This is great. I live in a dense suburb and the weekend is just a cacophony of deafening noise from leafblowers, table saws, pressure washers, lawn mowers, etc. My neighbor had a team of people grinding and cutting stone slabs for their patio from 9-5 last weekend. Wish people would chill out.
I haven't put any leaves out on my curb the last two years, but I still use my electric leaf blower. Early in the season, I use my mower to mulch some of the first leaves to fall into the lawn(which consists of grass, clover, violets, dandelions, etc). But eventually there are just too many leaves to mulch, and they would smother everything, so I blow them into big piles. Then we take the dogs out to play in the piles. Then we scoop them up with tarps and spread them all over my veggie garden beds. In fact I went and "stole" some leaves from the piles my neighbors put on the curb to get even more coverage of my gardens. Now in spring, the leaves have smothered a lot of weeds, and the ones that do grow are easier to pull because they have to push through the leaves and the dirt stays moist longer. I see a lot of insect life living in the leaves as I go about weeding. I hope the ecosystem is strong to balance out pests on my veggies.
The United States carpet bombed North Korea, destroying 90% of all structures in the country, killing 300,000 in the immediate bombing run, then sanctioned them into the stone age. The relative poverty of North Korea is not a consequence of Chinese support.
It's a consequence of the political system in North Korea and Chinese support. The industry of the north grew faster and was larger than the south after the war. Eventually the economy of South Korea overtook and then greatly surpassed the north. China support = not wanting the Americans to win but wanting to keep North Korea a poor subservient state. Is America as bad? Sometimes, but I think plurality of leadership thought in the West always tempers the worst instincts of their nation states.
The US did worse to Japan. Also South Korea was similarly carpet bombed in the process of re-taking it from the north as almost all territory was lost and the north almost won.
Are you serious?!! All of these countries I listed had democratic governments couped out or attempted to be forced out by USA. If you are ignorant about history you can just Google it.
Or maybe staying ignorant is part of your grand strategy to avoid cognitive dissonance.
How is USA imposing a dictatorship on Chile nonsense?
> I read section about US involvement, and nothing there says they organized coup in 1973, in opposite, this link says that many sources confirm US were not involved into the coup
Out of curiosity, skimmed the article a bit and look:
> They [Kissinger and Nixon] do take credit for creating the conditions that led to the coup. Kissinger says that "they created the conditions as great as possible." Nixon and Kissinger also discussed how they would play this event with the media and lamented the fact that, if this were the era of Eisenhower, then they would be seen as heroes.
Are you playing dumb or what?
> My critics is directed to your propaganda-troll alike communication style: thrown 10 names of countries without any specifics.
Seeing a spec in somebody's eye, but not noticing a log that's in your own, eh? Don't forget your tinfoil hat.
Sure, I guess sanctions were there just for the kicks without any real substantial effect. Just so that poor non-commies don't feel bad about themselves, since, you know, evil commies had already supposedly restricted all the trade. /s
This has nothing do with "rules of engagement" and everything to do with the fact that the US had dominated global economy as a result of WWII and since then had been doing everything in its power to eliminate any possible opposition, competing, or really anyone who doesn't bow to Pax Americana.
Then maybe US obviously shouldn't aggressively burn down every other country, if it can't take responsibility?
And let's be honest, the US can't build an autonomous functional democracy in another culture, period. It's not the matter of resources, it's just theoretically impossible.
Just look at Afghanistan. It's "democracy" fell like a house of cards just as soon as the US had withdrawn it's troops.
Yeah, bad Afghan people didn't want to fight for a pro-American regime that supported brutal warlords and child prostitution (literally the first populist thing Taliban did was to crack down on bacha bazi - imagine being worse than Taliban). What a propaganda troll rhetorics.
Sorry to break it to you, but what had happened was exactly Afghan idea of freedom. A freedom from a puppet regime installed by invaders. It's not a healthy one, but there is nothing healthy to be expected after 20 years of being fucked by invaders "for their own good" under fancy fake slogans of "freedom".
But hey, war is peace, freedom is being a neocolony with a puppet government. Nothing new here.
> it would be better indeed to put these resources into Ukrainians
And those resources would be a shit load of NATO weaponry left to Taliban, eh?
US is not perfect and you can find many more cases of misbehavior or conspiracies, but they are many times better in respecting human rights and development than regimes of Mao/Stalin/Putin/Xi.
You are using whataboutism, personal attacks and ignorance as a debating tactic - including quoting the CIAs words to exonerate the CIA. Its as if USA did not overthrow democracy to install the Shah of Iran. It's as if US government did not democratically elected Allende in Haiti. It's as if CIA backed banana republics don't exist in Latin America. It's as if USA did not send ships to the Indian ocean to assist west Pakistan with the genocide in Bangladesh to overturn the election results.
You are calling historical facts "fantasies" and making personal attacks on everything else. It's your commitment to ignorance even when all the information is shoved in your face that is impressive, much more than the whataboutism.
I in good faith checked your example with Chile, found that you didn't read link you posted, you compromised yourself and don't deserve any level of trust anymore nor any effort to check your other pointers. Again, what are you trying to achieve from this discussion?
> As I said I won't read your links anymore. Anything else? :-)
The commitment to ignorance and refusal to read, with celebratory, fait accompli, victory smiley is not surprising :) The stereotype of the proudly ignorant American isn't without reason.
You were obviously arguing in bad faith from the beginning. The links are for readers who will follow and are able to read, that would mostly be non Americans :)
You absolutely didn't read the Chile article in good faith. Instead you searched for a statement absolving CIA by the CIA. You live in a fantasy land where banana republics is just a clothing store. Not once have you acted in good faith. You are pretending to a level of ignorance that is completely unreal.
Germany and Japan had competent post-conquer leaders and orderly civilizations that was/is useful. Iraq and Libya had incompetent leaders and useless people.
Don't give free world leadership examples in Chile, Haiti, Guatemala, El Salvador, Iran, Palestine, Egypt, Bangladesh etc. The nature of the US military and economic machine as a force of ultimate good beating up truant, naughty countries must not be questioned. USA does not practice censorship, instead we have manufacturing of consent here.
How did iraq fuck around the USA in 2003? What the fuck, are the 1 million deaths caused by the iraq war also just people who fucked around and found out? I usually am pretty good at sensing sarcasm but this time I'm confused
I don't think you have to reproduce an entire original work to demonstrate copyright violation. Think about sampling in hip hop for example. A 2 second sample, distorted, re-pitched, etc. can be grounds for a copyright violation.
The difference here is that the images aren't stored, but rather an extremely abstract description of the image was used to very slightly adjust a network of millions of nodes in a tiny direction. No semblance of the original image even remotely exists in the model.
This is very much a 'color of your bits' topic, but I'm not sure why the internal representation matters. It's pretty trivial to recreate famous works like the Mona Lisa or Starry Night or Monet's Water Lily Pond. Obviously some representation of the originals exist inside the model+prompt. Why wouldn't that apply to other images in the training sets?
>It's pretty trivial to recreate famous works like the Mona Lisa or Starry Night or Monet's Water Lily Pond.
A recreation of a piece of art does not mean a copy, I've personally seen hundreds of recreations of Edvard Munch's 'The Scream', all of them perfectly legal.
Even in a massively overtrained model, it is practically impossible to create a 1:1 copy of a piece of art the model was trained upon.
And of course that would be a pointless exercise to begin with, why would anyone want to generate 1:1 copies (or anything near that) of existing images ?
The whole 'magic' of Stable Diffusion is that you can create new works of art in the combined styles of art, photography etc that it has been trained on.
A work doesn't have to be identical to be considered a derivative work, which is why we also don't consider every JPEG a newly copyrighted image distinct from the source material.
As an example of a plausible scenario where copyright might actually be violated, consider this: an NGO wants images on their website. They type in something like 'afghan girl' or 'struggling child' and unknowingly use the recreations of the famous photographs they get.
It’s not quite a one to one. Copyright law isn’t as arbitrary as it would seem in my experience. Also there’s the conflation of two things here: whether the model is within copyright violation and whether the works generated by it are
The “color of your bits” only applies to the process of creating a work. Stable Diffusion’s training of the algorithm could be seen as violating copyright but that doesn’t spread to the works generated by it.
In the same vein, one can claim copyright on an image generated by stable diffusion even if the creation of the algorithm is safe from copyright violation.
“some representation of the originals exist inside the model+prompt” is also not sufficient for the model to be in violation of copyright of any one art piece. Some latent representation of the concept of an art piece or style isn’t enough.
It’s also important to note the distinction that there is no training data stored in its original form as part of the model during training, it’s simply used to tweak a function with the purpose of translating text to images. Some could say that’s like using the color from a picture of a car on the internet. Some might say it’s worse but it’s all subjective unless the opposition can draw new ties of the actual technical process to things already precedent.
Because you're silently invoking additional data (the prompt + noise seed), which is not present in the training weights. You have the prompt + noise seed for any given output.
An MPEG codec doesn't contain every movie in the world just because it could represent them if given the right file.
The white light coming off a blank canvas also doesn't contain a copy of the Mona Lisa which will be revealed once someone obscures some of the light.
OK so let me encrypt a movie and distribute that. Then you tell people they need to invoke additional data to watch the movie. Also give some hints (try the movie title lol).
If you distribute a random byte stream, and someone uses that as a one time pad to encrypt a movie, then are you distributing the movie?
The answer is of course not, and the same principle applies if someone uses Stable Diffusion to find a latent space encoding for a copyright image (the 231 byte number - had to go double check what the grid size actually is).
I think it boils down to one question: can you prompt the model to show mostly unchanged pictures from artists? Then it's definitely problematic. If not, then I don't have enough knowledge of the topic to give a strong opinion. (my previous answer was just an use case that fits your argument)
I mean no, it doesn't. It's like drawing something in Photoshop which is a copyright'd work: the act of creating it is the violation, it doesn't prove that Photoshop contains the content directly.
The way SD model weights work, if you managed to prompt engineer a recreation of one specific work, it would only have been generated as a product of all the information in the entire training set + noise seed + the prompt. And the prompt wouldn't look anything like a reasonable description of any specific work.
Which is to say, it means nothing because you can equally generate a likeness of works which are known not to be included in the training set (easy, you ask for a latent encoding of the image and it gives you one): equivalent to a JPEG codec.
> And the prompt wouldn't look anything like a reasonable description of any specific work.
I think this is the most relevant line of your argument. Because if you could just ask it like "show me the latest picture of [artist]" then you'll have a hard time convincing me that this is fundamentally different from a database with a fancy query language and lots of copyrighted work in it.
That's not true. As an example of a more recent copyright-protected work that Stability AI consistently reproduces fairly faithfully, I invite you to try out the prompt "bloodborne box art".
Longer term, by analogy, it will then of course turn into a "what color is your neural net" topic.
Which runs into some very interesting historical precedents.
((I wonder if there's a split between people who think AI emancipation might happen this century versus people who think that such a thing is silly to contemplate))
> No semblance of the original image even remotely exists in the model
What does this mean? It doesn't mean you can't recreate the original, because that's been done. It doesn't mean that literally the bits for the image aren't present in the encoded data, because that's true for any compression algorithm.
Do you have any examples of recreating an image with these models? Something other than Mona lisa or other famous artworks because they have caused over fitting.
there are some artists with very strong, recognizable styles. if you provide one of these artists' name in your prompt and get a result back that employs their strong, recognizable style, i think that demonstrates that the network has a latent representation of the artists work stored inside of it.
So, what you are saying is that it is illegal to paint in the style of another artist? I‘m no lawyer, but I‘m pretty sure that is completely legss as long as you don’t claim your paintings ARE from the other artist.
That seems to indicate to me that the original work is actually not under copyright, since if it is the only method of achieving such an image in such a style, then there is no originality to be copyrighted.
Perhaps different media has different rules? You can’t necessarily apply music sampling rules to text, for example. Eg I don’t think incorporating a phrase from someone else’s poem into my poem would be grounds for a copyright violation.
"Copyright currently protects poetry just like it protects any other kind of writing or work of authorship. Poetry, therefore, is subject to the same minimal standards for originality that are used for other written works, and the same tests determine whether copyright infringement has occurred." [1]
it's funny, VR doesn't give me motion sickness but browser parallax stuff does. i don't think i've ever seen a good implementation of this and i've seen tons of attempts. highjacking scroll just never feels good.
i wonder if there's a constant average reading speed (in terms of bits per second) across written languages. chinese is obviously more dense per symbol than english, but is each character processed more slowly as a result, preserving constant data rate across languages? also interesting to consider how much harder real time, verbal translation would be if data rates differed.
Scientists started with written texts from 17 languages, including English, Italian, Japanese, and Vietnamese. They calculated the information density of each language in bits—the same unit that describes how quickly your cellphone, laptop, or computer modem transmits information. They found that Japanese, which has only 643 syllables, had an information density of about 5 bits per syllable, whereas English, with its 6949 syllables, had a density of just over 7 bits per syllable. Vietnamese, with its complex system of six tones (each of which can further differentiate a syllable), topped the charts at 8 bits per syllable.
how can you encode 643 syllables using 5 bits? same for 6949 syllabes/7 bits?
If I understand this correctly, it isn't that they are uniquely encoding each syllable. It's that they are encoding the information in each syllable. Many syllables have very low information content and must be combined with other syllables to convey information. Many other syllables are redundant.
media molecule's Dreams has the most impressive user interface i've ever used. it uses 2d panels laid out in 3d space but the panels can be anchored and resized in screen space, object space or world space and it's extremely intuitive. i highly recommend people interested in ux/ui check it out
Pressing a key on a keyboard works better than using spatial input from a pointer device for those cases. To click a button/icon there is a feedback loop that requires vision processing, correction, anticipation, and hand eye coordination. You have to iterate many times in this loop to move to the correct location. Often the refresh latency makes this problem even harder.
This problem is better understood when you have a virtual keyboard and only a single pointer.
Ctrl/Cmd-C is easier and faster than moving the mouse to Edit -> Copy.
Creative professionals rely on physical keyboards and buttons on the mouse for actions and only use spatial input for things that are spatially relevant (pan, placement, zoom, rotation, selecting surfaces, edges, vertices, etc).
Creative professionals rely on physical keyboards and buttons on the mouse for actions and only use spatial input for things that are spatially relevant (pan, placement, zoom, rotation, selecting surfaces, edges, vertices, etc).
That used to be true. Autodesk put a lot of effort into interfaces for engineering in 3D. In Inventor, you only need the keyboard to enter numbers or names. They managed to do it all with the mouse. Try Fusion 360 to see this; there's a free demo.
You mean the right-click radial context menus? Yes, those are really nice. They require much less of a feedback loop and often don't require any visual processing at all. They are also more gesture based not buttons. The affordance is very generous compared to buttons and icons.
I use a 3DConnexion Space Navigator (6 DOF) in my left hand and mouse in my right for selection when using Fusion 360 and often use the gestures on the mouse.
I guess that brings up an exception. The context switching cost. Moving from pointer to keyboard is very slow so gestures really help out in that regard. If my hand is already on the keyboard then I have less reason to want to use the gestures.
That, plus the ability to rotate, pan and zoom while you're in the middle of a selection. That's a huge win. In 3D work, you often need to select 2 or more things, and those things may be small and need precise selection. Precision multiple selection is hard.
Before this was worked out, most 3D programs offered four panes, with three axial projections, usually wireframe, plus a solid perspective view. Just so you could select. Now we only need one big 3D pane.
you really dont use context menus for everything. a lot of functionality is based on modal key chords. so, different sequences/chords do different things based on the selected mode. watch this video and tell me this is inefficient: https://youtu.be/31A0s9HDRHU?t=483
As a side note, leafblowers aren't just disruptive to the sound environment: https://twitter.com/touchmoonflower/status/17242321111447757...