@sun@shitposter.world
Still can't get a printer to work on arch, same printer worked OOTB on mint.
I'm so annoyed.

@mangeurdenuage @idiot @Zergling_man @lanodan @sun You also need a lot of money especially if you want your hardware to be affordable, and much, much better governance than the FSF for example.

For profit companies have an implicit advantage there, if they completely screw up they stop making money. Whereas Gnu Hurd is still on its road to nowhere thirty four years after it formally started.

But with money also comes temptation, see the Raspberry Pi for example (switched over to prioritizing companies that embed their stuff, is now doing an IPO).

I can't think of any high volume open hardware projects that either avoided that fate (of course Raspberry Pi actually wasn't very open, but that's in part on Broadcom etc. and the latter's making too many of a certain chip) or got anywhere, see for example lowRISC which was founded to make a very interesting RISC-V chip, as in they talked about what it would cost to make N-thousand working ones, and ended up being a consulting company.

See also how much it costs up front to make good injection molded plastic parts for packaging, that's generally needed for high volume parts for consumers. Every time you see something that's metal, CNC machined, and/or 3D printed you know it's going to be extra expensive.

@idiot @Zergling_man @lanodan @mangeurdenuage @sun Cross our fingers on the FTC non to non-competes, but its prospects aren't generically good since its head is an activist who doesn't care if what she orders is legal, and is being regularly slapped down by the courts.

Would also severely screw over California for what that's worth....

Speaking as someone who's worked for a few hardware companies back in the days, and did some fiddling around with electronics before college (OK, got some mad soldering skills out of that) and my family was also in the business, I think much of the rest of what you say is wrong including on salaries, but I don't have real knowledge of the latter today.

So I see "sharing schematics, drawings, and BOMs, and assembling the computer/printer/whatever together your own bad self...."

And I also see "and then a miracle occurs" where a bunch of surface mount parts get placed on a (multilayer??!?!!!???) PCB and soldered into place.

See also connectors, cables, and all the fun that comes with them and how they too can fail. I see people tearing their hair out trying to debug all the things which can go wrong.

You're also going to either go crazy trying to buy the parts in tiny or normal quantities which will be insanely expensive, or now you're back to a company buying them in normalish quantities and parceling them out into kits, along with PCBs that I hope they test.

And not entirely cheaply because they'll have to do something about customer support for those debugging issues, which can also include replacing damaged stuff.

Software is massively more tractable.

@romin @idiot @Zergling_man @lanodan @mangeurdenuage @sun Yeah, RISC-V is totally a Worse is Better, New Jersey macroarchitecture, they even go so far as to say "well, C and Java don't care about this..." in the official architecture document not long after they launched.

But you're already going to have to compromise a lot to have open hardware, so adding instructions to check for integer overflow etc. probably isn't a huge thing if you can get a fast enough part to begin with.

As to your two previous comments, we're talking about the 1% who for example already run a Linux/BSD desktop. And we have to start somewhere, there's always going to be a supply chain, software or hardware.

@mint @PurpCat @lanodan @Zergling_man @sun my refurbished laserjet works fine as well. complains about almost empty toner for half a year now, still prints. scanner works too, only the document feeder is kinda broken and jams often.