good morning/afternoon

good night (I should have gone to bed hours ago)

How do you say “I love you” in your native language? Boosts appreciated

@gavi "I love you" (english) or "Я тебя люблю" (russian)

@kemona_halftau nini!

meow

morning

similarly, the existence of "edward" implies the existence of "viward" and "emacsward"

the existence of "awkward" implies the existence of "sedward" and "grepward"

nini fedi

Nicolium's interaction circle thing doesn't work for me for some reason, it just gets stuck on "Fetching avatars"

@lumi it depends. standardization is good sometimes. please do not make yet another incompatible and legally questionable license just because you didn't write any of the existing ones, for example. also, standardization on USB-C is mostly a good thing and ISO 8601 is a good date format (please don't start inventing your own and using YDYD-MM-YY or something stupid like that)

for communication protocols standardization is also a good thing, the fediverse would not work if every implementation had created its own protocol instead of using activitypub

@thing @17lifers @ivy all the information I can find on that says it's amd, but you said it was intel? what cpu does it have?

@thing also there's autoport, but iirc that only supports up to haswell

@thing For porting coreboot, it’s not very well documented and much information is stored only in the heads of coreboot developers, but you should read these links:

• https://aprilthe.pink/coreboot.html (probably the best one currently, contains info from the 38c3 talk)
• https://media.ccc.de/v/38c3-corebooting-intel-based-systems (video version of the above)
• https://review.coreboot.org/c/coreboot/+/85912
• https://web.archive.org/web/20230331154746/https://www.coreboot.org/Motherboard_Porting_Guide

Basically you should obtain some dumps (preferably with all SATA/PCIe ports filled), find a similar already-supported board, and try to adapt it using information from the dumps you obtained.

But first, you should make sure that your computer doesn’t have bootguard (the first guide I linked explains how): anything ivy bridge (intel 3rd gen) or older won’t have bootguard, haswell (4th gen) won’t have bootguard if you have a socketed CPU, and anything with MEv11 (skylake/kaby lake/kaby lake refresh (6th gen, 7th gen, and some mobile 8th gen)) might have bootguard but it’s broken and you can just use deguard to bypass it.

The only board I have ported (optiplex 5050 sff (I should finish that at some point)) was relatively similar to an already-supported board (optiplex 3050 micro) so it was relatively easy. If it was more different (and had an unsupported SoC or SuperIO or something) I would probably have no idea how to port it.

For AMD, unless it’s quite recent and supported by OpenSIL (which supports Genoa, Turin, and Phoenix), or a chromebook, it will be much more difficult to do than for Intel. The only supported AMD boards are chromebooks, some quite-recent OpenSIL-supported stuff, and some old pre-zen stuff (fam15h and possibly others) that coreboot dropped support for years ago but support is still maintained in the coreboot-15h fork. I don’t know as much about porting AMD boards as I do about Intel.

Adding a board to libreboot requires that you port coreboot first but otherwise is much easier. You just need to add a few configs basically. You can see my supermicro x11ssh-f/ln4f port as an example (and also the lbmk maintenance manual).

@kemona_halftau nini!

@ielenia I use this, but it only supports one specific logitech mouse (G203)

@freeplay I don't really see the point of the EUPL. It's a strong copyleft license (and attempts to close the SaaSS loophole similarly to the AGPL) but at the same time explicitly provides for compatibility with several other licenses, including those with weaker copyleft (like the Eclipse Public License and the GPL), which makes the stronger copyleft provisions pointless.

I would probably use AGPLv3-or-later.

@niko my server's SSDs are about this fast to scrub (3.62GiB/s), but my other laptop's NVMe SSD is a lot slower for some reason (1.33GiB/s) (they're both LUKS-encrypted, and the server is RAID1 which shouldn't be directly relevant to scrub speed since it checks both copies but does mean that it has to encrypt/decrypt everything twice since the keys are different for both SSDs)

@kemona_halftau there is, network.http.sendRefererHeader (but disabling it will break stuff)