social.noyu.me

Ah, there is a huge technical dream of ours (that a few of you may know of) — of the operating system variety — that we might have figured out how we can make happen without it killing us (we can't carry three worlds on our back, two is already too much).

Something we've been wanting to do for roughly… a year now? Maybe a bit longer than that? Something we have confided in a few people over (some in DMs, some IRL), but generally been quite quiet about. Something we have sought funding for and been rejected.

In fact that's probably the only way to get this to work, because as impressive as touchHLE was, it was a feasible one-person project; patching WINE, ReactOS, the Wayland server used by GNOME, packaging it all for Debian, and so on and so forth, that's a huge project…

…too large a project for a single person, regardless of how weird she might be, but individually, it's just a few weekend projects for a dozen or so FOSS maintainers and suddenly you have something half-usable…

We will not get a delusion of grandeur over this, we would literally just be an “ideas guy” if you will, but there's a … there's a way to make this work.

In at least two months' time.

When we're better-rested and in control over our life again…

Here is a hint about what we're talking about:

Did you know that already, today, without patching anything, with a few lines of Bash you can, on any modern Linux distro, get it to boot into an entirely-WINE environment that bears a striking resemblance to Windows 2000?

We are currently in no state to talk of the dream any more than this, but if this sounds interesting to you and you have skills/experience/connections in the relevant FOSS projects, please uhh… email us at hikari@noyu.me and mention “loss32” in your email as a keyword…

…and we'll try to get back to you and co-ordinate efforts a tiny bit in maybe a month or two when we're not, y'know, struggling to just exist.

Oh, and as a teaser, and as proof we're not delusional about this, please check out the many screenshots in this thread where we were quietly proof-of-concepting this:

https://x.com/hikari_no_yume/status/1794500608923455848

(We would really prefer to link the fediverse version of this thread obviously, but our instance's search is broken and it's from May last year, and we haven't been able to find it on Google or DuckDuckGo, so it'd take a long time to dig up alas!)

Someone kindly dug up the Fedi thread!! Check it here: https://social.noyu.me/@hikari/statuses/01HYRYZXMSDQHA2NPAFHS31Y9C

It seems like a trusted collaborator is planning to make their own post about this containing roughly the same info, but which will probably be far better written, so our Fediverse followers may want to hold off on retweeting/boosting for a bit, you'll know her when you see her!

Given the state of our mental health, I can promise you all one thing: we are not planning to be a “project lead” on this exactly; what we think is that we might be able to articulate a vision well enough, and do a tiny bit of co-ordination work, that we don't have to.

@hikari Well, more like Win32 on top of GNU, which is... wait, like Windows 2000 running Win32 on top of a NT kernel.

@anthk the GNU component is a lot less significant than the Linux component when WINE is involved, and you can probably outright remove all the GNU without it stopping working

@hikari Oh hey, I remember you talking about this a while back.

Sadly I lack both the win32 expertise and distro packaging knowledge (especially in the modern day's distro infrastructure) to be much help to the project (and am too perpetually exhausted in the foreseeable future to realistically contribute beyond maybe testing stuff?) ...

...but as I think I said before, it's a laudable goal and concept, one that's both kinda weird in it not already existing (AFAIK) but also incredibly understandable that it doesn't exist - despite being a "glue 2 things together" project, those still take a lot of time and effort, and there aren't enough motivators/resources for it to just spring up outta nowhere.

(e.g. for most folks Linux + Wine is sufficient, or is that plus some helpers to true and smoothen that experience like some of the more "Windows-user-friendly" distros that have come and gone tried to do - and ReactOS tries to be a full windows reimplementation rather than running on a unix-like base, which is also an admirable goal, but does leave the gap of something between the two)

I've also seen at least one example of a "wait, this didn't exist already?" project recently that was basically writing the glue between two things that already existed to make the obvious-missing functionality start to work - there was that project for running full X11 desktop environments in Wayland I saw on Fedi recently?

But yeah, it's a cool project, I hope it gets off the ground, agree that even without regarding recent circumstances, you not being the "project lead" but a contributer and someone getting the ball rolling sounds wise.

Will follow progress on this and keep an eye out for if there is anything I can do to contribute