Russell Coker

Bryan, what's your opinion re: the stability of the LLNL port of ZFS on Linux? The 0.6.0 release is apparently just around the corner and several high-profile people in the community, Russell Coker for one [1], have started using this on production systems. I'd love to make the move to ZFS as well but am still afraid of data loss. I started using XFS on Linux in 2002 and my gut feeling is that ZFS on Linux is now at or around a similar point of maturity. If you could share your opinion I'd be really grateful. Thanks so much. [1] http://etbe.coker.com.au/2012/07/31/zfs-debian-wheezy/

Many good points, and I largely agree. But for developers and designers working on user-facing software: I recommend keeping at least one poor to profoundly shitty monitor in your setup. Both resolution and color rendering accuracy matter here. If you have everyone on fancy MacBooks and excellent high end monitors, you will tend to miss defects in your design that will affect your customers - things like poor color contrast (I don’t think the accessibility tools care about contrast of non-text elements), elements that are too thin to be seen comfortably or lead to weird aliasing, etc. Microsoft famously made this mistake with their Fluent UI design (or some version of it); 1px borders with huge corner radii that ended up looking “cut off” on standard corporate monitors (like the ones 99.999% of their target audience are subjected to). P.S. to have some empathy for the corporate IT department, if they are handing out Dell laptops to thousands of people, up until relatively recently they very likely would struggle to handle two or sometimes even one 4K monitor. And Windows high DPI handling is a potentially huge support burden if there is any sort of diversity in the software in use. Edit: turns out the author themselves wrote about how painful good/4K monitors could be to use. Apparently it was essentially unusable for them in late 2019, although that doesn’t match my personal experience - I was using 4K monitors in windows with only occasional issues as early as 2018.…

When Google was struggling with various people being locked out of email and not being able to get back in (without an alternate channel its really hard to verify you are who you are without a bad actor in the loop) I proposed a project for creating an identity token. I was playing with the Yubikey [1] as a possible approach. I certainly think something like that has the basic concepts, physical possession of the device which can be queried to provide strong identity verification. It was amusing (in a greek tragedy sort of way) to watch the conversation get de-railed by reductio ad absurdum arguments about perfect security. When in fact it would benefit over 80% and perhaps as many as 95% of the Internet users to have something like this. Still have the business plan. Its a bit capital intensive since it functions best when everyone has it, but the basics were build the base system in a "open" documented kind of way (so that folks could verify for themselves it wasn't snake oil), create a licenseable fob capability that is compatible with a variety of communication mechanisms (this was pretty straight forward). Have a 'key' implementation cost less than $5/key in quantity going to less then $1/key (Difficult but doable I believe) and three key partnerships. The 'competitors' are soft systems which are implemented on handheld devices or other token schemes. Russell Coker wrote up a good summary set [2]. The 'secret' sauce of this effort, or perhaps the key learning that…

It's almost like they're slowly increasing the friction for other packaging formats (flatpak, deb) other than snaps. Bubblewrap, the sandboxing tool for Flatpak, is broken out of the box on 24.04: https://etbe.coker.com.au/2024/04/24/ubuntu-24-04-bubblewrap/

Russell Coker's reports of his experiences with BTRFS give me the screaming heebie-jeebies, no matter how up-beat and positive he stays about it: http://etbe.coker.com.au/tag/btrfs/

> Can you insert [X] into a running pipeline? Yes [0]. It needs a bit of tooling to be more user friendly, but even the gdb stuff is pretty self-explanatory. [0]: http://etbe.coker.com.au/2008/02/27/redirecting-output-from-a-running-…