Okay, I didn't come here to shill my software or anything. Probably won't even publish it. But I was wondering would you read a book like this? I was messing around with AI to see what you could create, and I tried to create a study app for Aristotle's metaphysics. On the left it has the table of contents where you can jump to the book/parts of the book. In the top-center it has the text where each sentence is numbered. Below it has an AI generated commentary for every single line. On the right it has a chat window where if you have a loaded ai model it will detect it and you can interact with it. You can feed it where you're currently at in the book as context. You can also create bookmarks and there is a search function. There are a couple rough spots, but I mostly got it to do what I wanted it to do. So, I was wondering, if you wanted to study a text would you consider doing it in a format like this? Or do you think this is overcomplicating things and it's better to just pick up a scholarly version of the text and open up ChatGPT in your browser?
>>24678351I just don't see the point in involving ChatGPT at all. I'd wait at least 5 to 10 years before involving AI into any serious interpretation -more so with Aristotle-, even if it's just used to bring resources into the reading. Idk man just pick up Heidegger if you want a deep reading of Aristotle.
>>24678369It's more for the purpose of learning the text. Personally I find it invaluable to be able to ask the AI about any verse and it can explain to you in very simple language what is being said. Commentary can do that but, unless it's the bible, there are often lines that I don't really understand and there's not really comprehensive commentary. The point was for it to be a learning tool not really a way to develop new interpretations of Aristotle. I find Bible websites and software so convenient cause you can just dive straight into the book based on chapter/verse and easily get to that part on the book cause it's all numbered. In normal books you kinda have to flip through it a while til you get to the part you want. And being able to just ask the AI without having to alt-tab all the time seems convenient to me. And the text always being fed to it without having to copy and paste it in. I don't know, seems beneficial to me. But that's why I was asking, is something like this the future you reckon.
>>24678402>it can explain to you in very simple language what is being saidAnglo-simple language-brainrot will be the doom of us all. I now see the monastery existence truly is the only choice.
>>24678402>is something like this the future you reckon.It's horrible, so yes. It might be the future.
>>24678429Often times complex ideas can be explained in much simpler language. Not everyone has a deep background of philosophy. I think stuff like this is good for people who either want to really understand the text or people who don't have much of a background in philosophy or the writer. I don't understand the hate. I mean that's essentially what lectures try to achieve. Boiling down ideas and I don't hear people hate on those very often.
>>24678451>Not everyone has a deep background of philosophy.Philosophy is not for everyone. The simplification of the expression of complex ideas commonly takes away from those complex ideas. Understand concepts and ideas on the grounds thinkers present them if you truly want to understand them or don't pretend to want to understand them (STOP THE LARP)
Let's be serious. If you have to ask AI you are way in over your head anyway. Go back to reading Harry Potter
>>24678451>I don't understand the hateIt stems from the usual and erroneous simplification that comes from AI with relation to philosophy. It turns it to shit. That's why waiting 5 to 10 years like >>24678369 stated would be smart, AI is just extreme shit with relation to these topics.>I mean that's essentially what lectures try to achieveyeah but a lecturer will actually teach you what a thinker is expressing with the complexity worthy of the topic. AI is just not at that level at all.
You niggas really telling me you read through the Organon without needing external help? I'd think something like this would be invaluable. Also, generally, I find AI to be way more useful than lectures because I can ask it about highly specific things. And I never claimed to be some deep philosopher so I don't really know where this LARP is coming from.
>>24678488You've been oneshotted.I bet you'd commit AI guided & assisted suicide like all those kids nowadays
>>24678351Automatic ChatGPT summaries sound like the sloppiest of outputThat said, imagine Cursor for ebooksYou can ask it questions and point the LLM at specific parts to talk aboutI don’t know how you’d manage context windows and on the other hand aren’t all the good books part of the corpus anyway
>>24678351this is a good use of AI. i've been thinking of doing shit like this. i guess i'm embracing da beast. oh well.
>>24678402On the other hand, if an LLM makes up some bullshit (above and beyond what some Redditor or Tumblrina thought about it), how would you know?>alt-tabYou’re going to have to press something to change focus from the text to an LLM input window, and alt-tab ain’t a bad one, maybe with command-shift-o or whatever it is on Windows to make a new ChatGPT Q&A session
>>24678351Hi, I'm an ex ancient philosophy scholar: this definitely sounds interesting and has some potential. If you can keep developing it and find a way to also integrate the original text (or make it so that AI can show you the greek while you read the original and you can ask it questions about the language) this would improve research in the subject immensely.Good job for coming up with this. I also suggest you get in touch with people studying digital humanities either in UK or US universities - their departments are dying and they are on the lookout for people who could come up with stuff like this. Check out their programs, send a couple of emails to professors and see if you can find a way to get inside (if you aren't already).
>>24679555Another thing - usually for ancient philosophical texts there are specific ways of quoting pages which refer to specific editions people take as standards - e.g. the Bekker edition for Aristotle's metaphysics. You can find both the Greek text and the Bekker structure integrated in Perseus Digital Library. People working on that website might also be genuinely be interested in integrating something like your model, so do reach them out, there should be contacts somewhere on this page:>https://www.perseus.tufts.edu/hopper/help/faq
>>24678351This looks like absolute ass only a Linux daily user would even open. Content wise it looks like bible study programs that already exist with actual commentary written by experts. You could do this for other works but if the “feature” is just AI sloppa and not real commentary then what’s the fucking point? In the off chance I need CGPT to word vomit “analysis” based on wikipedia and reddit (genuinely the majority of the content is sourced from this) then I can copypaste the part I need commented on itself or just ask about the work.
>>24678488> I find AI to be way more useful than lectures Jesus Christ dude.
This is insane. Why would you choose to use AI when every word of Aristotle has been commented on for 2500 years by actual humans within the commentator tradition? OP, this could be amazing if instead of ChatGPT garbage, it was Alexander of Aphrodisias, Simplicius, Aquinas, etc juxtaposed with one another and the text by Bekker number.
I don't know. I more did this to figure out how far AI has advanced and see if I could build anything with it. I'd need to give it some thought if I wanted to make it into a more serious project. If I did that I'd probably try to turn it into a more general Aristotelian study tool including the major books. For when you open up the program it gives you the option to choose between the different books. If I got even more ambitious I could probably universalize it where you could just hand it a couple text files of any book, and if it's formatted correctly, would display. Seems to be a lot of hate for AI in this thread though. Think people are a bit closed minded, it's truly an excellent tool.>>24679561Implementing different languages too I don't think would be too hard. You could feed the AI both the English translation and Greek text as context and could ask it how the author translated certain words for example. The Bekker numbers I think would be a bit more tedious because I'm not sure there is a good way to automate it like I could with just having a simple script that checks for ending of sentences and assigns it a line number. The displaying of the text in the program is easy enough it's the getting the numbers right in the file which it reads from would probably take time. >>24680056I just used AI commentary cause I think it's the most useful for me personally. If I was more serious about this I'd figure out my way to get my hands on or at least borrow a more powerful gpu so I could run a bigger model. And I think it would be relatively simple to implement a drop down box and you could switch to other commentary. The other problem is those translations, probably many of them, their translations aren't in the public domain. Which if they're not would essentially mean I couldn't distribute them.
having multiple translations would be way more helpful then an AI commentary, maybe something that explains teh different usages of specific greek terms in this work and in other ones, and an index so you cna look up how a word is used/translated throughout the book if it's unclear
>>24679436What is the difference between a human making up bullshit vs a computer? I would prefer a human's interpretation, by far, but the fact remains that I take from the explanation what I will; in both cases I selectively and critically choose the analysis that joins my own.
>>24680661>What is the difference between a human making up bullshit vs a computer?There isn’t. Anons here will bitch and moan but none of them can compete with a machine’s explanation. They can poke holes at it for flaws, sure, but their own explanations will rife with similar flaws. And at the very least an AI will admit to being wrong and try to find a correction. Already that makes it more honest and useful than the vast majority of people here.
>>24680674you have never heard of AI hallucinations
>>24680750>ai makes up things>people also make up things>or just flat out lie Retard
AI is the death of thinking. Sad end for humanity.
Knowledge without discipline will ruin you. Not because the weight of it is immense, but because you will be holding something without knowing the significance. And when the time comes for it to be applied, you will flounder and fumble for not knowing the path: you just know that the clearing exists, not how to get there. Machines will not replace cognition solely because we, as of now, do not know each other and cannot know each other. Tell me you know another man besides yourself and you’ll look like a fool. Because what interior have you been within besides your own? There is nothing you can perceive outside of that. There is speculation, accurate guesses and prediction: but you will never see another soul.
>>24680767Tell it to your teacher
