RugnirViking

I think we like simple and elegant solutions because we understand the perils of leaky abstractions. LLMs benefit hugely from this also; short scripts that work well can save them hours of faffing about trying to discover how to do something. I imagine if they were able, they too would prefer elegant solutions

Very busy ui. Mmo levels of stuff vying for the players attention

it seems to me that would only be true if the amount of knowledge needed to complete a task far exceeds the context limit. Why would that be the case? I find most people are using horribly inefficient contexts, and a naive approach to dark factories/software eng swarms seems to only compound this

> It's not very good, it's not optimized for maintenance or extension

Well yeah. Prototypes are not products, the last 10% takes 90% of the time, and so on.

Practically half the wisdom written about software in the last 20 years has been about the fact that just because you can make something that kinda looks like the thing at a glance very much does not mean you can make the thing. All those "small tweaks" you think you need to be finished usually turn out to require large redesigns, else you'll tank performance, stability, security, usability, or all four at once.

Meditate on the fact that people used to (and still do!) go to game jams and code games with freshly made assets in a day. I just did one myself without llms at the weekend. Whats the difference between that and making a game that people buy? Why does it take so much longer?

I really enjoy lemmy with summit client on my phone. Never been on it from my desktop though. Probably a good thing all things considered.

It's got pretty good coverage of most things related to my country and interests, but if you go too far afield/niche theres not many users on many topics

Ive been looking at educating myself on some version of the trivium recently. Never had any worthwhile humanities education on philosophy, rhetoric, history of ideas, or logic from my high school education in the UK. My degrees have been all STEM. It's hard to push through alone though, id love some kinda evening course I could do.

All this to say, I get it. Its in the air at the moment, I think.

the best part is how they don't help at all for the hard part (collecting and curating high quality data) and instead help with the part any intern with python and a stem undergraduate could have done already (make a pretty graph)

armed with the free version of chatgpt, pretty much anyone capable of understanding or composing the question can summon such a graph nowadays, given the data.

I want to offer a counterpoint to the above. I think that for some people in some circumstance it is true, but you can absolutely have fulfilling relationships with people you spend a lot of time and attention on, both friends and romantic partners. I think a lot of people get stuck in the classic avoidant + anxious attachment styles, where people chase what they can't have, but thats not everyone.

I'm not sure how much I actually agree with attachment style theory, but I do observe the pattern parent points out in some people, that they chase people who are distant or unavailable. But its often unhealthy for both people

gibberish no, made up facts probably in about half of all responses it gives me dozens of times a day, things that dont make sense (internal contradictions in a single response) multiple times a day.

I find there is quite a difference between something like "what do I do next" for a specific problem, and actually trying to learn something hard from an AI. The first it is quite good at, the second it is not. Surface level is easy, going into the weeds is difficult

I understand its hyperbole but are there people seriously trying to benchmark how good ai is at seducing people? lol

no. That's not a given at all? I think most reasonably people now agree that AI intelligence is spiky, that is to say, there are some areas it is amazingly good, and some areas where it spectacularly fails.

One of these areas it's not good at is sadly it's social ability. It's terrible at explaining things - just try asking ask it to explain code !

the author, I imagine, would object that the bridge should have specified that it is not supposed to collapse.

And if it did, that the specific load that made it collapse should have been specifically mentioned in the spec.

And if it was, well just wait until agents become the magical machine gods he assumes they are, then bridge collapses will be a thing of the past.

but let's be real, he and everyone else are machine generating their specs anyway, and the relevant spec didnt mention anything about bridges collapsing but instead 250 pages of llm waffling about the tensile strength of different structural materials, load patterns from different tyre treads, the effect of climate change on el nino and weather patterns for wind, and the bridge still fell down, because nobody had any idea what they were doing

you can't write emotions on paper either. You can write things that evoke emotion in the reader, but code can do that too. It happens quite often to me, at work. Often negative but occasionally positive

is that true? from looking at the map it seems there are workers and customers in europe, but usually these are separate? most europe workers working for US based entities, and most customers buying from different US entities

it has always been so that being able to read and write code is the first step on the rung. Once you get experience, you start to focus more on higher level concepts of what won't work, what will be annoying to maintain, what will be slow, what will be annoying to extend, and what will be acceptable to the other groups of humans you work with.

I think learning to code will still help with the other things here - its a good first step, even if its now obsolete on a direct level. Like how its useful for mathematicians to be able to calculate by hand even though calculators exist - it gives intuition for how to operate the tools and check results

how do you turn a prototype into a spec? do you find it ossifies incidental assumptions the ai made when making the prototype?

where are you working that bans llm usage in house? can I work there too?

buildings may be quite hard to destroy, but the relevant question is whether it is proportionally harder to build one.

required codes and standards may raise one (difficulty of removal) or both (difficulty of removal + difficulty of construction) in varying proportion.

this is clearly vibecoded though?

this morning I tried a different approach to speculative decoding by not writing any code at all and in fact going out for a walk around the lake near my house? What do you mean specifically?

several reasons. For some, because we think it might lead to useful results. For some, because we think the human pursuit of knowledge is worthwhile. For some, because we want to discover truths about logic and the universe. For some, because we are playing a status game (not just the mathematicians themselves but countries, universities, companies etc). For some, because we get personal fulfillment in understanding, and there being experts at something who can help others to understand.

For most, all of the above in various measure. To act as though one is the only thing thats important is wrong. The crisis is because the balance between these that existed is being shifted.

there was already far more to learn than any genuis could understand in a hundred lifetimes. There is no nobility in simply increasing the amount of things that can be learned

you get "assaulted" "all the time"!? surely this is hyperbole. If not, perhaps consider moving somewhere far away from where you live now, immediately!

Thats not how jev works? We're talking about a decision model that gives a number next to a series of options that, taken together, add up to 1. There is some sensible debate here about whether they should be taken as probabilities, but I can't follow what you're getting at if you did actually understand how jev works

50% heads, 50% tails. What would you have given for predicting my coin flip?

Have you tried asking jev to predict the result of a con flip? My one and only prompt to jev so far has been that. After I 90% heads as a result, I figured I'd wait a bit before checking again to see if this tech has become useful

the "paper windows" idea doesn't seem to have been done with as much soul as the rest of the ui. Just a turned up corner. I reckon something done there could really tie this together

damn, I would definitely have volunteered. But quite a crazy environment, seems to me mechanical switches with protective caps to prevent accidental activation on would be better suited to such things. Though it also depends on the possible consequences of the wrong input I suppose, if it were like a microwave or a clothes washer I guess its fine

I have simply not experienced 1) here in denmark. I don't know what to say, I live in a city that's not copenhagen, but if it exists it seems to me it must be rare.

as for 2, yes young people sometimes speak to each other in a mix of english and danish. It's using danish grammar, and critically, not english. You couldn't follow at all if you only spoke english (and with significant difficulty if you only spoke danish for that matter). I consider this similar to the process that creates slang

the other part of 2 regarding people that don't rush to learn the local language. Yes I know people who do this. After living here long enough, especially if they plan to stay, they tend to change their tune. I think there is a lot of motivated reasoning on this topic in particular, including stuff I have said to people myself in the past that friends like to remind me of - Whne I first came here I literally used to talk about how its so hard to practise language because locals won't help while being surrounded by locals who are happy to help. I won't speculate as to why people change after staying somewhere long, its just a tendency I have seen.

regarding 3, it is seemingly often associated with the far right yes, especially when bound with nationalism and themes of invasion/protection, but not unique to those contexts. I have experienced this in leftist political spaces personally, up to the point of votes being held and passing, restricting language in spoken meetings to danish only, explicitly for the comfort of those attending. And of course, I have also experienced this in entirely unpolitical spaces like hobby clubs, discussions at parties, and at work.