2025-12-05 10:09:39 This is actually really cool https://en.wikipedia.org/wiki/Operator-precedence_parser#Full_parenthesization 2025-12-05 10:10:54 How Fortran I did operator precedence 2025-12-05 10:11:22 In 1957 2025-12-05 10:41:31 I put a lot of thought into reducing overhead for accessing the stack on my calculator forth since a crash takes down the whole system. I've been using a circular stack and only just now realized I could do just one check at the beginning of every block like the body of if or do 2025-12-05 10:42:37 just have to put down a hidden primitive at the beginning of the block then update its argument when the word ends or a word occurs where the stack picture can't be predicted 2025-12-05 11:14:09 Can you explain that a little more deeply? I'm not quite clear on what you mean, but I'm interested. 2025-12-05 11:16:45 for DROP, for example, you need at least 1 stack item, otherwise it's an overflow 2025-12-05 11:16:51 *underflow 2025-12-05 11:17:51 so you can have an if statement in the DROP primitive to check the stack depth but it's probably faster to just AND the stack pointer with a mask and add it to the base so that you get a circular stack 2025-12-05 11:18:42 so if the stack address is 0x4000 and your index is 0, then doing DROP makes the index -4 (assuming a 32 bit stack) which is now out of bounds 2025-12-05 11:20:44 but if you AND -4 with a mask like 0xFF if becomes 0xFC 2025-12-05 11:21:37 and the address of the stack item is 0x4000 + 0xFC which is the last stack item 2025-12-05 11:22:08 which avoids a stack underflow and also avoids the overhead of checking the stack before every DROP 2025-12-05 11:23:34 what I'm thinking about now is when something like `: foo X 5 + DUP ;` appears, lay down a primitive after foo starts that checks the stack depth followed by an argument for the expected stack picture that is later overwritten 2025-12-05 11:24:16 then record what the next 4 primitives do to the stack and write the expectations to the argument 2025-12-05 11:24:47 so that when foo starts, the check is done once and everything else there runs with no check 2025-12-05 11:24:56 my brain just melted trying to follow MrMobius 2025-12-05 11:27:26 lol 2025-12-05 11:28:12 I get the need to check that the stack has an argument for a DROP tho, you made that clear 2025-12-05 11:29:05 and I get the general idea, man there is a LOT to writing an efficient Forth! 2025-12-05 11:29:34 tp_: in the example of `: foo X 5 + DUP ;` if x is a variable and pushes one value then we need at least two spots left on our stack not to overflow, right? 2025-12-05 11:30:42 yes 2025-12-05 11:32:04 one stack spot is the variable (X) + 5 the other is the duplicate 2025-12-05 11:32:05 in order to prevent overflow, the code for X and the code that pushes 5 both check for overflow. also, the code for + checks for underflow and DUP checks for overflow. why not combine all of those into one check before those 4 things run to save time? 2025-12-05 11:34:02 I dont understand the process well enough to say 2025-12-05 11:34:21 because you could replace those four checks with just one check at the beginning to make sure there are two spots left 2025-12-05 11:35:29 I'd tend to have a stack check associated with each Word individually I think 2025-12-05 11:36:31 something that is self sufficient and accomodates all possible variations ? 2025-12-05 11:37:01 but I definitely dont have a Forth implementors mind 2025-12-05 11:37:22 Im just a Forth user ... 2025-12-05 12:21:39 : foo X @ 5 + DUP ; 2025-12-05 12:22:36 forthBot: VARIABLE X 2025-12-05 12:22:51 forthBot: 5 X ! 2025-12-05 12:23:55 forthBot: X @ . 2025-12-05 12:23:55 5 2025-12-05 12:24:19 forthBot: : foo X @ 5 + DUP ; 2025-12-05 12:24:32 forthBot: foo .S 2025-12-05 12:24:32 <2> 10 10 2025-12-05 13:06:04 Ugh - I'm getting so weary of the anti-AI-content crowd in the Reddit community I help moderate. We had a huge discussion about this a couple of years ago, and even ran a community poll to give the community an opportunity to support a prohibition of AI content. The vote failed, so our moderator position is that unless and until there are either laws about it or a higher level Reddit policy about 2025-12-05 13:06:05 it, the content is allowed. We did add a "flair" that should be used to label the content, so people who don't want it can easily avoid it. But the folks who lost just won't stop sending in "reports" on almost every post made that has AI content in it. One popped up yesterday that was nothing more than a casting suggestion for one of the characters, but they'd used AI to put the actor in a 2025-12-05 13:06:07 character related costume. I've had to reject about a dozen reports on it so far. 2025-12-05 13:06:28 Meanwhile, it's been upvoted going on 200 times, but that doesn't seem to matter to these people. 2025-12-05 13:10:45 I've become rather torn over this whole issue. When it first gurgled up a while back, I initially felt like the "anti" case resonated fairly well. But since then I've thought more about it, and currently I rather lean the other way. AI image generators don't literally pick up pixels out of anyone's art and move them to the generated image. And think about how a human artist develops. Most of 2025-12-05 13:10:46 them study the work of other artists, draw inspiration, study techniques and methods, and then apply all of those things to their own work. It feels to me like what AI generators do isn't too horribly different from that. The output is INFLUENCED by the inputs, but not "copied from" the inputs, and that's pretty much the case with a human artist too. We don't accuse them of "stealing the work" 2025-12-05 13:10:48 of the artists that they drew influence and inspiration from. 2025-12-05 13:12:04 I'm coming to feel like the whole thing is really just yet another "industry" getting worried that computers are becoming capable of doing their work. 2025-12-05 13:12:59 I totally get that that is a concern to them, but it's hard for me to see that it's very different from, say, the impact PowerPoint had on the graphic arts business, or that spreadsheets had on the bookkeeper profession. 2025-12-05 13:13:20 Just another way software tools are lowering the demand for this or that profession. 2025-12-05 13:14:56 But most of the lay public probably doesn't understand that the image generators aren't literally copying input data. They don't really understand how the tools work. 2025-12-05 13:15:22 So it's easy to "sell them" on this "stealing argument." 2025-12-05 13:18:29 It's actually sort of an interesting question in general - where does "inspired by" end and "stolen from" begin? 2025-12-05 13:18:56 In extreme cases we know it when we see it, but I have a feeling there's a very blurry zone in there somewhere. 2025-12-05 13:24:04 The real copyright issue is they are unlawfully copying a lot of content to train 2025-12-05 13:24:30 If I downloaded loads of copyrighted works to put on my PC I'd get in trouble, in theory, especially if I was a business 2025-12-05 13:24:41 But if you're big enough then you can just sort of ignore the rules 2025-12-05 13:25:09 The fact that they're not distributing it verbatim is irrelevant because the copies they make initially to do the training are unlawful 2025-12-05 13:25:25 They have no license to do this 2025-12-05 13:30:12 That's a fair point. But it's something that could be taken to court, of course. And what a mess that would be - how would you prove that? Maybe we wind up with an arrangement where the training process has to be documented in some way, etc. 2025-12-05 13:30:22 An audit trail of the materials used for training? 2025-12-05 13:30:52 The best is when the AI is trained on stock images and thinks that's supposed to be in there then either puts a company's name or icon on the generated image or something resembling them 2025-12-05 13:31:48 :-) Cases like that would be easier to resolve, I guess. 2025-12-05 13:31:50 Also funny with the recent drama about a guy who used AI to generate art then blasted other artists for not being original because their AI art looked like his AI art so he accused them of stealing his prompts 2025-12-05 13:32:19 Oh gosh. Prompt IP. 2025-12-05 13:32:35 I hadn't thought of that, but why not? We fight over everything else, don't we? 2025-12-05 13:33:59 But I guess crafting clever prompts would qualify as a skill. 2025-12-05 13:34:07 I remain dubious 2025-12-05 13:35:06 Yeah, same. I'm just having trouble making a well formed case aagainst the idea. It's a little like clever search queries. I had a boss once that commented one day that I was particularly good at finding useful things online. 2025-12-05 13:35:59 It just never occurred to me to think about it as something that might be "protectable," but like I said - those things do seem to qualify as "skills." 2025-12-05 13:36:16 I think we get into the same sort of arena as not issuing patents for "obvious" inventions. 2025-12-05 13:37:05 Very short works aren't usually copyrightable I think 2025-12-05 13:37:34 So prompt IP is going to depend on how detailed it is probably, but also if it's just 'inspired' there's no protection at all 2025-12-05 13:37:44 Yeah, it seems like we need some "practicality lines" drawn somewhere, just to keep things from bogging down in their own complexity. 2025-12-05 13:38:41 I definitely think "downloading and distributing your copyrighted work so it can be used to generate other work / revenue" is something you need a license to do 2025-12-05 13:38:47 Sounds like a class action waiting to happen 2025-12-05 13:38:49 At some point we just stop focusing on the nitty gritty legal wording and "be sensible" about things, and I guess figuring out where that is will wind up in court like it always does. 2025-12-05 13:39:35 Yes, I do think a creator should be able to control whether his or her work is used in that way. Definitely. 2025-12-05 13:39:46 I think there's a strong precedent to say that if they're not obeying licenses and processing the art to make money then they're violating copyright 2025-12-05 13:40:08 And a massive class action is the only way that could really go to court 2025-12-05 13:40:19 Because any individual damages are going to be extremely small 2025-12-05 13:40:32 Yes, I agree with that too. But how do we show that they did use a particular work? At the current time they generally don't share their training data, do they? 2025-12-05 13:40:40 Discovery 2025-12-05 13:40:50 Like I said, we may come to a point where that has to have an audit trail. 2025-12-05 13:40:51 Wouldn't the same short prompt produce the same AI art if it's run on the same AI by different people? 2025-12-05 13:40:56 So that discovery could even work. 2025-12-05 13:41:05 MrMobius: Nope it's random 2025-12-05 13:41:16 Right - there are random seeds in play too. 2025-12-05 13:41:34 The algorithms used by these 'AI' tools are non-deterministic 2025-12-05 13:41:58 Although often the result is surprisingly deterministic 2025-12-05 13:42:13 Putting in a system that would allow such an audit trail would almost certainly increase training costs, and that's the real thing they're trying to avoid. 2025-12-05 13:42:32 Or, wait - that's the thing it's LEGITIMATE for them to wish to avoid. 2025-12-05 13:42:32 Especially the output of LLMs, e.g. "write me a cover letter" produces the same slop every time and I know you did it so you go in the bin 2025-12-05 13:42:49 They're also trying to avoid being vulnerable to getting into trouble if they break the rules, but that part is illegitimate. 2025-12-05 13:43:31 someone emailed us at work and at the end of the email had left the prompt saying "make it a little more professional in tone" 2025-12-05 13:43:39 The lack of audit trail isn't what's illegitimate though, the technique they're using guarantees they will violate copyrights 2025-12-05 13:43:44 lmao 2025-12-05 13:43:52 lol 2025-12-05 13:44:19 I would have emailed back saying "I'm sorry you don't think my tone is professional enough, I will try harder sir" 2025-12-05 13:44:30 :-) 2025-12-05 13:44:36 lol 2025-12-05 13:44:40 my bad 2025-12-05 13:45:55 Oh, and just to be totally clear about that Reddit thing I started with, of course the guy who posted it isn't making any money off of it. At least not there on Reddit. 2025-12-05 13:46:24 He was just having fun - he thought that actor would be good for that character (I agreed) and thought it would be fun to put in "in character." 2025-12-05 13:46:33 put "him" 2025-12-05 13:46:54 Just a totally innocuous thing, and the reaction it provoked was fairly ridiculous. 2025-12-05 13:48:09 veltas: are random processes involved in training too? 2025-12-05 13:48:21 Or will the same training data produce the same weights if you re-run? 2025-12-05 13:48:42 Of course you could document the random seeds too - surely we could get to a point where a training process could be repeatable. 2025-12-05 13:49:13 Then the whole audit trail thing could work: Here are my weights, and here is how I got them. 2025-12-05 13:49:41 Yeah I believe the training is ND 2025-12-05 13:49:44 But they regard a lot of that as proprietary, don't they? 2025-12-05 13:50:49 But we've fought this battle before, with souce code and application binaries - seems like we ought to have a legal template for how to strike a balance between the necessary openness and the necessary confidentiality. 2025-12-05 13:51:50 And given that they just crawl the net to find the training data, there's going to have to be a way to have everything marked as to whether it's usable as training data or not. 2025-12-05 13:52:13 That license has to be accessible and readable by the training tools. 2025-12-05 13:53:52 And even if we eventually get everything in place for all that, what do we do in the meantime? 2025-12-05 13:54:06 Other than rage at one another. 2025-12-05 13:55:20 Oh, and let's go back to the human example. Should an artist be able to say "You can look at an enjoy my art for pleasure, but you're not allowed to carefully study it to try to learn my methods." 2025-12-05 13:55:28 Can we ban HUMANS from doing that? 2025-12-05 13:56:16 Because those things seem rather similar to me - that's the point that has caused my thinking to change on this to some degree. 2025-12-05 13:56:53 A case can be made here that what the software is really doing (as it's being trained) is emulating the mental activities of a human learning the craft. 2025-12-05 13:57:25 It's just doing it in a precise numerical way rather than a fuzzy mental way, but it's actually the SAME THING. 2025-12-05 13:58:46 If I were to sit down this morning and start trying to write an urban fantasy novel, I'd almost certainly find myself emulating aspects of the style Jim Butcher uses in The Dresden Files. At what point would I be stealing from him? 2025-12-05 13:59:38 Like I said earlier, extreme cases would be obvious. But what if I just imitated his general pacing and so on? Is that stealing? 2025-12-05 14:04:02 So one way of looking at it is "We're trying to design software that mimics the development of human expertise. Are we going to subject the software to requirements that we don't subject humans to? If so, how do we justify that?" 2025-12-05 14:06:35 I think the whole thing is a lot deeper than any one surface issue. I'm always reminded of Dune and the "Butlerian Jihad" in its back story - that was a situation where their culture got into real trouble with AI, and it eventually led to a violent upheaval and an utter rejection of the whole technology. They got a "commandment" out of it: Thou shalt not make a machine in the image of a man's 2025-12-05 14:06:37 mind. 2025-12-05 14:08:33 So you saw all kinds of fancy tech in Dune. Spacecraft, lase guns, shields, etc. But no computation / data processing. It was forbidden. 2025-12-05 14:10:44 KipIngram: The Software/Binary precedent is much more damning for 'AI' because we're saying machine data generated *from* a copyrighted work should be copyrighted too 2025-12-05 14:11:09 So if that's a fair precedent then the 'AI' content should be joint owned by the owners of the source material 2025-12-05 14:13:10 As for the difference between machine learning and human learning, the difference is that a human apparently has a right to view and learn from art, but the precedent with machines is that their data is another kind of 'copy' so needs licenses 2025-12-05 14:14:07 And copyright is there to grant rights to authors or businesses that generate copyrightable work, not machines 2025-12-05 14:15:07 Why should the state give rights to machines that impact the rights of artists/authors? 2025-12-05 14:17:30 I think there's a balance of competing interests that will influence states, and the most compelling is adversarial 'AI' 2025-12-05 14:17:54 A state will prefer to leap ahead and support 'Ai' as much as possible because they don't want to lose that race, it could be existential 2025-12-05 14:24:32 Environment for cleobuline inactive, freeing... 2025-12-05 14:42:04 Yes, I think precisely what I'm questioning is the actual basis for discriminating between humans and machines in that way. If the machine is essentially modeling the human learning process (and I think a fair case can be made that they're at least extremely similar in abstract), then why is one ok and the other not? Of course, we're allowed to just arbitrarily declare that, but we should be 2025-12-05 14:42:06 clear we're doing it. 2025-12-05 14:43:01 And I think we should be clear on why. Just to be clear, I'm not really expressing "support" for AI art here - I've just done some thinking on the deeper aspects of the issue and am sharing them, that's all. I don't think it's as simple and obvious as people make it out to be. 2025-12-05 14:44:18 I don't think what the AI tools are doing is "clearly and obviously wrong," because I think it's highly similar to what human learners do. We look at a lot of examples of what we're trying to learn (focusing mostly on art fields here), and somehow we "internalize" that in a way that allows us to produce new work of the same nature. That's PRECISELY what the AI tools are doing. 2025-12-05 14:45:46 We can just decide we're not going to allow it in spite of that similarity, and I won't be broken hearted either way - I don't really care. But I would regard it as an arbitrary distinction we'd made. 2025-12-05 14:47:21 I think you're being philosophical 2025-12-05 14:47:32 And I do think it's quite different from the source->binary step in application compiling. That's a direct conversion, and I really regard the source and the binary as the same thing conceptually. The AI art thing is different, because there's no such one-to-one conversion. 2025-12-05 14:47:41 I absolutely am being philosophical. 2025-12-05 14:48:01 I'm trying to skin it down to its bare bones as much as I possibly can. 2025-12-05 14:48:51 And I'm being realpolitik 2025-12-05 14:48:52 I'm trying to get down below the surface level issue that people are upset about - I've never believed in dealing with controversial issues in a case-by-case way. We need some system of values and principles that we base our decisions on. 2025-12-05 14:49:26 People who just kneejerk decisions case by case usually wind up being wildly inconsistent and can't point to ANY values other than "brazen self interest." 2025-12-05 14:50:24 But self interest is never going to align across a whole population, whereas values at least potentially can. 2025-12-05 14:50:50 Values are a luxury 2025-12-05 14:51:17 I think they're absolutely necessary for the success of a culture long term. Without them we're apt to tear ourselves apart. 2025-12-05 14:52:36 It's interesting you think that given that most things people hold as dear values today weren't guaranteed for most of human history and yet here we are 2025-12-05 14:52:55 It's hard to have them, though - holding values can sometimes mean making decisions you don't actually like in that particular instance. 2025-12-05 14:53:20 "Doing the right thing, even when it hurts." 2025-12-05 14:54:20 Yes, and those were some extremely chaotic times. I don't think that the human race would vanish without values. 2025-12-05 14:54:57 It's hard to get rid of all of us - honestly I don't think even a plague or a nuclear war would. SOME folks would make it through, and they'd... start over. 2025-12-05 14:55:06 But ouch. 2025-12-05 14:55:42 I think personally our 'values' are a luxury afforded by success, rather than the other way round. If we want to keep them we definitely need to improve the economy, productivity etv 2025-12-05 14:55:46 etc* 2025-12-05 14:56:10 Yes, I think we've mismanaged a number of things for sure. 2025-12-05 14:56:25 In the UK we're discussing removing jury trials for about half the current cases that would go to jury 2025-12-05 14:56:43 It's not that the economy isn't productive, but it's structured in an ugly way. Too much power centralized in huge corporations. 2025-12-05 14:56:46 In my opinion. 2025-12-05 14:56:51 This is a right guaranteed by the king in the Magna Carta 2025-12-05 14:57:11 Yes, I worry about some of the stuff you guys are doing too. 2025-12-05 14:57:17 And we can't afford it anymore so we're binning it for half those cases where people would like their peers to have the final say because they worry the system will gobbel them up 2025-12-05 14:57:45 And freedom of speech seems to be in real trouble over there, though I may be poorly informed. 2025-12-05 14:57:49 The truth is I might do the same in Lammy's position, because another concern is our criminal cases are taking so long that a lot of them fall apart 2025-12-05 14:58:17 Rather than poorly informed you may be succesfully propagandaad 2025-12-05 14:58:49 You've got competing rights and ultimately our state is failing to provide some 2025-12-05 14:58:56 Fair enough. I think it's really hard to get accurate information, in spite of our miraculous planetary communication network. 2025-12-05 14:58:57 So we need to pick and choose 2025-12-05 14:59:00 We use it to lie to one another. 2025-12-05 14:59:30 If we have a more effective economy and state one day we may be able to restore the right to a trial by your peers 2025-12-05 14:59:32 Manipulate one another's emotions. 2025-12-05 14:59:54 The US has a very different precedent for freedom of speech than the UK 2025-12-05 15:00:20 It's a matter of competing rights, in the UK we have more rights against slander/libel, but less freedom to say whatever we want 2025-12-05 15:00:27 It's been that way for a very very long time 2025-12-05 15:00:49 I personally like our balance 2025-12-05 15:01:13 I think it's good that you can't accuse people of damaging things without backing it up in court, lest you get sued 2025-12-05 15:01:18 Well, we have slander and libel laws here too - the freedom of speech thing isn't universal and that balance has been worked out in court over time. Often going all the way to the Supreme Court in particularly difficult cases. 2025-12-05 15:01:26 And you can't incite people to violence 2025-12-05 15:01:40 If you say something that damanges someone, and they can prove it was false, then you can get convicted of slander. 2025-12-05 15:01:50 But... if it's true, then you can't. 2025-12-05 15:02:13 In the UK it's different, the onus is on the person making statement to prove it's true 2025-12-05 15:02:28 If you can't prove the statement is true then you are liable 2025-12-05 15:02:57 Honestly I'm not 100% sure which way that goes here. Not sure who has the burden. 2025-12-05 15:03:17 In the UK your reputation is innocent until proven guilty, so if someone slanders you the onus is on them to prove it 2025-12-05 15:04:20 That doesn't necessarily seem wrong to me - particularly given the penchant people have for just brazenly lying online. 2025-12-05 15:04:52 It doesn't seem unreasonable at all to me that if you're going to say something you should be prepared to document it somehow. 2025-12-05 15:05:18 I think if people understood the UK legal system better then there would be less people saying we are losing our rights 2025-12-05 15:05:33 But people love rage baiting etc 2025-12-05 15:06:02 Sure - the certainly are, on all fronts. 2025-12-05 15:06:21 There are probably many parallel things going the other direction across the pond. 2025-12-05 15:06:26 People say you shouldn't be arrested for a tweet... well I think it depends what you say. If your tweet is offering money for someone's head maybe that should be illegal 2025-12-05 15:06:28 And then of course there's the ever present problem of money. 2025-12-05 15:06:42 The truth is there is a lot of nuance and nobody is interested in that 2025-12-05 15:06:44 Someone with money can haul you into court, and can ruin you financially whether they can actually "win" or not. 2025-12-05 15:06:50 And I have no idea what to do about that problem. 2025-12-05 15:06:51 They just love moaning about the UK 2025-12-05 15:07:27 I don't think the monaning is limited to the UK - pretty much everyone has SOMEONE attacking them. 2025-12-05 15:07:38 Everyone significant, at least. 2025-12-05 15:07:45 I would say the UK is by default one of the least popular countries in the world 2025-12-05 15:07:52 We just love to attack, apparently. 2025-12-05 15:08:13 Yes, it's a major Western power. 2025-12-05 15:08:27 The English speaking world gets it particularly rough, I think. 2025-12-05 15:08:30 I'm thinking about writing a compiler because a compiler I am interested in has blocked the UK completely from their website 2025-12-05 15:08:34 At least the US and the UK do. 2025-12-05 15:09:06 There are two kinds of countries, the countries that aren't important and the ones everyone complains about 2025-12-05 15:10:23 Also recently Trump threatened to sue the BBC for lying about him. I have in conversation heard a number of British people essentially saying he should, and has the right to 2025-12-05 15:10:36 But I think a lot of pro-Trump americans would be shocked to know this 2025-12-05 15:10:58 But I think most people in the UK agree that if the BBC lies about Trump then they shouldn't get away with that 2025-12-05 15:11:22 Whereas in the US I don't think Trump would have a lot of success suing media 2025-12-05 15:11:37 I don't understand the Trump "phenomena." He seems to trigger some kind of derangement in people. 2025-12-05 15:12:49 I think I understand part of it, a little - I do think the political left has been fairly ugly to conservatives in America for decades. Disrespect, disdain, and so on, and meanwhile they were winning on a lot of front culturally. It bred a lot of hurt and hate in the right. And now a lot of those people smell blood in the water and want payback. And they see Trump as offering them that chance. 2025-12-05 15:12:53 It's a very ugly thing. 2025-12-05 15:13:26 I think the hurt feelings were justified in some ways, but to respond to it with this kind of behavior... I just can't condone it. 2025-12-05 15:14:45 And the thing is SOME people on the right deserved that treatment. We have people that are hateful and racist and always have been. But the problem was the treatment got delivered to the whole of the right, not just the "bad ones." And a lot of those people did NOT deserve that kind of disrespect. 2025-12-05 15:15:25 I've watched the whole thing just get worse and worse starting in the 1990's. 2025-12-05 15:15:53 I can't help thinking that the ability to communicate on the internet has accelerated it. It facilitates "mob formation." 2025-12-05 15:16:25 Mobs compmrised of extremely uneducated people who've never paid any attention to trying to actually think these things through. 2025-12-05 15:16:48 It used to be that you didn't get a public mouthpiece unless you had some amount of "training" in the whole business. 2025-12-05 15:16:54 But now everyone has a mouthpiece. 2025-12-05 15:21:36 Maybe "training" is the wrong term there - you didn't get the mouthpiece unless you went through a sort of "filtering process." Because there just weren't enough mouthpieces for everyone to get one. 2025-12-05 15:29:11 Anyway, I think the US gets guff because, well, we're the "superpower." And the UK is pretty much our most important ally, so guilt by association. And then there may be European dynamics that get the UK targeted totally independently of the US. 2025-12-05 15:31:04 We definitely drew our own ire before the US was the main superpower 2025-12-05 15:31:08 I mean it was us before you 2025-12-05 15:31:09 I guess I better go load the crock pot up with tonight's dinner. Chicken vegetable soup, with cabbage instead of rice or pasta. Very healthy. :-) 2025-12-05 15:31:18 Sounds alright 2025-12-05 15:31:25 Right right - I'm not claiming all the credit for us. :-) 2025-12-05 15:31:34 After all, before we were the superpower, you were. 2025-12-05 15:31:53 Starting pretty much with the industrial revolution. 2025-12-05 15:32:31 It took two nasty world wars to beat you down from that, while we hunkered over here behind our ocean and didn't get bombed. 2025-12-05 15:33:08 We've been a big deal since we defeated the spanish armada 2025-12-05 15:33:24 Yes. One of my favorite history stories. 2025-12-05 15:34:08 I love how the absence of modern tech changed history - my understanding is that those two fleets passed remarkably close to one another out on the high seas a time or two. But - just out of sight, and if the battle had happened out there it might have been very different. 2025-12-05 15:34:25 A lot of big naval battles were like that 2025-12-05 15:34:32 battle of trafalgar too 2025-12-05 15:34:41 Yeah. Which one was Francis Drake? 2025-12-05 15:35:15 armada 2025-12-05 15:35:28 That's *Sir* Francis Drake :P 2025-12-05 15:35:37 Yes, of course. :-) 2025-12-05 15:35:38 Sorry. 2025-12-05 15:35:55 And well deserved, too. 2025-12-05 15:37:04 I would say the founding fathers of US would have considered that part of their heritage and the US's heritage too 2025-12-05 16:13:40 Well, yes - the heritage of the English speaking. We really owe it all to you guys. 2025-12-05 16:14:13 I regard the US and Australia / New Zealand as "offspring" of the UK, so to speak. 2025-12-05 16:14:30 Culturally, historically, all that good stuff. 2025-12-05 16:15:33 I even read a science fiction story once that mentioned that outright - the bad guys had been trying to dominate the world in a "shadow" sort of way for centuries on end, and one of them said that England had been no end of trouble, first in their own right and then for having spawned the U.S. 2025-12-05 16:16:20 England had been the thorn in their side that just wouldn't go away. 2025-12-05 16:17:00 I found the idea quite satisfying. :-) 2025-12-05 16:17:40 I wouldn't mind reading that again, but I don't know if I could possibly identify it at this point, unless that anthology still happens to be in one of the boxes in my attic. 2025-12-05 16:19:42 There's probably at least a thousand old paperbacks up there. No doubt a fire hazard, if one got started somehow. Maybe I should offer them up on eBay or something. 2025-12-05 16:21:36 That attic is accessed by a full size door on the second floor you can walk through - not a "ladder atic." It's over our garage. When we bought the house there was like a four foot square floor there for the water heater, and the rest was open beams. 2025-12-05 16:22:52 But I spent a week many years ago putting in decking, so turned the whole thing into storage space, and in one nook where the joists were particularly strong I put in shelves. The idea was to gain storage space that the house was pathetically lacking when we bought it. But... it's full. Reference the conversation from a day or two ago about too much junk. 2025-12-05 16:23:20 Toys and games from when the kids were little, etc. 2025-12-05 16:23:40 Old dishes we don't use anymore. More or less anything you can imagine that might show up in a home. 2025-12-05 16:24:39 I once thought I might convert part of it into a radio shack, but it got filled up with stuff instead. 2025-12-05 16:34:15 A good tactic is to go through and make 2 piles, stuff to keep and stuff to chuck 2025-12-05 16:34:22 And try to keep them about the same size 2025-12-05 16:34:42 Much easier to chuck stuff when you are trying to justify keeping something else 2025-12-05 16:35:38 Unfortunately most of the books are probably not interesting to anyone, or worth the effort putting on ebay 2025-12-05 17:04:10 I don't know - I just bought the whole Nero Wolfe detective series off of someone on eBay. 2025-12-05 17:04:54 My 11th grade homeroom teacher introduced me to the character - I read a couple back then. A few weeks ago it crossed my mind and I decided to revisit it. They're actually quite good. 2025-12-05 17:08:41 https://www.youtube.com/watch?v=zHIsGqJaXXw 2025-12-05 17:09:13 Wolfe is a morbidly obese genius who almost never leaves his New York City brownstone, and grows orchids in his roof greenhouse. He employs a fellow named Archie Goodwin, who narrates the stories first person, to do his running about on cases. 2025-12-05 18:47:10 Reminds me of Monk's nemesis 2025-12-05 20:13:47 There have been a couple of TV Shows over the years; in 1959 they made a pilot for one, but the network turned it down so that's all that exists. William Shatner was cast as Archie, and I thought he did a good job of it. But if the show had succeeded then we might never have gotten him as Jim Kirk, and that would have been bad. 2025-12-05 20:32:22 agreed 2025-12-05 20:32:40 Shatner was the perfect Kirk imho 2025-12-05 20:37:14 well, my great NixOS experiment is over at last, great concept, 100,000 packages, but no arm-none-eabi which I need to compile Mecrisp-Stellaris for STM32 2025-12-05 20:37:50 so today I try out FreebSD-15 which does have the Forth support apps I need 2025-12-05 20:39:15 I've had a great AI holiday the last year or so with NixOS, tried out all kinds of AI locally and remote, but now it's back to full on embedded projects for my retirement time hobbies 2025-12-05 20:41:20 I found AI interesting and novel but nothing I "must have" compared to the apps Ive used for years, 'gEDA', 'gSCHEM', 'PCB' general unit utilities, ARM cortex-m support, Forth support and editors 'nvim', 'helix' etc 2025-12-05 20:42:43 and in my areas of expertise, electronics, embedded, Forth etc, AI has little to contribute I dont already know 2025-12-05 20:45:37 I dont denigrate AI, or insist that it's useless as it obviously has tons of useful areas where it's very useful, document summaries, proofreading, document writing suggestions etc 2025-12-05 20:46:43 Ivbe used AI for web scraping and information location and the results have been very beneficial to me 2025-12-05 20:47:41 I realised today that I haven't used AI for about 3 months, havent felt the need or had something that only AI could solve 2025-12-05 20:48:12 I wonder if we are heading for a AI bubble 'bust' now ?