Hmmm I think you're sort of right but not entirely. It's true that a novel consists of a valid organization of tokens, and that this sequence can be feasibly made to be output from a model. But when you say this:
> So there should be a prompt that can cause that sequence to be output
Is where I think I might disagree. For example, the odds of predicting verbatim the next sentence in, say, Harry Potter should be astronomically low for a large majority of it. If it wasn't, it'd be a pretty boring book. The fact that it can do this with relative ease means it has been trained on the material.
The issue at hand is about copyright and Intellectual Property - if the goal of copyright is to protect the IP of the author, then LLMs can sort of act like an IP money laundering scheme - where the black box has consumed and can emit this IP. The whole concept of IP is a little philosophical and muddy, with lots of grey area for fair use, parody, inspiration, and adaptation. But this gets very odd when we consider it in light of these models which can adapt and use IP at a massive massive scale.
> Unless your strategy is to create a photo-lab-like screen in pure black and red, or wear deep-red-tinted glasses, it’s unlikely that a pure colorshift strategy will cut out that big of a chunk of the spectrum.
I absolutely think this is the right approach. The glasses which do 'blue light filtering' which barely change your perception are clearly placebo, but a very strong redshift I think is obviously a different creature.
Absolutely, although dark orange seems to work well enough. If you can put them on and still tell the difference between most colors, they aren't working. I use my pair for one purpose: reading in bed with a backlit e-reader. I can't imagine trying to do much else with them on, they have plastic wings to block light from the side and they're not light.
Fantastic idea! I didn't see my https://words.zip game here so I'll submit it, even though it didn't get mucht traction on HN - it's up to 288,000 words found now!
Just a thought, maybe a "tunnel vision" mask toggle? So I can focus on the center of the screen without all the peripheral noise. It doesn't have to be a significant loss, but dark edges and corners with a nice fade to the visible circle in the center. Whatever that equates to..
Oh that's not a bad idea, and easy enough to implement! I'll definitely get to it later today - I think because I play it on my phone the contrast is less obtrusive, but on a big desktop screen it's a bit boggling to look at
This is amazing. Many years ago I started a similar project that but with a more traditional word search style. This one is much better as I couldn't get the interface the way I wanted. Congrats!
Yup! A fairly simple deterministic RNG such that it can be generated both client and server side. This strategy greatly reduces the burden of transmitting all the letters as you scroll around, only needing the subset of found words
Take a look around the map! Some people are making drawings, some are just trying to fill a dense area as much as possible, some restrict themselves to only 4 or 5 letter words. Just interesting little projects people are working on
A lot of people have called out some interesting things - one thing that I notice is how the cold water ports shut down in the winter (in the northern hemisphere). It's one of those things I've always heard and known about, but to see it visually conceptualized (and the implications on economy and national interests) is very cool
One of the notable seasonal events here in Minnesota is the last barge of the season before the Army Corps of Engineers closes the locks on the Mississippi River. Then again, each spring, the first barge though is celebrated when the locks open again.
It's not earth-shattering, but it generally makes the news ;-)
Something that will change in the next century. I'm curious how that will affect shipping. I'm imagining it won't have much impact on container shipping but natural resources probably?
> Something that will change in the next century. I'm curious how that will affect shipping. I'm imagining it won't have much impact on container shipping
No, on the contrary! The effect on container shipping will be absolutely massive with the Northeast Passage opening up. It's going to cut thousands of miles off shipping routes, since its the much shorter route from American East Coast ports and the Euros to Asia.
You don't have to do anything - if it's more important to you to bring your authentic beliefs and opinions in front of others - you're free to do so. But if your number one priority is getting ahead at work - or simply aiming to develop social cohesion with a group around you, conforming to that group's beliefs and culture are helpful. This applies not just to the 40 hours you work, but to any group you want to be welcome in, whether it's your school, country, church or sports team. And after a little while, it's not really pretending anymore.
> Maybe we should tell kids that in school
Half of the reason we have school is to socialize kids to these very concepts. We obviously lecture about peer pressure, but most of this is learned innately. Society couldn't exist without these social pressures, even though they're not always beneficial.
> Half of the reason we have school is to socialize kids to these very concepts.
School is a brainwashing machine that limits the range of acceptable dialog and therefore leads to a culture of mediocrity.
Don't get me wrong, you should always be polite and empathetic towards others. But the most world-changing ideas don't come to you when you're dedicating huge amounts of your processing power to self-censoring or arbitrarily creating constraints because you're worried that the group will cancel you over them.
There is an actual range of acceptable dialog though. The workplace is not the right forum for discussing politics and religion, or your sex life. Not discussing that stuff is necessary not only for politeness but good rapport and productivity.
>But the most world-changing ideas don't come to you when you're dedicating huge amounts of your processing power to self-censoring or arbitrarily creating constraints because you're worried that the group will cancel you over them.
The most world-changing ideas are not hot button ragebait talking points or political soapboxing. Furthermore, it should not require a lot of processing power for you to stay on task at work and not inject controversial narratives into everything. You might be lucky enough to have some friends at work who enjoy talking to you about that stuff but have a little respect for everyone else who isn't in the mood for your shit.
I concur - the cost of this conformity is that creative outliers can be crushed, and their ideas (maybe important ideas that could save the organization from some dire outcome) can be lost.
If 100% of the fuel we use comes from this technique - where it is imbued with carbon that was atmosphere-bound (flue gas) then we have decreased emissions significantly, since the alternative was that we release the flue gas CO2 AND the burning fuel CO2.
> So there should be a prompt that can cause that sequence to be output
Is where I think I might disagree. For example, the odds of predicting verbatim the next sentence in, say, Harry Potter should be astronomically low for a large majority of it. If it wasn't, it'd be a pretty boring book. The fact that it can do this with relative ease means it has been trained on the material.
The issue at hand is about copyright and Intellectual Property - if the goal of copyright is to protect the IP of the author, then LLMs can sort of act like an IP money laundering scheme - where the black box has consumed and can emit this IP. The whole concept of IP is a little philosophical and muddy, with lots of grey area for fair use, parody, inspiration, and adaptation. But this gets very odd when we consider it in light of these models which can adapt and use IP at a massive massive scale.
reply