throwaway4aday

Cities need to rewrite their laws on rooming houses. It's been many decades since they were put in place and their current form is a hindrance rather than a help. There aren't many in between options for people falling on hard times. Roommates, cheap apartment, motel room or you stay at some form of shelter. Proper rooming houses could fill that gap by being less expensive but still affording you your own space and perhaps be easier for some people such as the elderly or disabled if meals are provided.

Oddly enough, the only way you can really make housing a commodity item would be to build capsules or other types of mobile homes that can be purchased separately from the land they are parked on. The Nakagin Capsule Tower comes to mind[0] but it's far from a solved problem since you'd need similar infrastructure to a container port to move the capsules around but since you don't move them often the infra itself has to be movable so it can go where the demand is, which is a pretty big ask if you're only moving one capsule. The only existing format is the traditional mobile home which is limited to single story and low density which means it can't be located near urban centers that need it the most. So, you're kind of stuck with the problem of it being easy to build small homes but really hard to find a place to put them.

[0] https://en.wikipedia.org/wiki/Nakagin_Capsule_Tower

I wonder if you could dedicate some of the probes to serving as a relay network stretching all the way back by only partially accelerating them so the bulk of the probes will go as fast as possible and then a tiny fraction will go a little slower and another tiny fraction slightly slower than those and so on so that they spread out into a chain as they go. Of course, you'd only need to do it that way if you had to launch them all within a short window. If you were to continuously launch probes then you could maintain an evenly spaced chain for as long as you could keep launching them.

Do the One Punch Man routine minus the run.

yikes! a cheapo generator on Amazon is like $10 and you can probably get them for less on AliExpress. Not sure if you'd have to amplify the signal for either LEDs or headphones but I imagine you'd want some kind of volume control for the headphones. Total cost could be $30 or less and would make a fun weekend project even for electronics noobs like me.

Could you just buy a cheap signal generator and hook it up to some LEDs and headphones to make a hardware version of this?

Really burying the important takeaway at the bottom:

Starting in 2025, the Vera C. Rubin Observatory in Chile — funded by the National Science Foundation — will catalog the solar system from the ground...

"It took us 200 years to discover all the asteroids we know to date, about 1.2 million asteroids," Mario Jurić, the Rubin Observatory's solar system discovery team lead and the director of the University of Washington's DiRAC Institute, told Astronomy. "In the first three to six months of Rubin, we will double that."

That's one additional ground based observatory coming online in the Southern hemisphere. If we get our act together and build a lot more of these and other space based observatories we're going to see the true scope of the "shooting gallery". We really need to raise public awareness of this, as a planet based civilization we are taking our chances by ignoring the problem and assuming everything will be fine for a hundred or a thousand more years. All of our efforts at averting a climate catastrophe will be for naught if we get smoked by a rock big enough to fill the atmosphere with particulate matter and drop global temperatures for years afterward let alone anything bigger resulting in global firestorms and tsunamis.

Back in the day people made tea out of anything they could get their hands on that wasn't outright poisonous. https://en.wikipedia.org/wiki/Herbal_tea#Varieties

Also, if you're in constant pain you'll try all kinds of random stuff to make the pain go away. If necessity is the mother of invention then desperation is its father.

A bit hyperbolic, I'm sure the different variations are finite but you can produce a very large number of examples with essentially the same information content.

At a certain point you just need to take a statistically representative sample of the output for review.

No, it's an effective technique that has been used by Microsoft Research among many others.

There could be an infinite amount of junk, in fact the amount of junk available online before LLM output was added to the mix was already incredibly huge. At this point it no longer matters because you can take 1 or 2 examples of some knowledge you want to train a new model on and you can then feed it to an LLM with instructions saying to create 10,000 variations on this text while still retaining the factual details, you then filter this list and throw away any output with errors or other garbage and what you're left with are a good number of high quality examples of the information you want to add to your training set. Incredibly, there is new research that shows even this might not be needed at some point because there is evidence that transformers can learn from a single example.

Agree, Copilot is a huge speed boost for stubbing things out, auto-generating the dead simple parts, and refactoring. ChatGPT is awesome for rubberducking, pair programming, brainstorming, and figuring out what that thing that you kind of remember is called. We aren't at the point where we can just say "robot, do my work" but it's for sure good enough to take care of the boring stuff and be a sounding board.

This is wrong for at least two reasons:

1) The data scraped from the web is filtered, deduped, ranked, and cleaned in a variety of other ways before being used for training. While quantity is necessary for a model to learn the structure of language, quality is even more important once you have a model that can produce coherent output so a lot of work goes into grooming the data.

2) There have been a bunch of papers and training runs that show synthetic data created specifically for training a model is as good or better than scraped human produced data. The importance of scraped web content is quickly declining because you can now generate infinite higher quality examples using existing trained models. The only relevance it has now is for knowledge about new developments and that is a much easier stream to filter since most of the important stuff comes from official sources and you don't need as many variations since you can just generate your own using one or more examples.

Going to drop this here for others who haven't heard of it https://invidious.io/

Now, how do we fix this? YouTube's ad model sucks. Their algorithm sucks. Their front page sucks. They've captured a bunch of creators though so often YouTube is the only place you can find someone.

I want those creators to benefit from me viewing their videos. I want the fact that I view a video and like it to help other people find that video in their recommendations. I want an algorithm that shows me things that are interesting and relevant not one that promotes the spammiest and most ad heavy videos that barely have anything to do with my watch history.

Having an alternative front end is nice but I don't want to rob YouTube of the money they spend on hosting the videos.

So, how do we do this?

Peer to peer fails when there is little interest in something or when most people leech and it sucks for archiving old content.

Hosting it all in one place is super expensive and hard for a small group to manage without turning into YouTube.

Maybe we could find a way for the creators to host their own content and get paid when people view it while being part of a large federated network for easy discoverability?

Please list any projects you know of, I'm sure there are a lot of people here who would be willing to contribute or donate.

We can fix anything with another layer of indirection!

Interstellar travel is only going to work if we're using much more powerful propulsion techniques than we currently are. I don't think gravity assists will help much in that case. I still think it's a boon to have a lot of interstellar bodies though, they are certainly easier to reach than the nearest stars and they could be just as interesting. Who knows what kind of weird, icy worlds might exist out there heated only by radioactive isotopes and tidal forces?

Well, that entirely depends on if the theory and models being used are correct. This article seems to indicate that many of those beliefs did not match reality. I think we should keep looking rather than discount out of hand. There are more things in heaven and earth, after all.

I'm guessing this is due to poor translation combined with lack on knowledge on the reporter's side resulting in the error not being caught.

If there are that many Jupiter+ sized objects then how many rocky planet sized ones are there? Would they even be detectable via infrared or gravitational lensing?

I don't think you can get away with throwing out the baby with the bathwater on this one. In any market where there is more than one supplier you need some form of advertising even if it's just just a couple signs that say "Bob sells shoes" and "Alice sells shoes" though in reality customers need more information than that. What kind of shoes? Men's shoes? Women's shoes? Children's shoes? What color are they? What weather are they good in? What are they made of? etc. etc.

Granted, advertising has been taken to the extreme and there wouldn't be harm in cutting back on it to a certain degree, you just can't get rid of all of it so it is in fact one of those jobs that need doing.

The point of my comment is that you can say the same thing about every pair of thing someone wants to do and thing someone needs done and the essential part is that it doesn't matter what the second thing is, people would still complain that all this talent for A was being wasted because they were working on B instead but it's not a waste for a variety of reasons one of which is that there is not enough demand for A. You could easily say someone is a psycopath or antisocial if they choose to spend all their time doing something that isn't needed or not wanted instead of doing the things that other people need done. You would at the very least say they are selfish.

In a way, mural art in churches and cathedrals is advertising. It's advertising for that religion, showing you the promised benefits of subscribing or the terrible things that will happen if you don't sign up now during this limited offer.

You could really say this about anything, the only moral to the story is that people have to work to pay the bills and the work that pays is the unglamorous stuff that needs doing.

I believe this has been true throughout human history. The greatest artists and thinkers throughout history have earned their keep by providing services to the wealthy or teaching paying students (also often wealthy).

There are upsides, the artists get to eat but they also have all the benefits of being connected to their employer/patron including time spent practicing their craft. If they were to quit and focus on pure art then they would likely earn far, far less and possibly spend less time making any kind of art commercial or not.

It reminds me of this survey that made the front page not long ago about author's that contained this illuminating paragraph:

> While 80% of respondents considered themselves to be professional authors, only 35% said they were full-time authors while 53% said they were part-time authors (with the balance being one-book authors or undecided). The primary writing occupation of part-time authors outside of publishing books was professor/academic (8.5%), followed by book illustrator/author (4.2%), editor (2.9%), poet (2.4%), journalist (2%), teacher (2%), and entrepreneur (1.5%).

So even people who consider themselves professional authors are unable to work at it full time. It makes sense, there's only so much one person can produce and it's quite difficult to find an audience willing to pay for what you make. This likely holds true across all professions i.e. you have to do the work that they want done and not the work that you want to do.

https://www.publishersweekly.com/pw/by-topic/industry-news/p...

A budding Kardashev IV civ https://kardashev.fandom.com/wiki/Type_IV

It seems like it would be much, much slower to CNC all the plywood and insulation and assemble the boxes and then put the boxes together than it would be to just frame a wall, sheet it and put insulation batts in. Having access to and operating a CNC also seems harder than learning to use a skill saw and hammer a nail. It feels like they're trying to apply abstraction to construction where you just have one unit that can be adapted to work for everything but they don't realize that already exists and it's called dimensional lumber.

The best form of open source houses are free and complete framing plans with material and cut lists and detailed instructions that don't assume the builder is a professional. Add some guides on exterior and interior finishing, basic plumbing and electrical and walk-throughs on how to obtain permitting and being your own general contractor and you'd have something genuinely useful for everyone.

That should be done without appropriating the voice and image of the deceased without permission.

Probably the most exciting thing about this is that it shows the hard work of labs to discover the tacit knowledge necessary to produce single layer graphene is bearing fruit. Developing the real world industrial processes for putting theory into practice is always the hardest part of creating new technology especially when doing so on atomic scales. It seems slow and underwhelming compared to whizbang announcements about wonder materials but this is the type of gradual progress that the future is based on.

I didn't say it should provide citations for every statement. We're talking about instances where the model reproduces text verbatim. If it is paraphrasing or creating new text then obviously it doesn't need to cite any more than a human would.

That is the worst possible way to implement a liberalization of drug laws. I'm in favor of decriminalizing practically all drugs but this is just insane. They need to be treated like the currently legal drugs in some cases and more tightly regulated in others where the substance is more dangerous to the user. There needs to be a spectrum from dispensaries where the cashier is some college kid and pharmacies where your access and education is handled by a pharmacist or doctor. Physical addiction, while complicated by several factors, is still a real phenomenon that must be identified and treated in order to avoid ruining lives.