kanodiaashu

Interesting take..

I would really want to use this, maybe about once a week, for major PRs. I find it absurd that we all get AI help writing large features but very little help when doing the approx same job in reviewing that code. I actually would even read my own PRs with it, as my workflow with AI is to prompt it to acheive building some feature/goal, then only review the code once things work (this is an oversimplification).

Thank you for noting this. That's an unexpected bug, will fix.

Oh you can actually do that in this! the nodes of the map if you click them have references to the actual 10Ks which are readable as html in the space. Not only can you read the original text, but the html has been 'chapterized'; you can also read summaries of chapters. Further, you can query anything in chat - where if you select the document and ask, it retrieves the right passages to answer your questions too.

Hi all,

I got sick of copy pasting everything I'm reading through deeply into ChatGPT and getting lost in walls of text.

I'm building a new map like interface to understand any topic deeply. As you chat, your map gets built out, kind of like a nice notebook. You can search the web and pull in/upload documents in any format (pdf/epub/html), create maps over those docs for deeper understanding. You can also listen in podcast mode to your map for complete control over what you're listening to. You can semantic search docs, but also the interface has chapter level awareness of each uploaded source.

Samples are also available at proread.ai/pgessays , proread.ai/books , proread.ai/llm-papers.

Here's a couple of recent spaces I was consuming lately 1. economics - https://www.proread.ai/share/3fd6fa9d-5aea-4dd1-a190-eb83a6a... 2. the dollar weakening - https://www.proread.ai/share/3ca2565f-96ed-4abf-a523-e32b591...

Thank you! Its not open source, no. I need to check those out, I have not.

Well, on the flipside of writing with AI, I've been making an app to read papers with AI! https://www.proread.ai/community/ab7bd00c-e017-4de2-b6fb-502... ; Please give me feedback if you try it!

I’m working on something inspired by Cursor and Obsidian — but for research, reading and note-taking.

I’ve not been satisfied with how AI is bolted on after the fact — reading in one place, notes in another, and AI as a separate assistant. So I’m combining them.

It works over EPUBs, PDFs, and HTML — you bring your own sources (BYOS), and it turns them into a AI scaffolded, structured, hierarchical reading experience.

Already using it for both reading and deep research style deep dives, and its already way better than my otherwise broken workflow!

This list is so cool!

This is nice!

This looks exciting! What are the key improvements over the ebooklib python package?

There's some fascinating research by Rachel Griffith which shows that advertising can be significantly welfare reducing for not only customers, but also for companies themselves (they just overall make lower money taken together); it is just another dimension of competition, like pricing/positioning, and adding a meaningful dimension is costly.

Could you say more? Is there a tutorial I could follow?

did you compare with Zoho mail?

How do you actually speedread? And how much time does this take for about a 200 page book? I really like the explaining to friends part; wonder if you could give me a couple of examples of specific books where you did this.

I had kindles for a while until I read that the higher resolution on ipad makes it as easier on the eye and is possibly more important than lesser light on e-ink. Ultimately I settled on the ipad due to the amazing pencil and versatility while reading pdfs of papers/color material, and the ability to search on device far more easily. I've often wanted to switch easily between reading and listening but nothing seems to provide that.

ReadEra is interesting, thanks. I find it hard to only rely in Gutenberg etc for non fiction books.

Didn't know of HathiTrust - interesting, thanks. When you do ask do you manually type/copy paste (i assume so)

that's a large large display..

Two big things immediately: 1. I often ask it to summarize specific chapters or parts of the book after a particular point because the book has started feeling repititive from that point; and then I go into rabbit holes about that content 2. I ask ChatGPT to find related podcasts/content online for specific parts of the book

This reminds me of the search engine aggregators in the old days that used to somehow install themselves on internet explorer and then collected search results from multiple providers and sometimes compared them. I wonder if this time these tools will persist.

I wonder if education/teaching is fundamentally inseparable from, at least to a serious degree, caring. Every single teacher who impacted my life seems to have cared for me. Maybe not. I have learned a lot on my own too. It would be interesting to see the minimum amount of care needed for a successful education; that would directly increase how scalable education can be.

I would 1. Make it more capable as a full GUI - currently I have trouble displaying jupyter notebooks with images/progress bars etc in it. 2. Make an 'emacs app store' where you can get paid plugins (I'm going for Cursor). 3. Make collaborative editing possible.

'Data and Reality' is cool!

oh; thanks for trying it - our linkedin api provider is imperfect. If you could try to put in your resume that would be great. I'm happy to get on a call and talk about it but no pressure.

I'm so reminded of Seeing Like a State (James Scott) where the author describes how much of society as we observe it is a function of designing it for data collection. I feel like there's a whole pedagogy on the philosophy and practice of 'data', and I wasn't aware of it.

Its interesting to me how you start with career maps. Maybe this is advertising, but I made a career mapping app here - https://www.moveup.ai/for-individuals - I wonder if you would find it interesting.

interesting. 'startup idea'.

Well once ads kick in, Idk if there is a way to keep a pro sub - google certainly doesn't have one.

but wouldn't that normalize over time? I think this may already not be true for gpt-4o-mini?

I ssh from a mac into linux servers for my everyday work.

I went from tmux + vim TO (doom) emacs + tramp + projectile. I think I outgrew vim, and emacs EVIL gave me VIM but also magit and org and projectile. I tried to work in the emacs shell for a while. I realised I need tmux persistence still. Emacs shell also just doesn't do well with EVIL mode. So now I just do all development on emacs and manage runs etc through my tmux, occasionally still using vim on the tmux to manage config files when I don't want to load them into emacs because its a one off.

I dk about kitty, but I'm a fan of Kovid Goyal's calibre. I can't imagine how I would live without a. multiplexing b.persistence c.returning me to my context. And I never have had any tmux problems.