Hinori

joined 7 months ago
[–] Hinori@scribe.disroot.org 1 points 18 hours ago (1 children)

«It's all your fault!» — Sony to Sam Altman, probably

Also, I have a strange déjà vu: wasn't there a native Bloodborne recompilation a couple years ago? I remember watching a few videos about it. If I recall correctly, there was a strange cape animation, as it's internally rendered in 30 FPS, or something like that. Or was it an emulator?

[–] Hinori@scribe.disroot.org 2 points 18 hours ago

While I mostly sit around and watch the sunset, lowering my expectations to nearly zero and abandoning hope and any long-term goals, I also passionately use Free Software, write some unapologetically apophenic theories about VTubers, and slowly translate some good games and VNs into my native language. This is my way of how to stay sane and connect the world!

Spoiler:It's not very effective, though.

[–] Hinori@scribe.disroot.org 2 points 3 days ago

Good question! I'd say there should be a mod manager on the receiving end. Something like Limo, that has both generic and some game-specific deployers. But that's only part of what I want: there should also be a middleware to manage differential downloads between it and the origin.

And when I think about it, it becomes clear that all of this has to be backed by P2P of some sort, as there is no such service that could reliably provide even a simple direct link to an arbitrary binary file, let alone allow you to run some odd directory diff protocol. Unless you have your own server, but this is obviously not an option. So it seems like P2P is the only viable option.

So there should be at least 3 components: a mod manager, a directory diff middleware, and a P2P daemon, and they should all work together. And I just realized that the «directory diff middleware» sounds just like OSTree! But it should be able to pull/push from some kind of P2P remotes instead of HTTP ones. Or something like that.

[–] Hinori@scribe.disroot.org 2 points 4 days ago

At least for YouTube clickbait there is crowdsourced DeArrow, from the same developer as SponsorBlock. And you can use it in FreeTube!

[–] Hinori@scribe.disroot.org 2 points 4 days ago (2 children)

Let me first describe a very specific problem that I'd like to solve.

So, I make mods for some games and VN (to be more specific, it's unofficial translations). Sometimes I have to update it and release a new version (and sometimes I just release it in an ongoing mode, part by part, when the next chapter is translated). In either case, I have to ask my audience to manually re-download the entire archive and re-apply it on top of the game directory. Although I manage to make patches pretty lightweight, it still feels so wasteful to re-download the whole archive, even if only a few files are changed!

So, I'd like to have something that could help upload these patches, distribute them, and apply them easily. It should be able to calculate the diff between the local copy and the remote origin, download only required files, and then apply them on top of the game directory. Something like Steam Depots, but standalone. And it should be easy to use, so rsync doesn't count!

Does all of this make sense, or am I making up nonsense?

[–] Hinori@scribe.disroot.org 2 points 3 weeks ago

I'd like it to be built entirely around a P2P network of easily self-hostable instances as a distributed storage/indexing backend, and a local desktop app as a suggested entry point, but that's not the case, unfortunately.

With its current architecture, I simply haven't managed to go trough the trial of finding a suitable instance! So if I'm watching PeerTube videos, it's only occasionally, when a link was posted elsewhere, as I'm lacking a proper entry point to continuously interact with it, let alone uploading.

[–] Hinori@scribe.disroot.org 1 points 5 months ago (1 children)

I'm not entirely sure, but I feel like you know something about nondestructive editing in Krita! Can I ask you some questions about it?

Is it possible to use the smart patch tool in a nondestructive manner? I'd like it to take samples from the underlying layer(s), but apply the result to the upper (empty) layer. Just like the clone brush can do. Is it possible to do it in this way?

[–] Hinori@scribe.disroot.org 1 points 6 months ago (1 children)

Hold on, I know how to make it really infuriating! Almost the same image as in the starting post, but instead "Linux™ 7" the title is "Linux™ 6". So putting everything together, it reads as "Linux 6 7"

[–] Hinori@scribe.disroot.org 2 points 6 months ago (1 children)

But out of curiosity... What exactly it looks like?

the OS provides simply the base software interface to your hardware, and does so in a separate layer, while your own software and data are another segment you don’t want to mix with the OS.

My own software? Sounds like a bunch of containers on top of a base system.

[–] Hinori@scribe.disroot.org 2 points 6 months ago (1 children)

So, using an immutable system, one could achieve... having one less backup target, apparently? Which is good, but seemingly not good enough for this particular case. ¯_(ツ)_/¯ As far as I understand, it could certainly save the day if I mess up its settings, but in case of FS corruption it effectively just adds a different kind of procedure that I'll have to execute for its full restoration? And splitting my system and data to a different partitions... well, this approach in itself have nothing to do with immutability, isn't it?

Meanwhile, I managed to recreate exactly the same system with all my data and stuff, down to ctimes, mtimes, hardlinks and reflinks, as if nothing happened at all. Just saying!

[–] Hinori@scribe.disroot.org 5 points 6 months ago* (last edited 6 months ago)

How do you know it was btrfs, are you living in my walls?!

Well, jokes aside, it was a power outage. In my defense, I've survived dozens of power outages while using btrfs with almost no negative consequences!

[–] Hinori@scribe.disroot.org 3 points 6 months ago (5 children)

I mean, it sounds both tempting and terrifying at the same time, but even assuming it's a good idea, how to reproduce my data, then? Not just the data per se, but all these volatile stuff in /var and $XDG_{CONFIG,DATA,STATE}_HOME as well. I like the idea of being able to (re)produce the entire system from a tiny "seed", but my system is still not the same without those parts, as it will behave slightly differently. So I still have to backup all my data, haven't I?

Also, using different partitions might be less fragile, but also much less convenient in terms of free space management, especially when you only have a single medium-sized SSD. So I just use a single rootfs (with subvolumes instead of partitions) for the system and all kind of data.

 

TL note: "keikaku" means "plan"

view more: next ›