Admiral Patrick

I’m surprisingly level-headed for being a walking knot of anxiety.

Ask me anything.

Special skills include: Knowing all the “na na na nah nah nah na” parts of the Three’s Company theme.

I also develop Tesseract UI for Lemmy/Sublinks

Avatar by @SatyrSack@feddit.org

  • 155 Posts
  • 2.16K Comments
Joined 2 years ago
cake
Cake day: June 6th, 2023

help-circle



  • One particular spite house in Boston: https://en.wikipedia.org/wiki/Skinny_House_(Boston)#History

    According to local legend, the structure was built as a “spite house” shortly after the Civil War:

    … two brothers inherited land from their deceased father. While one brother was away serving in the military, the other built a large home, leaving the soldier only a shred of property that he felt certain was too tiny to build on. When the soldier returned, he found his inheritance depleted and built the narrow house to spite his brother by blocking the sunlight and ruining his view.

    Another source states:

    Not much is known about the city’s narrowest house. Legend has it that … its unnamed builder erected it to shut off air and light from the home of a hostile neighbor (also nameless) with whom he had a dispute. … Believed to have been built after 1874







  • Not sure about Android, but on iOS, when one scans a QR code it shows the web address on the screen that the user then taps on. For the average user, I doubt that they are going to question what the URL is before following through to the website.

    Android does the same. The problem is most of those QR codes are encoded short links which tells you nothing about where they’re taking you.

    https://short.link/au1034gha could take you to a PDF on the restaurant’s Wordpress site or it could take you to malware or somewhere else you really don’t want to go.

    In that case, I blame the people generating the codes for using URL shorteners. My org uses them in flyers for the public, and I always have to chastise them and re-create the QR codes because they run the URL to our website through bit [dot] ly. 😡





  • I suspect zram’s swap device only consumes RAM when it actually contains swapped pages, but I don’t know for sure. Can anyone link an authoritative statement on this?

    I’m wondering the same. I haven’t read anything authoritative, but it definitely seems like it only consumes the RAM it’s using. Otherwise, it wouldn’t be able to create block devices that exceed the physical memory. I started to wonder when I had it set to 50% (4 GB) and gave it a stress test. The 4 GB it allocated filled up but was compressed to just about 1 GB, so I thought “surely this isn’t wasting 3 GB of RAM to hold 1 GB of pages.”

    The guidelines I’ve read seem like there’s some guesswork involved in the planning. Basically you can make the zram device as large as you want so long as the compressed data is less than the physical RAM (not all pages compress equally as you mentioned).

    I’ve since bumped it to 200% of system ram (16 GB), and I think that’s probably good enough for my use cases. I’m seeing about a 4:1 average compress ratio, so I could go higher, but 8 GB has been plenty usable up until now. :shrug: I also left the original swap file in place with a lower priority as a spillover (I’m not really missing the 4 GB of disk space that uses, so might as well keep it).