people on a network run mostly by volunteers and expect everyone to start operating it and using it for your benefit instead of for the ideals of their 20+ year community sounds a lot more entitled to me.

The whole idea of purchasing a business for its customer base is pretty disturbing in the first place. I as a customer do not consider my relationship with any business to be a transferable asset.

A mechanic is in a good position to notice that when a car fails, it tends to fail in certain ways. Far too often, though, they then generalize to claims that the car is likely to fail in that way, but that is not necessarily the case. It may be that they fail 1% as often as competitors and due to a relatively fewer number of causes. Without being in a position to see the absolute failure rate, a mechanic may incorrectly conclude that those parts they see failing are especially badly designed even if they are more reliable than the industry average.

Absolutely agree, but that’s a whole new can of worms and outside the point I was making (which was that it is not obvious that they made this change for the SOLE purpose of preventing repairs). Do we force them to sell the M1 module as well? Individual chips from that module? I’m all for that, personally, but good luck getting that passed. And assuming we can’t make them do that, what is the line between those parts and the less special ones like this one?

I understand about the frustration of not being able to source the part, but a change in a single IC or connector’s pin mapping absolutely can be a performance improvement alone.

I’m lobbying at work now for one such case. Sometimes the existing pin assignments are just bad, and that can destroy signal integrity, increase layout complexity (and thus board cost) and negatively affect EMC performance. I don’t know whether that is the case here but I wouldn’t automatically trust someone’s judgment on it without knowing their level of expertise on high speed layout design.

It still doesn’t suck?

Saleae Logic. Their USB3 “pro” analyzers are fantastic and the software is clean, fast and user-friendly. Using host RAM for capture buffer is great and completely changed my workflow. I can capture hour or days of trace and sift through after a fault occurs.

We have the usual other test equipment from the usual big names, stuff that costs 40x as much. For many jobs, I’d rather have the Saleae.

The entire model many large web companies use of automating moderation and support to extreme degrees is, I feel, fundamentally wrong. I wonder what the ratios of customers to support staff look like for all these multi-billion user platform companies.

Did any one person say both of those things?

A lab I used to work at had a light field microscope system. Sorry, though, its existence is pretty much all I know about it since I never personally got to use it.

Isn’t that a federal crime in the US under CFAA?

AVR does this. I can attest to the “interesting” debugging...

DDG makes me nervous because I don’t actually understand their business model. Which isn’t to say they don’t have a well-known and viable one but I haven’t personally looked into it and as a result my gut feeling is that they are probably not an exception to this.

I use them anyway because they at least claim to be private and haven’t yet given me specific reason to doubt it. I probably should at some point take the time, though, to try to actually understand how they can viably exist in a way that isn’t going to succumb to the same corrupting incentives as google.

The fact that Facebook appears to consider user consent an existential threat to their business model tells me everything I need to know about them right now.