queerlilhayseed

I have never wanted to shake anything as bad as I’ve wanted to shake a printer.

We can’t rule out that they may be a plant.

That sounds really nice. The peer network is definitely more of a long-term thing, but it would be really nice to have a strong municipal peer network in addition to long-range providers. I’m not saying ISPs need to go away entirely, but I would like the world to be in a place where they could disappear and the internet could still function.

Easy is the goal, but it turns out mobile OSes are hard to make, harder when the entire for-profit ecosystem is actively trying to undercut free solutions. Right now I’m working with old devices that, while they are running postmarketOS, are still pretty janky and have a lot of missing functionality. Still, a lot progress is being made, by an army of regular folks who pitch in when and where they can, and somehow the boulder of free software slowly but inexorably rolls uphill. It’s very cool to watch.

Not incoherent at all, that sounds really nice. One of my ambitions is to have my “main” cell phone just be a postmarketOS device that can receive calls and texts and select encrypted data from home base. Eventually I’d like to be able to just flash an image onto a cell phone and have it hooked into my VPN and basically just have it communicate solely with my home server, but that dream is still a ways away.

It would be really nice to have a community where everyone had household compute that was their “main” computer. They could be the hubs for communications that’s always on, and you could set it up to communicate with an otherwise anonymous public handheld, or just leave it until you get home and decide to get online. handhelds wouldn’t need to be repositories of sensitive data, they could just be thin clients that could be safely wiped and restored (without needing a corporate entity’s help or permission) in the event of damage or theft. Home networks could become overlapping network nodes, reducing or eliminating the need for regional or global internet provider services. We could really run the internet ourselves.

It looks like a dog ate bird poop and threw up on a biscuit. It’s really good though.

When I was younger there was a hot new book called The Artist’s Way. I never finished it, I don’t even know if I would recommend it, but one of the practices it recommends is morning pages, and morning pages helped me.

What you do is you get a pen and a notebook, and write three pages every morning. If you miss a morning, wait til next morning and write three pages. It doesn’t have to be good or even sensical. A lot of my early morning pages were line after line of “I don’t want to do this, this is bullshit, I’m just kiiiiiiiilllllllling trees and wasting space this is dumb”. Eventually I got bored of writing that stuff and started writing other things, but it took a long time.

The trick is you gotta do it every day, even on days you don’t want to. No one but you can make you write so if you don’t do it, you won’t do it.

Oh it’s pandemonium. My nipples can launch a thousand ships per minute. Each.

My alt-text isn’t showing up and I spent too long on it to give it up so I’m putting it in a comment:

It’s such a defeated attitude. Back in my day the internet Nazis at least pretended to give a shit about their ideology, now it’s just “we can’t let them take away our wallowing drivel, otherwise we would have nothing left to wallow in” like, Really? the best defense you can come up with is “it’s better than nothing?” When was nothing an option?

Oh right, when nazi ideology dictated that the only legitimate culture is its own. God damn, nazis, it must be bleak to live in a world where you’re not allowed to enjoy any good art. No wonder you’re such cranky assholes all the time.

LMAO “not rehosting nazi content” -> “we’d have no content” is a wild statement.

We could be the great scourge of this part of the universe. Expanding recklessly through galaxies in the local group, leaving only the dead husks of stars that have been stripped of all usable energy. The exponential nature of scientific discovery means that not only do we have a head start, our head start compounds as time progresses. We become a horrific but very efficient war machine for the sole purpose of controlling and exploiting all available forms of energy for profit. We seem like we’re on that path.

Specifically for writing textbooks, you might enjoy looking at https://www.americanyawp.com/ for inspiration. Their documentation is light on technical details but you could email the contacts at the bottom of https://www.americanyawp.com/text/about/ with questions about how to structure your specific project.

I haven’t found anything better than codeberg. The following novella is a reflection on my experience with a similar problem.

I have all my poetry and short stories (and D&D snippets and board game / video game design snippets and screenplays and draft policy proposals and diatribes on the nature of being and other misc stuff that’s hard to categorize) in a single giant git repo. It works okay though I find it really hard to organize everything in a strictly hierarchical structure and I often have a hard time finding things I wrote because I can’t remember if it’s in the “boardgames” dir or the “video_games” dir, etc. I don’t think anyone but me would ever be able to use it. Not because it’s too esoteric, but because the organizational structure grew up organically over the years and makes no sense when viewed in its current form without an understanding of the history of the repo. It’s a very similar phenomenon to opening up an old code repo for the first time and being overwhelmed by the messiness of it, when someone who has been maintaining the repo for years has a built up schema of how the repo is acutally organized and can navigate it somewhat more effectively than someone new to the project

What I’m saying is that keeping git repos organized sensibly over time is really hard, even when dealing with something as highly structured as source code, and IMO it’s much harder with more loosely-categorized creative writing snippets, if for no other reason than there’s not a strong tradition of creative writers using collaborative editing tools like VCS (though there are, I think, other good reasons why it’s harder).

If I were to reformat my repo, I would probably start by trying to come up with a more formalized type system, a formalized metadata system, and a linking system (e.g. I would like to be able to create a link from a scene in a script to a character bio or a macguffin description). You can do links with markdown if you’re familar with it, I sometimes do that but I am pretty inconsistent with it, most of the time its too much bother to maintain them when I keep shuffling documents around (because I don’t have a well-defined type system…). Obsidian also has this feature; I haven’t been able to get myself into a natural flow writing Obsidian documents, but it’s worth checking out I think as a tool for creative writing on top of a VCS like git. If none of these sound immediately useful they can be put off until they’re needed since it’s easy to overcomplicate projects in the beginning by anticipating future problems that may never materialize, these are just things that I want for my own writing style.

Whether or not you use something like obsidian, I think for creative collaborative projects it’s perfectly doable to use git, but it makes sense to spend some time at the top thinking about the shared rules that collaborators will need to know in order to keep all the contributors from stepping on each others’ toes. These rules should be in the repo itself in a README and should be regularly reviewed by collaborators because you will almost certainly find better ways of doing things as the project grows. These rules should cover things like:

• How are files structured? Does every project get its own top-level dir?
• Do projects share a common structure or does every project get to define its own structure? if the latter, should each project be a separate repo?
• Do you have a metadata system? If so, what types of metadata do you anticipate and how should it be formatted?
• Who gets to commit to each project? Is there a review process? The answers to these can be “anybody” and “no” but I would recommend at least thinking about it before
• how do edits work? Are you going to commit everything to main before it goes to editorial review? or do you want to have feature branches that editors review and approve before merging to main?
• How does the team update the rules? which collaborators can submit PRs for the rules document? what is the mechanism for making sure that the rules are being followed, at least well enough to keep the project usable?

If you are like me, all this might sound like extreme overkill for something as simple as loosely structured creative writing. That’s because it is, until a project grows beyond a certain size, and historically I do not recognize that size boundary until I am well past it and my repo is an unmaintainable mess. It might help to discuss with your collaborators what they think the project would look like when it’s a few years and a few complete projects old: what problems will you want to solve then? Can they be anticipated in a way that doesn’t unduly burden the very difficult task of getting the whole endeavor started? A little of that can go a long way to preventing the project from losing steam because they repo just gets too unwieldy to be fun to write in.

I exclusively use black beans in my chili, I like 'em better. I think it’s a good idea.

I’ll enable FDE during the install for systems with sensitive data, but I don’t bother with secure boot. If I were deploying machines in unsecured areas (i.e. not my house) that also had sensitive data on board, I might look into it.

Maybe this is reflective of my media bubble, but I’ve never encountered someone claiming that LLMs should be used as calculators. Most of the advertising I’ve seen (not much) is mostly centered around natural language search and image recognition. I only really hear about them being bad at math from detractors, and I think it misses the mark of why AI companies are dangerous. The problem with LLMs is not that they’re bad at math, or even that they get non-math answers wrong sometimes. The problem is when they’re controlled by humans with a political axe to grind, who deliberately wish to obscure or distort the information their users can access, c.f. Grok.

I generally make sure the entire block is evenly coated in a thin layer (a shop rag helps), then stick it in the oven at 450F for an hour to let the patina cure. After a few coats you should have a nice, even surface.

I hear this criticism of LLMs all the time and I just don’t get it. They’re language models, they take language inputs and produce language outputs. They aren’t designed to do math. It’s like complaining that a reciprocating saw can’t do math.