• 3 Posts
  • 23 Comments
Joined 1 year ago
cake
Cake day: June 13th, 2023

help-circle



  • Here’s my take:

    1. We’re built for about 150 relationships max (Dunbar number), and yet we benefit from cooperation above that threshold. Rather than make it so we have to have a personal relationship with everyone who could possibly benefit us, we accepted a ramped down version of relationship we call “transactions”. This is a very weak replacement for a relationship, but it is a sort of “micro-relationship” in that for a brief moment two people who don’t know each other can kind of care about each other during an exchange. Through specialization, we can do something well that doesn’t just benefit the handful of friends and neighbors we have, but tens of thousands and possibly millions of people via transactions (e.g. a factory, starting an Amazon business, etc.)

    2. There is a process called “commensuration” in the social sciences, where people start to make one thing commensurate with another, even in wildly different domains. For example, to understand the value of a forest and to convey its importance to decision makers we might say “this forest is worth $100 billion”. It’s kind of weird to do this (how do leaves and trees and anthills and beetles equal imaginary humoney?) But slowly, over time, we have made many things commensurate to dollars at various scales. (I don’t think this is a good thing, but it does have benefits). In short, more and more things that were part of an implicit economy of relationships (e.g. can the neighbor girl babysit tonight?) have entered the explicit domain of the monetary economy (e.g. sittercity).

    .

    IMO, in order to participate in the huge value generated by this monetary economy, people sometimes lose the forest for the trees (so to speak) and forget what really matters (e.g. excellence of character, deep relationships, new experiences, etc.) because it seems like we might be able to put off those things until “after” we square away this whole money thing first. But maybe “after” never comes–and the hollow life of a consumer capitalist drains the inner ecological diversity of a soulful life.









  • This is the case for me as well. I tried NixOS this weekend, and even though it has more adoption than Guix, it still does not have 100% coverage of all software I wanted. That said, the packages I did install were pretty up-to-date. I guess NixOS is as close to “critical mass” as we’ve got when it comes to this type of OS. But if I were a wizard devops type person with more time, I’d probably enjoy Guix more.


  • Given encouragement to try tmux, here is what I’ve come up with as a “one-liner” (script) that does what I was originally looking for:

    #!/bin/sh
    
    tmux new-session -d -s split_screen_grep \; \
      send-keys "/bin/sh -c '$1' | tee /tmp/split_screen_grep.txt" C-m \; \
      split-window -h \; \
      select-pane -t 1 \; \
      send-keys "tail -f /tmp/split_screen_grep.txt | grep '$2'" C-m \;
    
    tmux attach-session -t split_screen_grep
    

    I use it as follows, first arg is a command, second arg is a pattern to search for:

    $ ./split-grep "cat big_file.txt" "tmux"
    









  • The short answer is “yes, but only as much as it needs to”. Flatpak had to make a decision between “do we guarantee the app will work, even with system upgrades” or “do we minimize space” and they chose the former. The minimum necessary dependencies will be installed (and shared) amongst flatpaks.

    Have you had the unfortunate experience of a utility or program losing its packaged status? It’s happened to me before–for example fslint. I don’t think this can happen with flatpak.


  • It’s funny, I do almost the exact opposite–whenever there is a flatpak version, I prefer it over a built-in apt package. The flatpak is almost always more up-to-date and often has the features and bug fixes I need.

    Examples:

    • Vorta (0.8.12 flatpak; 0.8.3 apt)
    • Pinta (2.1.1 flatpak; 1.6 apt)
    • Minder (1.15.6 flatpak; 1.13.1 apt)
    • Xournal++ (1.2.1 flatpak; 1.1.1 apt)

    .

    I don’t think it’s fair to expect the distro maintainers to be up to date with every software out there–the universe of software has grown and grown, and we just can’t expect them to wrap/manage/test every new release and version bump.