it is possible that NixWRT is doing a lot more compilation than it needs to (even given that there is no MIPS binary cache) because of I-didn't-understand-how-overlays-work when I wrote the overlay for it.

For example, where I wrote coreutils = super.coreutils.overrideAttrs (o: { doCheck = false } ) that changes the build system coreutils as well as the mips coreutils, so everything x86-64 that depends on coreutils is getting built from source instead of pulled from cache.

Leastways, that's how I now think it works. ICBW this time too

#nixpkgs #nixwrt

Daniel Barlow boosted

Wishing there were a standard data format for #activitypub services. If I want to change the software my mail server runs, I know I can get the data transferred in mbox (or maildir) format. If I want to change my gitweb presenter, the repos are there. To change from ActivityPub server program A to B, then unless A is Mastodon and B is Pleroma it appears I'm basically out of luck.

Someone tell me different? is abandoned, as far as I can tell (and no, I don't have time to maintain a fork)

Can't claim to have actually repaired anything for #InternationalRepairDay, but I have bondged a magnetic GoPro mount using epoxy resin and a magnet from an old hard disk drive, and I will later be fixing the broken bezel on my old thinkpad

Oh, but I did manage to make the believed-broken Sky remote work again. Etymologically I guess that counts as a re-pair

Think your Sky Q remote is broken? Try this before you buy a new one - 10SECOND.TECH>>

OK, it's admission time: I spent this week under the illusion that my Sky Q...


I have succeeded at , subtask "callbacks without random crashes", and the answer is "assign the callback to a local variable, otherwise the gc collects it". yes it's logical that the gc doesn't scan C address space looking for roots, I suppose, but it took an awful long time before I realised this was the problem


GitHub - q66/cffi-lua: A portable C FFI for Lua 5.1+

A portable C FFI for Lua 5.1+. Contribute to q66/cffi-lua development by...


added labwc support for "input inhibitor" (needed to make screen locking work) and fixed a bug with the window cycling, but now it's doing something extremely weird with blanking the portions of screen not covered by the active window (the wallpaper disappears and reappears when I move the mouse) and I can't for the life of me think what's causing it.

Reverting my changes hasn't fixed it either

Think I now have working support for primary selection (a.k.a middle-button click to paste) in labwc and am approaching the point where it doesn't have any really horrible gaps in as a works-for-me GNOME replacement.

A screen lock would be a good next thing to add before I next take my laptop out of the house

workaround Gtk primary selection bug by changing order of globals by telent · Pull Request #49 · johanmalm/labwc

This makes primary selections work at least with wayland-native Emacs (using...


Switched from gnome to labwc (wlroots-based wayland compositor) and kitty (terminal) and it has not been plain sailing: so far spent most of the evening trying to find out why the mouse pointer in kitty windows is teeny tiny. Changing the mouse pointer theme - apparently this is a thing - seems to have fixed it

XCURSOR_THEME=Adwaita XCURSOR_SIZE=32 - command line password manager

A small ruby script that works like "pass" but uses age instead of gpg. Written mostly because there seems to be no clear winner among the other previous attempts at doing similar, and partly because it was a nice simple exercise

I like to say WIBSNM ("was invented by someone not me") because I wasn't the first to think of "NIH"

GitHub - telent/moss: Maybe-OK Secrets Store

Maybe-OK Secrets Store. Contribute to telent/moss development by creating an...