post_below's recent activity
-
Comment on Git at any scale in ~comp
-
Comment on Git at any scale in ~comp
post_below LinkCursor and daddy Musk want all that sweet coding data for themselvesCursor and daddy Musk want all that sweet coding data for themselves
-
Comment on Older adults who typically had longer intervals between meals accumulated chronic diseases more rapidly over time than those with shorter intervals, according to study from Karolinska Institutet in ~health
post_below Link ParentAgreed, they picked strange numbers. And also: The study found no effect in people under 78 years old, and there are just way too many confounding factors for the result for 78 and above to mean...Agreed, they picked strange numbers. And also: The study found no effect in people under 78 years old, and there are just way too many confounding factors for the result for 78 and above to mean anything other than "more study needed". And of course the reason there are too many confounders is that they used an existing dataset, it's meta analysis rather than a study designed to test the impact of fasting.
Which is ok, nothing wrong with meta analysis. But the bulk of available research shows that there are significant metabolic benefits from fasting, including on markers strongly correlated with disease.
-
Comment on Young adults in the US are increasingly wary of AI, concerned it will take jobs in ~tech
post_below Link ParentI agree that AI will automate some jobs. And that it could be a profound number if the tech keeps improving at a steady rate. I don't think that's a foregone conclusion though. It all depends on...- Exemplary
I agree that AI will automate some jobs. And that it could be a profound number if the tech keeps improving at a steady rate. I don't think that's a foregone conclusion though.
It all depends on whether they can figure out how to get an ever increasing amount of quality training data. Fine tuning will continue to improve, harnesses will get better, but every significant capability leap so far has come with a size jump (excepting the introduction of tool use and reasoning fine tuning) and you need pre training data for a size jump.
The reason the tech would need to keep improving, for some time, is that it can't fully automate the majority of jobs right now. In software engineering, sometimes cited as the field most in danger of replacement, agents can oneshot boilerplate and do an impressive job on more complex tasks. But you still need a human in the loop, and not just any human, a domain expert. Even the scarily smart models are dumb a non-negligible percentage of the time... And in increasingly hard to catch ways.
There really aren't many fields where current tech is threatening many jobs. In knowledge work, they still hallucinate pretty often, no way you can rely on ouput without human direction and review. In writing they suck quite a lot. In health and law it's grossly irresponsible to trust agent output without expert review. Even in digital art, which seems particularly threatened, you still need a human artist if you don't want to put out slop and piss people off.
So it's really about where the tech will be in X years, and right now there's no one who can make a credible prediction.
I have no doubt though that LLM tech will change the world in all sorts of ways. I think of it as another chapter in the digital revolution following the internet and personal computers.
But it's useful to offset the hype: Jobs might be replaced at some point, but so far it doesn't seem to be happening.
-
Comment on Why your Amazon order confirmation emails have become so unhelpful in ~tech
post_below Link ParentMaybe their response to the coverage will be that it's meant to protect customer privacyMaybe their response to the coverage will be that it's meant to protect customer privacy
-
Comment on Why your Amazon order confirmation emails have become so unhelpful in ~tech
post_below Link ParentGood point. But you're thinking like an engineer. If it's true that they nerfed confirmation emails globally to protect their metrics, that's a management or c-suite call. I doubt that was the...Good point. But you're thinking like an engineer. If it's true that they nerfed confirmation emails globally to protect their metrics, that's a management or c-suite call.
I doubt that was the main motivation though, more likely they want to encourage clickthroughs into their walled garden. Probably motivated in part by a larger initiative to create more opportunities to expose people to their AI tools
-
Comment on Virgin Galactic wants your help naming its new Delta class spaceship in ~space
post_below LinkStrange, none of the names mention EpsteinStrange, none of the names mention Epstein
-
Comment on I'm not ready to let this place go in ~tildes
post_below Link ParentI've never been to a forum inspired meetup, but I've seen plenty reportedly work out great over the years. I think maybe the only barrier is not everyone wants to tell internet strangers where...I've never been to a forum inspired meetup, but I've seen plenty reportedly work out great over the years.
I think maybe the only barrier is not everyone wants to tell internet strangers where they are. Maybe an anonymous location only form would be a good way to test the waters? Then you'd know how many interested people are in a given area.
-
Comment on I'm not ready to let this place go in ~tildes
post_below Link ParentI want to add one more similar reading of recent posts, this is what I heard: "I haven't been making or planning any major updates for a long time, that isn't going to change, but I see no reason...I want to add one more similar reading of recent posts, this is what I heard: "I haven't been making or planning any major updates for a long time, that isn't going to change, but I see no reason Tildes can't continue indefinitely."
I think the drama is a symptom of a lot of people caring about Tildes. And maybe some of it is drama for its own sake. But I think it's evidence that Tildes doesn't need more features... though I'm also getting that updates written by community members could potentially be merged so it's not completely off the table. In any case as others have said better and at length: Tildes is about the community, the platform is already doing it's job because the community exists. Doesn't have to be perfect, nothing humans do is ever perfect.
I'm pretty sure it's all going to be ok. We all still have invites to use if we ever feel like the population is dwindling.
-
Comment on I love Tildes in ~tildes
post_below LinkI'm not sure there's anything like Tildes on the modern internet that isn't focused on a niche. The thing I appreciate most is that every time there's been an influx, the culture of good faith has...I'm not sure there's anything like Tildes on the modern internet that isn't focused on a niche.
The thing I appreciate most is that every time there's been an influx, the culture of good faith has survived!
-
Comment on Is Flowers for Algernon sci-fi? in ~books
post_below LinkSci-fi or not, Flowers for Algernon is a beautiful book. There are lots of valid criticisms that can be made about it but as a whole it works in a way few stories do. The first time I read it, it...Sci-fi or not, Flowers for Algernon is a beautiful book.
There are lots of valid criticisms that can be made about it but as a whole it works in a way few stories do. The first time I read it, it felt like discovering some sort of lost draft. It's unpolished, maybe a little clumsy at times, and more of a short story than a novel. But it's a great premise, Charlie's arc draws you in, and the humanity of the story overwhelms the imperfections.
Even reading it a second time years later it still felt like something special. Despite the low hanging cynicism in the social commentary, and the tragedy, the story feels uncynical, sweet and somehow hopeful.
-
Comment on Any significant changes to Tildes are extremely unlikely in ~tildes.official
post_below Link ParentI've thought about this a lot too, possibly since before I discovered Tildes. But we've all seen what volunteer moderation looks like, over time. The politics that develop, and so on. Safe to say...I've thought about this a lot too, possibly since before I discovered Tildes. But we've all seen what volunteer moderation looks like, over time. The politics that develop, and so on. Safe to say that it would a huge accomplisment to create a community like Tildes with distributed moderation/control.
Part of the reason Tildes works is that the moderation is quick, decisive and mostly stays in the background. Once the politics of moderation and governance leak into the community it can drag everything down.
Still, it would be fun to try in some future with more time in the day.
-
Comment on Neuroscientists find logical reasoning does not involve language-processing parts of the brain in ~science
post_below LinkI love empirical evidence of intuitive concepts... The linked study was paywalled though so it's hard to say how good the evidence is. But of course it makes sense, we could no doubt reason and...I love empirical evidence of intuitive concepts... The linked study was paywalled though so it's hard to say how good the evidence is.
But of course it makes sense, we could no doubt reason and perform logical analysis before we developed language, most animals can reason. Some level of reasoning is a basic survival skill in complex organisms.
And now we've developed pseudo-intelligence that reasons exclusively using language (in a manner of speaking) as a shortcut around the much harder problem of developing true reasoning.
-
Comment on The real energy use of agentic AI in ~tech
post_below Link ParentThat's true, in the US things are (hopefully temporarily) completely fucked federally when it comes to not destroying the environment. It's down to states, counties and towns, some of which are...That's true, in the US things are (hopefully temporarily) completely fucked federally when it comes to not destroying the environment. It's down to states, counties and towns, some of which are doing encouraging things.
-
Comment on Browser fingerprinting demonstration in ~comp
post_below Link ParentOn the other hand, consider things like the iphone. A particular iphone version provides huge swaths of users with identical hardware and software which makes fingerprinting significantly more...On the other hand, consider things like the iphone. A particular iphone version provides huge swaths of users with identical hardware and software which makes fingerprinting significantly more challenging.
The reason browser fingerprinting is useful at all is that the primary use case (outside of fraud/bot prevention) is advertising targeting, which doesn't need to be anywhere near perfect to be profitable.
-
Comment on Browser fingerprinting demonstration in ~comp
post_below Link ParentPossibly the reason you don't vibe with it is that the whole project, including most of the prose, is vibe generated.Possibly the reason you don't vibe with it is that the whole project, including most of the prose, is vibe generated.
-
Comment on The real energy use of agentic AI in ~tech
post_below LinkI love that the author took the time to do these estimates since there's so little non-vibey data on this topic. As always when it comes to climate change: Individual impact will not solve it...I love that the author took the time to do these estimates since there's so little non-vibey data on this topic.
As always when it comes to climate change: Individual impact will not solve it whether it's driving an electric car or avoiding LLMs.
I'm not saying that people shouldn't reduce their footprint, I reduce mine in all sorts of ways. I think it sends an important message to society at large about what our values should be.
But policy, regulation and infrastructure are the actual solutions, which we can see all over the world in countries that are successfully reducing emissions.
When it comes to AI labs and datacenters: Make them fully offset their impact. Make it expensive not to use renewable power, enough that it makes economic sense for them to fund new green energy development in some cases. Same for water use, it would only take a nudge to make recirculating cooling the right financial choice.
-
Comment on The magic is in the language in ~tech
post_below Link ParentIt's an interesting topic. Assuming I'm reading this right, I want to clarify a couple points. One of the key features of LLM agents is that they can be useful even outside of the direct context...It's an interesting topic.
LLMs would be useful, if they have been trained on the context they would have been used in. That means it works on the generic queries, but on the domain specifics LLMs either will not have enough training data, or will contort the "same words" from its training corpus, forcing a language from an aggregate of contexts
Assuming I'm reading this right, I want to clarify a couple points. One of the key features of LLM agents is that they can be useful even outside of the direct context they were trained in/on. They're more useful, for some things, when you're closer to the center of the training distribution but they can do something that looks like generalizing, some of the time, outside of the distribution
The reason for this appears to be that there's is more information encoded in the token relationships than you might imagine. One way to look at it is that words have both independent meaning and meaning that changes based on the context. Which means both that a word has self contained meaning that exists independently of any training data, and that dynamic meaning can be built on the fly out of word relationships as you write them. Because this is encoded in the weights, when you're running inference, meaning can survive into space that isn't specifically in the training. Even when there is no reference (so to speak) the model can blaze a path through the dark built out of token relationships and the meaning they encode. The words and the reasoning are the same thing, so they can go uncharted places.
It starts to fall apart at some point, but with a large enough model it can make it pretty far into the dark before that happens. That's especially true when it can generate feedback and incorporate the results into its inference, which is one of the defining characteristics of agents. The answer doesn't have to be in training for a model to find it as long as there's a way to test and iterate. In domains like code, math, biology and physics, deterministic testing is often easily accessible.
-
Comment on Share your favourite dal recipe! in ~food
post_below LinkLooking at the recipe that @phoenixrises posted, it looks great. You can do it a little quicker though, and still end up with great dal. Here are a few things I've learned... Butter/ghee and cream...Looking at the recipe that @phoenixrises posted, it looks great.
You can do it a little quicker though, and still end up with great dal. Here are a few things I've learned...
Butter/ghee and cream are key. Mostly that's just about fat and viscosity, so if you want it vegan instead you can use coconut milk (full fat with cream) and a neutral oil like refined coconut or avocado. Some recipes call for adding the coconut milk towards the end, similar to how you might do it with cream. I like to add it earlier, it helps mellow out the coconut taste.
You don't actually have to toast whole spices, or make your own garam masala. It does make a difference but it's pretty subtle. A good pre-made garam masala blend and a few other jarred spices gets you at least 90% there. Add the spices at the end of the veggie saute step to toast them a little.
Speaking of the veggie saute step, since the linked recipe doesn't include veggies... Consider sauteeing, at the least, finely chopped or grated onions in the recipe's called for oil before adding the other ingredients.
You can pre-cook the lentils in an instant pot (pre-soak totally optional) to save time.
The truly important ingredients are black lentils, some form of tomato (paste, diced, strained, etc..), cream/coconut milk, butter/oil, garlic, ginger and garam masala. You can just kind of experiment with everything else until you dial it in. I like to be conversative with the spices for dal, too much can distract from the staple soul food goodness of a fundamentally simple dish.
Oh and mash the lentils a bit towards the end with a potato masher or immersion blender, it improves the emulsification, texture and mouthfeel. You want to leave the majority of the lentils whole though.
-
Comment on What JavaScript framework should I use? in ~comp
post_below Link ParentThe way I see it, the main selling points of frameworks are standardization and compatibility. They make it easier to hire a [framework] developer to replace the previous one and they come with...The way I see it, the main selling points of frameworks are standardization and compatibility. They make it easier to hire a [framework] developer to replace the previous one and they come with varying levels of built in patterns that help stop interchangeable developers from making messes. Once upon a time they also helped solve cross browser compatibility, but these days, for most use cases, that isn't near as much of an issue as it once was. The last selling point is also outdated: adding functionality that's hard to implement in, for example, vanilla css/js. It's been a long time since I've encountered something that's significantly easier to do with a framework than it is with modern css/js. YMMV of course.
It's true that you'll end up writing something like a framework if you build a vanilla UI, but whether or not that takes more time depends on various factors. It definitely doesn't have to. And unless you really mess things up, your bespoke UI is going to be lighter weight than the framework alternative.
I'm not sure I see the maintainability angle. Are you thinking because many people aren't familiar enough with vanilla and so have a hard time understanding or writing it in a logical way? It certainly doesn't demand any bad patterns by nature.
Something I think it's easy to forget is that html/css/js essentially is a framework. One that's been iterated for the use case (websites/apps) for 30 years.
Not that frameworks aren't useful in lots of cases but, as nothing more than language constructs, they don't add as much as I think people sometimes assume.
Also: coding agents. The time to do X in a framework has now become basically identical to the time to do X without the framework. The remaining difference is load time on a slow connection and resource usage on a cheap device. In that context frameworks are just adding weight. Provided of course that the coding agent is in the hands of someone who is fluent in the underlying languages. Otherwise it could potentially be better to use a framework you're more familiar with, or with more good idioms than bad in the training data.
Anyway, I appreciate the digression, but my curiousity was more specific to @delphi's situation. It seems like an ideal opportunity to hand roll the UI instead of spending that time learning a new framework. Especially when a frontier model is almost guaranteed to give you a solid answer to every version of "how do I do this thing in vanilla js" you're likely to encounter.
Or possibly GitHub was the next best option. Individual orgs don't need a 3rd party to host their git/repos, as you pointed out with Gitea, there are all sorts of good options that are easy to self host. Not to mention 3rd party hosters that aren't owned by Elon Musk. The selling point of GitHub is as a platform for getting work out there and attracting contributors. And now with the torrent of AI generated projects flooding GitHub I wonder how long it will be the best place to put your work out there.
It's already become too unreliable for professionals with the AI flood causing frequent outages.
From memory, GitHub commit volume increased to around 1 billion a year by the end of 2025, which caused them to celebrate and commit to a huge capacity expansion plan. But then by April monthly commits were over 1 billion and have doubled to almost 3 billion as of now. With their clunky infrastructure it's hard to see how they'll keep up. They're getting DDoSed by AI commits.