There are a couple I have in mind. Like many techies, I am a huge fan of RSS for content distribution and XMPP for federated communication.

The really niche one I like is S-expressions as a data format and configuration in place of json, yaml, toml, etc.

I am a big fan of Plaintext formats, although I wish markdown had a few more features like tables.

  • pHr34kY@lemmy.world
    link
    fedilink
    arrow-up
    17
    ·
    edit-2
    3 months ago

    I love this standard. If you dig deeper into it, the standard also covers a way to express intervals and periods. E.g. “P1Y2M10DT2H30M” represents one year, 2 months, 10 days, 2 hours and 30 mins.

    I recall once using the standard when writing a cron-style scheduler.

    I also like the POSIX “seconds since 1970” standard, but I feel that should only be used in RAM when performing operations (time differences in timers etc.). It irks me when it’s used for serialising to text/JSON/XML/CSV.

    Also: Does Excel recognise a full ISO8601 timestamp yet?

    • Jim
      link
      fedilink
      English
      arrow-up
      4
      arrow-down
      1
      ·
      3 months ago

      I also like the POSIX “seconds since 1970” standard, but I feel that should only be used in RAM when performing operations (time differences in timers etc.). It irks me when it’s used for serialising to text/JSON/XML/CSV.

      I’ve seen bugs where programmers tried to represent date in epoch time in seconds or milliseconds in json. So something like “pay date” would be presented by a timestamp, and would get off-by-one errors because whatever time library the programmer was using would do time zone conversions on a timestamp then truncate the date portion.

      If the programmer used ISO 8601 style formatting, I don’t think they would have included the timepart and the bug could have been avoided.

      Use dates when you need dates and timestamps when you need timestamps!

      • cout970
        link
        fedilink
        arrow-up
        7
        ·
        3 months ago

        Thats an issue with the time library, not with timestamps. Actually timestamps are always in UTC, you need to do the conversion to your local time when displaying the value. There should be no possible off-by-one errors, unless you are doing something really wrong.