Richard Hughes

What do you want to see in the fwupdmgr CLI when there's a long (5 minutes+) delay? Showing a progress bar means it's "stuck" on a specific character block for some time, and it'd would be easy to think that the process has hung. A spinner that goes back and forth with some kind of time estimation? Anything better?

I've been (ab?)using Claude to find bugs in my software, and although it's found quite a few false positives, it's also found a lot of actual real bugs. Nothing CVE-worthy, but still important to fix. Releases include:

* passim: https://github.com/hughsie/passim/releases/tag/0.1.11
* libxmlb: https://github.com/hughsie/libxmlb/releases/tag/0.3.26
* libjcat: https://github.com/hughsie/libjcat/releases/tag/0.2.6

#fwupd will follow, when all ^^^ has hit the various build-roots. If anyone wants to polish the translations in the meantime, it's the normal place: https://hosted.weblate.org/projects/fwupd/fwupd/

I'm pondering some interesting talks I could suggest for #OSFC this year. Does anyone want me to talk about anything specific for firmware, #fwupd or LVFS topics?

I'm thinking doing a pretty aggressive FU aiming at some of the major hardware vendors, but I'm not sure my boss would approve. Other ideas welcome.

Is everyone getting horseshit responses from #claude today? Queries that worked fine on Friday generate nonsense today.

How is there no standard MiB constant of 1024*1024 in the C standard? What do other codebases do?

Announcing #fwupd 2.1.1 -- a very big release with the usual bugfixes, new features and lots and lots of new hardware support: https://github.com/fwupd/fwupd/releases/tag/2.1.1

I'm also happy to announce we've got a new sponsor for the LVFS: #Framework

Although there are about a half a dozen OEMs that have promised to sponsor LVFS, Framework is the first to have actually signed the paperwork. You can see the new logo on https://fwupd.org/ and also the new "sponsorship pill" on various vendor pages like https://fwupd.org/lvfs/vendors/#framework