The style of the website reminds me of https://faircoding.com/ai/, which I suppose is an example of these recent LLM-generated websites you're talking about. It likewise has dark background and semi-transparent cards with thin non-transparent borders.
There are over 7000 languages in the world, around half of them dying or having already died due to linguistic domination, in large part English, each with its own set of culturally sensitive words.
To follow the above mode of reasoning without advantaging one or few languages, you would have to change an enormous amount of words in all languages, if not basically all. This is obviously not feasible.
If GIMP was a dirty word in a Native American language, or a native African language, there would be no debate. That we are debating this at all is because English has privileged status due to the Anglo-Saxon hegemony.
Hence, you are expecting us to give special, privileged treatment to the linguistic sensitivities of your dominant culture. Which is unfair, especially historically, because the hegemony was achieved by mass land steal and many genocides, which we shouldn't be rewarding by allowing further claims.
So yes, it should be expected from an adult anglophone to tolerate the existence of sordophones, words that are dirty in their dialect but not in others, especially in an international, multilingual setting. This is what it means to abstain from linguistic imperialism. This is what it means to tolerate and respect other cultures.
And to enforce tolerance, indeed it may be needed to view those who fail at this as childish.
I feel somewhat sorry to say this, but I need to be assertive here.
> And to enforce tolerance, indeed it may be needed to view those who fail at this as childish.
No, it's not necessary to denigrate other people under the belief you can police others by proxy.
"Is this derogatory or offensive?" is a basic localization question that is constantly asked in many languages. Yes, including Arabic.
I generally agree about the evils of linguistic imperialism. But I'm describing the world I live in, not the one I want to create.
But that's beside the point. "Linguistic imperialism" is the wrong lens to use here to defend the name. GIMP is not a sordophone, it's the opposite.
GIMP was named by American-born English speakers with the intent to have an edgy name. GIMP was chosen in reference to the full-body sex garment, because they were college kids and that's funny when you're 23.
The intent was offense. It worked well. It's no surprise that GIMP is only well-adopted where the word doesn't carry its offensive meaning.
>"Is this derogatory or offensive?" is a basic localization question that is constantly asked in many languages. Yes, including Arabic.
While it is pragmatic to chose new names to be appealing to members of dominant societies (I do that too), it is problematic when dominant groups view themselves as entitled to that, which is the case here, and which is why we have this discussion.
>The intent was offense.
First, I am not aware of any evidence that there was an intent to offend. The only source for etymology I know here is an old interview with one of the original developers where he said that he blended the words GNU and Image Manipulation Program, and soon afterwards realized that he heard that word before in a film. There was no suggestion there that he wanted to upset others.
And even if the name was really intended to be edgy, the current developers, who have inherited the codebase from the original authors over two decades ago, view it differently and dissasociate themselves from that etymology in the FAQ. This should be sufficient to close this line of reasoning.
Finally, regarding adoption: I can't tell for sure what it is like for graphics editors, but I haven't ever seen anyone not using SRAM memory and OSRAM lightbulbs in Poland because their names are sordophonic to Polish verbs about defecation (in fact, because of that OSRAM is the only lightbulb brand that I can name from memory). Or even anyone complaining about that, apart from being amused. And I wouldn't dare to demand for these names to change just because they have dirty associations in my language when read a certain way.
> problematic when dominant groups view themselves as entitled to that, which is the case here
That is not what is happening here.
> There was no suggestion there that he wanted to upset others.
As someone else pointed out, that's a misunderstanding of the interview. As I've said several times, the GIMP is named after the full-body sex garment. (It's just an unfortunate thing that the word is also a slur for someone with motor disabilities).
> the current developers ... view it differently
I would need a source for this. My understanding is everyone is aware of the name and has been steadfast by it for years.
> This should be sufficient to close this line of reasoning.
No, it is not. You imagined how the developers must feel. And even then, it does not matter how the developers feel.
> I haven't ever seen anyone not using SRAM memory and OSRAM lightbulbs in Poland
That's wonderful, but this is not an analogous situation. I don't think you're even reading my post. "Gimp" is not a sordophone, it's a derogatory term and the name of a full-body sex garment.
> I wouldn't dare to demand for these names to change
Congratulations for you, but nobody's talking about that. It's not the question at hand. The question is whether or not GIMPs adoption and investment was hurt because the images the name conjures up.
And to be clear, I don't think it's a given! The most generous interpretation is that they chose the name to deter users, contributors, and investment. These aren't necessarily measures of success.
For example, if a friend named their bicycle repair shop "Grandma's Diarrhea Yogurt Warehouse", I'd wonder why they chose that name, but I'd assume they aren't trying to run a profitable business. If they told me it was actually an elaborate acronym, we'd both know that they're acting facetiously. (Of course, this is not analogous, as 'Grandmas Diarrhea' is not as belligerent a term as 'gimp'.)
All I'm arguing for is that GIMP is less adopted and less used than it would have been if it were named better. I am describing things that we already know to have happened, and which I and others in this thread have observed firsthand. There's nothing to do thought experiments about.
Exactly. As I think about it, I believe we have a pretty good thought experiment. What if "Audacity" (a program that's doing pretty good that IMHO actually also has a pretty crappy UI) was called, like "flatulence" or "Impotence?" I doubt it would be sitting on the 100 million ish downloads it has today.
mikolajw> I am not aware of any evidence that there was an intent to offend. The only source for etymology I know here is an old interview with one of the original developers where he said that he blended the words GNU and Image Manipulation Program, and soon afterwards realized that he heard that word before in a film.
That's wrong on every count. The primary source is Peter Mattis' own words in the GIMP Gazette interview, January 1, 1997, by Zachary Beane:
Mattis> "It took us a little while to come up with the name. We knew we wanted an image manipulation program like Photoshop, but the name IMP sounded wrong. We also tossed around XIMP (X Image Manipulation Program) following the rule of when in doubt prefix an X for X11 based programs. At the time, Pulp Fiction was the hot movie and a single word popped into my mind while we were tossing out name ideas. It only took a few more minutes to determine what the 'G' stood for."
So the sequence was: IMP (rejected) -> XIMP (rejected) -> Pulp Fiction inspires "GIMP" -> they reverse-engineered "General" as the G. The Pulp Fiction reference was the generative act, not an afterthought.
The GNU backronym came later. Same interview:
Mattis> "the GIMP originally stood for General Image Manipulation Program, but has since been dubbed GNU software by Richard Stallman (with our agreement). Spencer and I decided that GNU Image Manipulation Program is a better usage of the 'G'."
He didn't "blend GNU and Image Manipulation Program." He didn't "realize afterwards he'd heard the word in a film." He was a college kid at UC Berkeley in 1995, Pulp Fiction was everywhere, they needed a name, and the word popped into his head. He says so plainly.
Note Mattis' original Usenet announcement uses the phrase "The GIMP" -- with the definite article, exactly like the movie character is called "The Gimp." That's not how you title software. You don't say "The Photoshop" or "The EMACS." You say "The Gimp" because there's a character called The Gimp, and everybody in 1995 knew exactly which one.
Peter Mattis' original Usenet announcement, comp.os.linux.development.apps, November 21, 1995:
Thanks for correcting me, I should have read the interview more carefully.
>That's not how you title software. You don't say "The Photoshop" or "The EMACS."
Nitpick: The Sims, The WELL, TheFacebook are all attestable. But yeah, fair.
>The number one association most of the population of Earth have with the word "GIMP" is:
>Bring Out the Gimp - Pulp Fiction (9/12) Movie CLIP (1994) HD:
I suppose that here you didn't mean that most of the population of Earth watched and understood the English version of Pulp Fiction, or that the majority of people in the world will associate the word "gimp" with anything else than the graphics editor (certainly neither would be true), but that the second most-common association among all anglophones taken together after the GIMP editor itself is the Gimp character from Pulp Fiction.
Gonna have to say this a bunch around here, but yours is yet ANOTHER comment shooting the messenger. You (theoretically) are championing an idea of freedom in language or something like that.
Look, people, this is PR. The author wondered out loud "why isn't he more recognized" and a reasonable answer is that "People like me, in America, who love free software and try to get people using it, run into trouble that could have been avoided if the name was changed."
You want your lesson out there on freedom of language, fine, that's what you all got. Just be honest about what you may have missed -- which I genuinely believe could have been a world in which Adobe was nowhere near as annoyingly powerful as it is (or at least had been).
The vast majority of the outside world does not perceive the name the way you do. Even the majority of English users doesn't, as most of them learned standard English as second language at school without being taught vulgar Anglo-American slang.
If you want to pursue linguistic sensitivity, the just direction is against anglophone domination, even if impractical. We should be taking power away from the most powerful and redistributing it back to the weak, not the other way around.
So, it is the anglophones who should stop calling people using a nasty word instead of expecting international, multilingual communities to adapt to their culture.
Let me be clear, this has nothing to do with pushing "linguistic sensitivity" for its own sake. Nothing.
Look, the author himself wondered out loud why he wasn't more deservedly popular. I put forth a reason why. PR is real. I understand if you really want to keep the name at the possible cost of popularity; but I don't think it's worth it. I wonder if he understood the possible tradeoff.
Tukaani website states "jiatan" as the nickname of the malicious code committer on Libera Chat.
WHOWAS jiatan provided me the following information:
jiatan ~jiatan 185.128.24.163 * :Jia Tan
jiatan 185.128.24.163 :actually using host
jiatan jiatan :was logged in as
jiatan tungsten.libera.chat :Fri Mar 14:47:40 2024
WHOIS yields nothing, the user is not present on the network at the moment.
Given that 185.128.24.163 is covered with a range-block on the English Wikipedia, it appears this is a proxy.
It's single-threaded, alpha-quality software, and still isn't compatible with many forums and themes. But it can export WARCs and may just happen to work for you.
Though I dislike DeVault myself, this is a well-researched blog post that clearly demonstrates that Stallman has an oddly-specific interest in defending adults having sex with adolescents. I was very weakly hoping for Stallman's insistence on distinguishing between children and adolescents to come from a position of a broader youth rights, anti-adultist advocate, but this long, recurrent pattern of very specific focus on the subject of sex makes it clear to me that, indeed, this isn't where it comes from.
I wish ML researchers (EDIT: and engineers and journalists) stopped using anthropomorphizing language. This has decades of solid tradition, but that's no excuse. Any comparison of a machine to a human misleads the public. Machines aren't like babies, artificial neural networks aren't like actual neural networks or brains. Machines shouldn't be given human names (PLATO is a borderline case).
I know this is like talking to a wall -- money requires hype -- but still, please stop doing that.
> Actual language used by the ML researchers: "Intuitive physics learning in a deep-learning model inspired by developmental psychology"
In my opinion, this is still anthropomorphizing the algorithms. The term deep-learning is a poor representation of what actually goes on. Someone please correct me if I'm wrong, but all ML does is statistical regressions (in essence). It doesn't "learn" like a person learns. Neural networks are not actually like brains (as far as we understand how the brain works).
I feel like the whole industry is inundated with aphorisms that are kind of true, but not wholly true. Evolutionary algorithms, neural networks, deep learning, deep mind, this stuff all reeks of anthropomorphizing fundamentally mathematical processes. I get it, it's a lot easier to get the gist of "the computer is learning/training" than "the computer is refining the weights and biases to try to optimize the output".
Well it's not called person-learning-machine is it? Why would something have to learn "like a person" to be able to use the word "learn", those two concepts are not attached one to the other. If they were, saying "learn like a person" would be a pleonasm, yet it isn't.
IMHO "learning" is a fine term, it conveys the idea of what is happening effectively and quickly.
Also, we don't know how a person learns anyway, it might very well be a similar process, just way more efficient and complex.
> Evolutionary algorithms
How would you propose calling them? You have a generation of agents, each with their own specificities, and from the agents most successful to accomplish the task at hand, we derive a new generation, slightly modified from their parent's.
It seems to me "evolution" is again the most suitable and efficient way of describing what is happening.
While I agree that there is definitely too much anthropomorphizing surrounding AI, I feel you are going way too far in the opposite direction. Not every word that can be composed with natural process/humans should be banned from being used anywhere else.
Who cares? Why should you hold the idea that it would? They are systems moulded after data, 'learn' seemed to be a decent label. If it is not, it is because "learning" is _active_, by philological analysis, and happily consistently with an aim of AGI (intelligent entities learn actively).
A computer does not compute like a human would. Yet, no problem.
For that matter, you are using 'person' in a very individual way - not even "personal" (a "person" learns according to individual nature, while you are using it as a collective term).
As already expressed - nearby I wrote 'biomimicry' -, what you are calling "anthropomorphizing" is a wrong direction: "evolutionary algorithms" were born out of keys after the observation of the natural world, and the terms express that - it is not that you saw the algorithm and went "It looks like my uncle Oscar"¹ (this side is active - it "learns").
(¹Those anthropomorphizing Hollywood cultists and all that sculpture...)
> a "person" learns according to individual nature, while you are using it as a collective term
There's a very specific definition for learn:
>> to gain knowledge or understanding of or skill in by study, instruction, or experience[0]
There's a few more, but none of the definitions treat learn in a non-collective way. I guess meriam Websters dictionary doesn't like treating people as individuals or something lol.
Additionally, all the definitions there are speaking in human contexts. They talk about learning in the sense of being taught, or gaining experience, or gaining knowledge. Sure a computer kind of does this stuff, but it doesn't really. And that falls into the category of attributing human characteristics to an inanimate object.
I probably shouldn't have said that everything in the short list I wrote reeked of anthropomorphizing processes. But the evolutionary algorithm was more in line with what I mentioned immediately before. My whole comment read:
> I feel like the whole industry is inundated with aphorisms that are kind of true, but not wholly true. Evolutionary algorithms, neural networks, deep learning, deep mind, this stuff all reeks of anthropomorphizing fundamentally mathematical processes.
An evolutionary algorithm definitely falls into the category of kind of true but not wholly true. But it's not anthropomorphic.
> intelligent entities learn actively
Also, this is a very loaded statement. What is an intelligent entity? If you Google "is a computer intelligent" there are various papers, articles, and other pieces of media all claiming that we can't call a computer intelligent, and some claiming that we can consider certain algorithms somewhat intelligent. This is anything but an accepted standard today.
Give us an example of some relevant label that would be "«wholly true»" instead of "«just kind of true»". Because metaphors, and the whole system of fuzzy pattern relations, are based on fuzzy pattern relations.
> none of the definitions treat learn in a non-collective way
You have misunderstood my post. I would prefer that you read it again.
You are complaining about loose use of the language: I noted that you yourself used the term 'person' more than loosely, with a dubious jump. When one says '«like a person learns»', that is supposed to be "like a specific individual in his own individual characteristics will learn" - instead you used to say "like people in general learn". A "person" is a "definite form", not a general individual representing common features - it is the opposite.
> very loaded statement
Which you are taking out of context. I said that you have to call the moulding of your functions something, and that "learn" seems a very acceptable term, since it is bottom-up instead of top-down, it is automated instead of encoded: it is developed against data, it "learns". And that if the term is disliked, there could be a very good reason, because 'learn' was born as a sort of a hunting term¹ - it really means something like "investigate" -, which is a happy coincidence because what is largely missing in AI is critical thinking, part of the active process of learning ("learning" is active as investigation is). And the day John will have to check «accepted standard[s]» to see how things are, I will be willing to comply to his sad request for mercy.²
¹Irregardless of what the Merriam-Webster will write, because you get a none-the-wiser relative notion but not knowledge from a "dictionary of use", as at the paragraph for 'life' you will not find the meaning of life.
²John must be, tautologically, an "active learner". (He will check personally.)
You're right that journalists use anthropomorphization much more. But AI researchers also have a long history of choosing terms that are anthropomorphizing or animating. Here the name PLATO -- which evokes an image of an ancient philosopher, a human, who is by cultural tradition considered smart -- is used in the original journal article.
Terms like "neural network" and "artificial intelligence" are frequently used by AI engineers and researchers despite the obvious image they evoke. Sometimes they even call their creations "brains". Also note the name DeepMind.
To add to that, often EDITORS are the ones who come up with the titles, for reasons beyond being clear, like using words that draw attention and to fit in a specific space.
My pet peeve is when AI researchers coin new terms for objects that can be described by well-established mathematical terms. For example, saying a neural network layer has "256 units" instead of "output dimensionality of 256".
But at some point you need to name things for brevity. I understand why people say "activation function" instead of "elementwise monotonic nonlinear function".
Misuse is also rampant, like using "inference" to describe evaluating a neural network on an input, even when the NN isn't part of a probabilistic model.
To be fair, give high degrees of interdisciplinarity and imperfect acquaintance with all the terminologies (and imperfect memory), and given that we mix natural language and conventional technical language, and with some continuity, and given that natural language itself mixes original core root meanings and posterior conventions, and given that even biologically the best term may be occasionally (polysense) hard to find, the mess is expected.
> artificial neural networks aren't like actual neural networks or brains
Just to zoom right in on neural networks:
People often say this, and I never see a solid argument.
I know very little about biological neural networks.
Clearly they are very different in some respects, for example, meat vs silicon.
But I never see a good argument that there's no perspective from which the computational structure is similar.
Yes, the low level structure, and the optimization is different, but so? You can run quicksort on a computer made of water and wood, or vaccum tubes, or transistors, and it's still quicksort.
Are we sure there aren't similarities in terms of how the various neural networks process information? I would be interested in argument for this claim.
After all, the artificial neural networks are achieving useful high level functionality, like recognizing shapes.
There are many ways one can argue for or against this comparison. This is mostly a matter of terminology. However the problem is that the field of AI has been for many decades consistently shaping its language to evoke human-like connotations in order to boost hype. This article's title is a yet another example of that.
There are a few conceptual differences where artificial neural networks conceptually diverge for computational reasons.
One is the notion of time and connectivity loops - overwhelmingly, ANNs use a feed-forward architecture where the network is a directional graph without loop and some input is transformed to some output in a single pass - and weights can be adjusted in a single reverse pass, which is very practical for training. We do know that biological brains have some behavior that relies on signals "looping through" the neurons, and that is fundamentally different from, for example, running some network iteratively (like generating text word-by-word via GPT-3). We have artificial neural network simulations that do things like this, and also simulations of "spike-train" networks (which can model other time-related aspects which glorified perceptrons can't), but we don't use them in practice since the computational overhead means that for most common ML tasks we can get better performance by using an architecture that's easy to compute and allows to use a few orders of magnitude more parameters, as size matters more.
It is not the case - this is just biomimicry: "let us try imitating feats of a living organism". Perfectly legitimate. Nobody is told to make unduly images out of it.
"DeepMind AI learns simple physics like a baby" clearly makes an unduly image out of it. Calling it PLATO evokes an image of an ancient human philosopher. No other field uses as many bold comparisons to humans as artificial intelligence (its name alone is one).
But you are supposed "not to evoke": that'd be sensational.
"The name of AI": we call "intelligent" in this convention that which finds solutions - normally the natural intelligence of a professional, sometimes the artificial intelligence of a computerized system. As easy as that. It works, no intrinsic issue.
This experiment: children seem to rely on expectations in learning and this ANN based system tries to implement some form of "expectation based learning". No problem.
The big one, OPT-175B, isn't an open model. The word "open" in technology means that everyone has equal access (viz. "open source software" and "open source hardware"). The article says that research access will be provided upon request for "academic researchers; those affiliated with organizations in government, civil society, and academia; and those in industry research laboratories.".
Don't assume any good intent from Facebook. This is obviously the same strategy large proprietary software companies have been using for a long time to reinforce their monopolies/oligopolies. They want to embed themselves in the so-called "public sector" (academia and state institutions), so that they get free advertising for taxpayer money. Ordinary people like most of us here won't be able to use it despite paying taxes.
Some primary mechanisms of this advertising method:
1. Schools and universities frequently use the discounted or gratis access they have to give courses for students, often causing students to be only specialized in the monopolist's proprietary software/services.
2. State institutions will require applicants to be well-versed in monopolist's proprietary software/services because they are using it.
3. Appearance of academic papers that reference this software/services will attract more people to use them.
Some examples of companies utilizing this strategy:
Microsoft - Gives Microsoft Office 365 access for "free" to schools and universities.
Mathworks - Gives discounts to schools and universities.
The other ones are smaller but not much worse according to their tests (oddly, in the Winograd Schema Challenge and Commitment Bank tasks, the largest model actually appears to be worse than much smaller ones).
30B parameter models are already large enough to exhibit some of the more interesting emergent phenomena of LLMs. Quantized to 8 bits, it might be possible to squeeze into 2, better three 3090s. But the models also seem undercooked, slightly to strongly under-performing GPT-3 in a lot of tasks. To further train the same model is now looking at > 100 GB, possibly 200GB of VRAM. Point being, this is no small thing they're offering and certainly preferable to being put on a waiting list for a paid API. The 6.7B and 13B parameter models seem the best bang for your buck as an individual.
Mathematica has been on my mind since high school because we got it for free. I went through the free trial process recently and tried a couple of things I have been too lazy to manually code up (some video analysis). It was too slow to be useful. My notebooks that were analyzing videos just locked up while processing was going on, and Mathematica bogged down too much to even save the notebook with its "I'm crashing, try and save stuff" mode. I ultimately found it a waste of time for general purpose programming; the library functions as documented were much better than library functions I could get for a free language, but they just wouldn't run and keep the "respond to the UI" thread alive.
So basically all their advertising money ended up being wasted because they can't fork off ffmpeg or whatever. Still very good at symbolic calculus and things like that, though.
I'm afraid of companies pushing large scale models as the end all for anything text related. Large language models are revolutionary but the last thing I want to see is everything being run through an API. I'm more interested in things like knowledge distillation or prompt tuning. The hope is that a medium size model with some training can match a large one large one using zero shot approaches
reply