Chris Siebenmann @cks@mastodon.social

Today I accidentally confirmed that my home desktop still locks up if it's too cold. 'Too cold' seems to be around 60F (at floor level).

(What can I say, I was taking the chance to thoroughly ventilate my work area for once.)

Toronto is gusty and windy right now, which is best appreciated while working from home so I don't have to bike in it. Especially since we may get snow squalls later.

It's the start of March so it's a great day to flush my inbox. How did it get so big? Well, one reason is that I keep using it to hold read-but-not-fully-handled things and conversations.

(I read all my incoming email right away, I'm a sysadmin.)

Bless my past self who (almost) five years ago did a decently detailed write-up for our group on how to test a new version of this very important core machine.

I feel like I should have opinions on (mechanical) keyboards, but I lack the passion to explore the space thoroughly. My current one with Cherry MX Blues feels nice and right (and sounds right). Maybe another keyboard with different switches would be an incremental improvement, but I don't know and it feels like a lot of work.

(Part of me thinks I would like a slightly smaller layout, like my old BTC-5100Cs had, while another part thinks the extra keys and so on are handy.)

This is my sad face when people run web servers that don't ping even when the web server is up.

(I almost wrote an email that would have been a little embarrassing before I thought to check the actual URL, not just ping the host. Maybe I should get a 'httpping' program.)

One thing that's different this work-from-home winter is that my thumb isn't getting chapped up, so it still reliably unlocks all of my fingerprint-based things.

(Yes my things are still fingerprint based, not facial recognition. I'm slow to buy new shiny things.)

I made some quality of life improvements to my (ancient) (N)MH and exmh setup today, so now I can automatically quote messages when I reply to them even if they're all MIME'd up. (It's rather complicated in NMH.)

(There are so many old bits in my NMH and exmh configuration. So, so many.)

Given that Ontario's current official expectation for when broad covid-19 vaccination will start is August, I now think we're going to be working from home until next fall at the earliest.

This almost certainly changes yet more plans for upgrades and deployments. In a nice world it would change hardware purchase plans too, but we don't have that sort of money.

Current status: Wow, that sure is a snowstorm. Sometimes working from home isn't so bad.

(Sometime I will get studded bike tires, but obviously not this year.)

It probably doesn't do any harm to restart my Firefox every so often, but I do wish I wouldn't do it by hitting Ctrl-Q instead of Ctrl-W by accident.

(No big harm done because I have Firefox set to remember all my windows/tabs and recreate them on restart, plus I have a script to reposition all the windows in the right place.)

My modern process for spelling corrections: stick the probably misspelled word into Google and pick the 'did you mean' spelling. (Usually.)

Blog post: SMART's 'threshold' numbers turn out not to be useful for us in practice https://utcc.utoronto.ca/~cks/space/blog/tech/SMARTThresholdNotUseful

As usual with SMART it's a data quality issue. Too many drives do too many weird things with their normalized SMART values, both the 'current value' and the 'threshold value for problems' ones.

One nice thing about having a blog is that I have some use for negative results. I can recycle them into blog posts, instead of at most writing an email to co-workers saying 'I looked at this and it's useless for us'.

(Today's negative results are about potentially using SMART vendor-supplied 'threshold' information to drive alerts.)

Every time I use (u)rxvt for very long I'm reminded that I'm extremely used to xterm's very specific selection behavior (especially for multiple clicks) and I can feel something is off even for small variations from it.

In April and May, it was possible to believe that people in Toronto were scared. Now, I think we're just numb and going to go through the motions. Maybe that will be enough.

(Intellectually, I know that people can only stay scared and adrenalized for so long. It's another thing to see and feel it in applied form.)