I really like this idea and the app, but beware when using your existing logseq folder it will mess up the structure/indentation/bullet-points of the notes.
Show HN: Reor – An AI note-taking app that runs models locally
81–90 of 107 posts
Re: Show HN: Reor – An AI note-taking app that runs models locally
#82I did my usual test for these things - I tossed in the Markdown source for my site, which has 20 years of notes ( https://taoofmac.com/static/graph ). Surprisingly, indexing sort of worked. But since I have an index.md per folder (so that media is grouped with text for every note) the editor is confused, and clicking on links always took me to a blank screen. Also, pretty much every question gives an error message th…
Also, it destroyed front matter on a few files I clicked on, which is a big no-no. Filed that as an issue.
Re: Show HN: Reor – An AI note-taking app that runs models locally
#83Earlier quoted context omitted.
That was the reason why I gave up Joplin very quickly. The last Joplin thread, here on Hacker News, has also shown once again that some still do not understand why "But Joplin can export Markdown from the database!" is not the same as simple, flat Markdown files.
May I ask what you switched to? Running into the same issue.
On desktop, my favorite text editor plus my favorite file browser. On Android, Markor.
Re: Show HN: Reor – An AI note-taking app that runs models locally
#84since LLM are computer intensive, you should add some hardware requirements in the README
Re: Show HN: Reor – An AI note-taking app that runs models locally
#85Great job! I played around with this on a couple of small knowledge bases using an open Hermes model I had downloaded. The “related notes” feature didn't provide much value in my experience, often the link was so weak it was nonsensical. The Q&A mode was surprisingly helpful for querying notes and providing overviews, but asking anything specific typically just resulted in less than helpful or false answers. I'm sure…
Yeah, this is exciting - I'd much rather have it as a plugin for Obsidian though! I have my workflow with that, all the features I need. Having some separate AI notes app isn't what I would like to use.
Re: Show HN: Reor – An AI note-taking app that runs models locally
#86Running with a Local LLM on a Mac M1, this completely locked up my system for minutes. I tried to let it run, because the progress bar was ticking every now and then, but after 10 minutes I gave up and killed it.
You are running the wrong model. Get a smaller one
Macs are usually ok with their performance since the GPU has access to main memory. A decent Nvidia GPU with lots of memory is still king though.
I run codelama:7b on a MacBook air, and even auto complete is partially usable.
Re: Show HN: Reor – An AI note-taking app that runs models locally
#87Does the future of knowledge management involve using lots of AI to organize pieces of knowledge? I think "here be dragons", and that over-relying on AI to do all your organization for you will very possibly (probably?) cause you to become worse at thinking. No data to back this up because it is still early days in the proliferation of such tools, but historically making learning and thinking and "knowledge managemen…
I don't think that the problem would be becoming worse at thinking, but I see a possible related problem. Each one of us has its own way of organizing things, that looks logical to us but not necessarily to others: think about how you organize things inside your home vs where other people put their stuff. A similar issue could arise with AI tools, that will classify and organize documents based on their logic, which doesn't necessarily align with ours.
Re: Show HN: Reor – An AI note-taking app that runs models locally
#88This is a good reminder of why storing Obsidian notes as individual Markdown files is much more useful than stuffing those notes in a database and having Markdown as an export format. The direct manipulation of files allows multiple apps to coexist and do useful things on top of the same files.
The OP and your comment just made me cancel my Milanote subscription, export all my notes to markdown and start using Obsidian (to later experiment with this Reor).
As a side-effect, I just noticed that I prefer a long markdown file with proper headings (and an outline on the side) than Milanote's board view, which initially felt like a more free form better suited for unorganized thoughts and ideas for writing that I had (I use it for my fiction writing).
I still can have documents as a list of loose thoughts, but once I am ready to organize my ideas, I just use well written and organized headers, edit the content and now I have a really useful view of my idea.
Re: Show HN: Reor – An AI note-taking app that runs models locally
#89Great job! I played around with this on a couple of small knowledge bases using an open Hermes model I had downloaded. The “related notes” feature didn't provide much value in my experience, often the link was so weak it was nonsensical. The Q&A mode was surprisingly helpful for querying notes and providing overviews, but asking anything specific typically just resulted in less than helpful or false answers. I'm sure…
Which model exactly did you use and how large? I feel like even the best 7b models are just a bit too dumb for most things that I have tried. A 70b model or Mixtral or sometimes 34b seem to be adequate for some things. But those are several times larger and don't run on my oldish hardware.
Openhermes 2.5 Mistral 7b
Re: Show HN: Reor – An AI note-taking app that runs models locally
#90serious question: when do you ever have your own notes and can't find the answer ? i would call it bad note taking to not be able to recall an answer you put into your notes. I like the idea of something like this but i've struggled to find a real use case.
Oh I definitely have cases where I'm quite sure that I have had something in my notes and it takes quite an effort to find them.
It is probably relevant to note that I have more than a decade of notes, and more than half of them are not written by me but are excerpts from various relevant resources, scientific papers, etc; so there are cases where, for example, I know that many years ago I had seen an interesting application of method X and should have in my notes, but I don't necessarily remember the exact keywords how it was discussed back then.