Didn't the same thing happen with authors and google 10+ years ago? That wasn't ever resolved, was it? Normally, I'd look it up, but I have a feeling I'd be spending time to be disappointed by the answer.

Wow, that was a wild read. I kept going to see if the man responsible for Radithor would get his after finding out it made him rich.

That's a big question, but I'll try my best to answer without getting too deep in the weeds.

I'll probably sound like a fanatic, but I use my PKMS for notes, logs, journaling, project and task management, snippets, and documentation. They all have their own structure and flow. It's a Gall's Law kind of situation where I started simple and it worked, so it was extended and slowly evolved to reach it's current complexity.

The beauty of PKMS over a notepad is the loose set of basic features (Wiki-links, tags, templates, etc) that be used in a personalized way to quickly capture, organize, and retrieve info that works best for you and no one else.

As a simple, but detailed example, in the context of learning linux, i might make a "linux" note and dump info there. I put everything in my own words unless I use md quotes (> quoted text) and I add useful links that I also bookmarked in my browser. When the "linux" page gets bloated, I migrate clusters of info into new notes, wiki-linked in the "linux" note. For example a "distros," note which might have some high level comparisons. I favor making new notes over md headers so it's easier to find and open notes by name (a "quick switcher" hotkey as it's called in obsidian). When I settle on a distro I might make a note for it to contain wiki-links of default components EG "apt (package manager)," "gnome (desktop environment)," "x (windowing system)" and dump relevant notes there. If I try wayland, I'd make a "wayland" note but also a "windowing system" note that both wiki-links "x (window system)" and "wayland," and is wiki-linked in each of those notes.

It could get very meticulous, and some folks setup is too much for me, and I'm sure mine is too much for others, but start simple, experiment, find what works, and add to it. In the beginning I had dedicated time just to developing my PKMS. The important thing is quickly recording and retrieving info.

Sometime i do have crazy scrawlings where i just need a notepad to dump info during a deep dive. That would be loosely zettelkasten style with a time-stamped name, sometime with a few extra works for context/search. Sections could be extracted into their own note later. The note itself could be linked to more organized, related notes.

As a more complex, but shorter example, to show how similar tools can be used in a different manner: I'll make a note for a command line program, for example, cat. I have a CLI template with a Useful Flags (options) section. Kind of like a personalized tldr. I'll also have specific notes for complex snippets (AKA one-liners. Real note example: "list-and-sum-all-audio-file-durations") and if it uses cat, i'll tag it #cmd/cat. The CLI template also has a Snippets section that uses dataview to automatically list, in this case, all notes with the #cmd/cat tag. I also have a "command line programs" note that uses a dataview query to list all notes that used the CLI template. Also, a Snippets note using dataview to list all pages created with the snippets template.

There are tools specifically for snippets and personalized tldr, and I may migrate to those eventually-- especially after I have my install script up and running with linked configs-- but the simple tools in PKMSs are really adaptable and make it easy to customize and integrate. Plus it's all md files in a folder, so it's easy to sync and access on multiple machines, including mobile.

I hope that's not TMI. Starting linux can feel overwhelming and I don't want to add to that. Quiet the contrary. I started my PKMS right before my last, permanent switch linux and I think it helped it stick, and 3+ years later I still use [my PKMS] all the time. As I said before, the simple tools that turn a notepad into a PKMS can add a personalized structure to the insane scrawings, making it quick and easy to navigate, find, edit, and add info. You just have to start simple and take your time. I hope that helps. Good luck with the switch!

I use Obsidian. Stores everything in markdown and has a nice sql-query-like plugin, dataview, that I've built a nice workflow around. Obsidian isn't FOSS, which has become more important for me, so I'm looking to migrate over to markdown oxide in helix. If I were starting from scratch I might try logseq or similar. Whatever you choose, I think it's helpful that it's stored in a portable format like md so you can change programs if you need.

1. As others have said, it's possible to play most steam games, but not all. You have to decide if you like those games more than you dislike MS and Goo. I find there's so many great games out there that I'll never get to all of them, so I'm ok with dropping some bangers that usually want too much access to my system.
2. Here's a useful resource if you need to understand slightly technical linux foundations https://linuxjourney.com/ It might not be necessary but it does help to have a foundational understanding, and honestly, the command line is awesome, powerful, and one of my favorite things about linux. Beyond having a basic understanding (and maybe having one of the books the site recommends on hand), before going to an LLM as others have suggested, have official sources of various components bookmarked and go there first. There's so much BS out there now, I actually like the fact that I can read technical documentation, test it out, and know if it's true.

one other tip: I'd recommend some kind of personal knowledge management (PKM) system to take notes. Linux gives you a lot of freedom-- that's what's great about it-- it can be complex and have a learning curve at times. It's absolutely worth it though. It's a totally different paradigm than windows. After a while you can really start crafting the whole system to your needs as an individual. I'm 3 years in and was using my first setup that whole time, i didn't realize how customized I had made it until trying to set it up exactly on a new workstation. Now I'm writing a script so to automate my setup (os settings, program installs, configs) by running a single command. Then I can really start experimenting.

Everybody's different and with a little basic knowledge, everyone's setup can be tweaked to their individual needs a little better than other "user friendly/polished" operating systems. I hope you find as much joy and freedom in it as I do.

I've been tinkering with many of these lately, but I've been surprised by the lack of interoperability. I've yet to work with a bookmark manager that can import and export a netscape html file, without dramatically changing its structure. Of the top 2 for my needs, Linkwarden doesn't export to html and linkding does, but loses the hierarchy.

As I've been working on an install script for making my setup more portable, this is handy and timely. Thanks for sharing!

PS I hate to be the UUOC person. I'm sure you're already aware and it was a deliberate choice.

Not that I have it all figured out, but it sounds like it would help to decouple backup from sync. I have syncthing keep a two-way sync, including deletes, but have syncthing's trash as a "backup" (items deleted after n-days) on each device in case I accidentally delete something. Then I have a nightly, encrypted backups with versions stored offsite (eg borg) which is only meant to be used if there's major failure like a flood or fire. HDD failure is covered by RAID10 NAS. Somewhere in there I have or need a data integrity/hash check, but at least it's a start.

based on nothing.

Your first comment took a detour to follow a valid post with confirmation bias-led ad hominem. I think it was a safe assumption based on your defensiveness, but keep those blinders up.

I often speak positively of specific ML and ai algorithms in this community, and I suspect the only downvotes I get are from ai fanboys who don't like hearing their favorite chatbot is a grift led by billionaires.

There’s no room for rational discussion about AI on Lemmy.

Seems to contradict your first comment, but sure, we're the irrational ones. We must be if we don't like ai /s

Do you have something to add, or do you just want to take a cheap shot at people who are critical of a thing you like?

Most people here have a better understanding of ai than ai consumers i know. I find this community to be anti ai, as big tech hype/marketing brand, not anti ai as a branch of programming that includes ML and efficient, well-scoped models.

Thanks for the rec! The anonymous branches and working-copy-as-commit subsuming git stashes is intriguing. I'll give it a closer look when I have a chance.

Great topic. I'm going to have to investigate some of these suggestions later.

Since my first pick, helix, was already mentioned here and i commented on it, I'll add gitui. Git can be very overwhelming for me. Gitui arranges frequently used git commands in a sensible, visual layout and makes it easy for me to understand and interact with git.

+1 for helix. I was new to linux and TUI editors. The vim tutor was a good intro to the concept of modal editors, but needed lsp and syntax highlighting. At the time I struggled a lot with configs, so neovim was out. Helix is just a fantastic, batteries included experience. Approachable for beginners, but feature rich for novices.

Edit: typo, grammer

Thanks for the info! All good points. I'll keep snikket bookmarked for when I'm more competent in my server/self-hosting abilities and revisit how I chat.

Few things that make my blood boil like Trump saying "J-6 hostages."

Understood. Thank you. It'd been some time since I've scrutinized Signal. It was a set-it and forget-it type situation.

Out of curiosity, what's wrong with signal?

Funny. I was thinking the ai post leaned towards Don Hertzfeldt's style, then saw the "original" and it look even more like Hertzfeldt (see hands), which means ai even manages to fuck up stick-figure fingers!

Edit: btw, good eye/attention to detail!

I appreciated her closing response and the distinction made between over-hyped LLMs and specific ML models.

AI is such an interesting word because it's sort of like the word transportation and that you have bicycles, you have gas guzzling trucks, you have rocket ships, they're all forms of transportation, but they all serve different purposes and they have different cost benefit trade-offs.

And to me the quest to artificial general intelligence has the worst trade-offs because you are trying to build fundamentally an everything machine, but ultimately it can't actually do all of the things. So not only do you confuse the public about what you can actually do with these technologies, which leads to harm because then people start asking it for things like medical information and instead getting medical misinformation back.

But also it requires all of these things that I described, the colossal resource consumption, the colossal labor exploitation. But there are many, many different types of AI technologies that I think are hugely beneficial. And this is task specific models that are meant to target solving a specific well scoped challenge, something like integrating renewable energy into the grid, weather prediction, drug discovery, health care, where you identify cancer earlier on in an MRI scan.

These are all very task specific. It's very clear what the use case is. It's — you can curate very, very small data sets, train them on very, very small computers. And I think if we want broad based benefit from AI, we need broad based distribution of these types of AI technologies across all different industries.

The library