All software is political, riddled with biases and potential security risks. Most of the time we ignore the policy of the software, because we either agree with that policy, or are conditioned not to clock it as a “policy”, because “this is just Common Sense™”.

I suspect, if the author would have been more honest with themselves, they’d write something along the lines of “turns out, software is a platform for political action, and it scares me” - an opinion that is very valid, valuable and thought-provoking.

I don’t know of any solution that would allow to scan entire pages, but here is a local tool to get LaTeX math from images via OCR https://github.com/lukas-blecher/LaTeX-OCR

Please elaborate, I’ve been interested in this for awhile - what do you use/recommend for someone who’s new?

Handwriting has been proven to enhance learning in humans, so you are doing great by keeping the habit!

I don’t have much to recommend, but so far this little tool was very useful for me and my math studies: https://github.com/lukas-blecher/LaTeX-OCR

I am not a student, but I learn like a student all the time. I also enjoy handwriting (got an e-ink tablet for that) and knowledge management. I am often dreaming of a “perfect setup” where all I write gets pushed automatically through OCR into my knowledge vault (Obsidian, Logseq or whatever I/my peers happen to use). Even came up with a plan. I hope this new year will leave me enough energy to execute something useful.

Would you like to collaborate on that perhaps?

I am in the process of learning about/choosing shells for my new setup. Can you please elaborate on gpl-vs-mit style - what do you mean? Is it just about licences?

Brace for a hot take.

Most of these points are completely void, not because Linux is the bestest ever, but because the domination of proprietary systems has conditioned most users to comply to a lesser image of “personal computing”.

Things evolve too quickly? Sorry, we have to stay on top on security updates, new standards, hardware support, new features and ways of working - the world is changing, and our tools follow. It’s not a problem, but a natural consequence of progress. The fact that so many people view this as a source of pain in their personal computing is a problem.

Things break? Well too bad, it’s tech - it’s supposed to break. And we a are supposed to be able to fix it. If most users think that fixing tech is “black magic” - that is a VERY big problem.

Way too many choices? No - you just don’t know what you need. It’s silly to expect a Windows or an OSX user to make an informed choice when it comes to software, because they had these choices picked out for them all their life by the proprietor. An abundance of options is never a problem - our inability to orient ourselves among them is.

TLDR: proprietary computing has normalized a lot of brain-dead practices and expectations, so we crave silly and shiny while turning away from smart and pragmatic. We need better computer literacy, better education and better default computing for everyone.

This is very believable, almost to a point when I would love to read something more than a meme

It looks like a disguised Umbrella logo. Creepy, I like it!

I’ve been running arch on XPS 15 a couple years ago with no problems. There was a problem with sound on laptop launch, but a couple of months in the kernel update fixed all the issues. Arch wiki has a lot of info on hardware compatibility, as well as steps you should take to enable something that doesn’t work out of the box

For me NVIM has several really cool advantages: NVIM is really fast. With a good terminal emulator I can open enormous log files and be able to navigate around/search immediately. I have recently pivoted to DevOps, and using VSCode to interact with large log files made me realize how slow and sluggish it actually is.

Motions and modal editing. Plenty of people have already said how fast it is, I will just add that it is also very fun and, if you dig around a bit, not that hard to learn.

Configuration using Lua - I like it because my configs are simple git repos, so the file structure and the logic of configuration is easy for me to work with. I always thought VSCode to be quite awkward to configure. Also, using Lua instead of JSON makes it incredibly flexible, and as a tinkerer I find a lot of joy in customizing things.

NVIM (or VIM) is ubiquitous. You can expect it everywhere, and every other IDE has VIM-like bindings. Learn VIM = be comfortable anywhere.

A personal perk for me personally is that NVIM is designed to be used without a mouse. Mice give me wrist pain, and switching to NVIM made my work a lot more bearable.

If you’re thinking about trying it out, I would recommend going for a community-maintained distributions like AstroNvim or ChadNvim. It’s also quite cool to go back to your preferred editor, knowing your preferences are now more refined after trying alternatives.

Anyway, good luck

The Linux Foundation hosts brilliant courses on OS virtualization in Linux - after that there is a pretty clear path at a cloud administrator career.

Every home alone past 2, every terminator past 2, the latest Matrix practically apologizes for its own existence. Examples are too numerous, the entire hyperreality we exist in is built on pointless repetition and self-cannibalization.

Our world is running out of resources to turn out profit, so it had started digesting itself and feeding us its over-processed and over-produced communion, like a sleezy street-food vendor dousing their meats in spices, so we don’t smell how spoiled their paska is.

That is why everything revolves around the nostalgia: it’s not us who are stuck in the past, it’s our culture experiencing rigor mortis, and we treat it as the final chance to see its original form, as if the chicken in our tavuk durum hasn’t been rotting since yesterweek.

I’ve been toying with the same idea. Having Logseq running in server mode + creating an LSP adapter should be doable.

Knowing Typescript is enough to begin and start a career, you are sure to pick up relevant vanilla JS knowledge along the way. There is no way around JS: you’ll see it in tooling, debugging, building etc.

Of course you can really focus and grok everything way in advance, but I would argue it’s not necessary.

Yes, It takes it’s time indexing the graph - and you have to re-index periodically if you want you queries and graph to be in shape. I have a pretty mature KB, and this process takes no more than a few seconds, so it’s fine.

Honestly, I love everything about it, except for the app itself. It ties me to the default editor, which is an Electron-based sluggish resource hog. I’d rather have some software scaffolding to work in an editor of my preference, but that is just me. I suspect most people in most use cases won’t find it as problematic.

I second Logseq. FOSS, completely local, stores everything in text - works well with Git, lots of plugins - it’s almost perfect