• queermunist she/her@lemmy.ml
    link
    fedilink
    English
    arrow-up
    4
    ·
    7 months ago

    Logic guys love assigning random values to things based on gut feelings. “Everything is 5x as hard to do at scale” means absolutely nothing.

    • HumanBehaviorByBjork [any, undecided]@hexbear.net
      link
      fedilink
      English
      arrow-up
      2
      ·
      7 months ago

      Big Yud is literally the founder of an assigning-random-values-based-on-gut-feelings religion. His essay on Bayesian reasoning has given perhaps thousands of blog-reading nerds irreversible brain damage.

    • Findom_DeLuise [she/her, they/them]@hexbear.net
      link
      fedilink
      English
      arrow-up
      2
      ·
      edit-2
      7 months ago

      My boss does this for “estimating” software project schedules. He built a goddamned spreadsheet* where he will rate the entire project on a scale of 1 to 5, with 1 being trivial/quick-win territory, and 5 being extremely labor-intensive.

      Two problems with this approach as used at my job:

      1. He assigns the ratings before requirements gathering has even started (if they ever get documented in the first place).
      2. He bases the final deadline around the calculator spreadsheet, and sends that date on to the business partners/project stakeholders within the company, and they usually pass it along to upper management.

      So, by the time we finally get requirements together and find out, oh, shit, this is actually way more complicated than a 2.71828 or whatever, the stakeholders have already told the Senior VPs of Thought Leadering that my team will be done by a specific date. The week before that date rolls around, boss goes into a panic, demands that I work on absolutely nothing else as I’m being pinged daily to put out random bullshit fires on other projects that were rushed through implementation before I even worked here. Between that and the low pay, I start really strongly considering pulling a no-show. I stay up late a couple of nights, project gets finished. Rinse. Repeat.

      I envy the dead.


      *: No, it’s not a Monte Carlo simulation or anything that fancy – he just multiplies the complexity rating by a set number of labor hours, and doesn’t bake in additional time for risk mitigation. They promoted his ass because this is so scientific and data-driven. Edit: and no, there isn’t a more detailed breakdown/implementation milestone schedule somewhere further down in the estimate. It’s literally “I feel like this is a… 2. You have a week. GIT 'ER DUN!”

      • wheresmysurplusvalue [comrade/them]@hexbear.net
        link
        fedilink
        English
        arrow-up
        1
        ·
        7 months ago

        My boss gives me the opposite. He asked me to give a work estimate on a year-long project, he added +25% buffer for unknowns and submitted it. When the work ended, we were so efficient that we only used 70% of the estimated budget, and this was a problem! Buddy, that’s why they’re called estimates, we can’t make perfect guesses before requirements are gathered.

    • lugal@lemmy.ml
      link
      fedilink
      English
      arrow-up
      1
      ·
      7 months ago

      Yes, I agree. I think it’s only 3x, 4x at best according to my gut feeling