← Back to context

Comment by mohamedkoubaa

20 hours ago

Principal engineer balks at bad UX when the PM should know better (it's their job)

2023: Ah well I guess we can't do it

2025: you're fired. Hey kid we hired two weeks ago, implement bad idea please

To be fair, it was already done by bad managers long before.

  • This is what I sees getting missed in a lot of LLM conversations. They're amplifiers. Full stop. If you have good practices they supercharge them, if you had bad practices, same thing.

    • Does our society generally have good, user-friendly practices?

      What will amplifying its output do?

  • I saw a trend of UX/UI designers coming with practice which I knew better were wrong. But they insisted. E.g hijack brosser native controls.

    Will never know whether they passed along some manager/PM commandements or were just incompetent.

    • > But they insisted. E.g hijack brosser native controls.

      [Rant-Example] The goshdarn ticketing-system hijacks alt-f, so that instead of opening the File menu of my browser, and instead toggles the favorite-status of whatever ticket I happen to be viewing.

      4 replies →

    • To be fair, the native browser controls have had too many quirks and features fox UX/UI consistency.

      Corporate needs their Brand™ look precisely as specified in their expensive Style Guide. IBM wouldn't want the Google vibes of Android Material Design TextFields, I imagine.

      Scratch beneath the visuals, and starker technical differences appear.

      Safari on iOS (used to?) has a 350ms debounce delay on every tap / click, in case you want to do a multitouch gesture.

      JavaScript (Frameworks) were the only way this arbitrary delay to user input could be reduced before 2015, when Apple finally released a native API for this.

      https://webkit.org/blog/5610/more-responsive-tapping-on-ios/

      1 reply →

Give me a break. This was already happening with Web 2.0 and things like "microservices".

  • I know it's just that we are in an accelerated state of Putt's Law

    • Had to look that up:

      Putt's Law: "Putt's Law: "Technology is dominated by two types of people, those who understand what they do not manage and those who manage what they do not understand."

      I suggest that this law does not give a complete description of what has been happening to software engineering in the past 2 decades:

      Putt's Law does not address the (new) phenomena we first saw when 'blog hotness' and minimal effort frameworks permitted practitioners with little practical experience or hard gained knowledge to manifest technical capability and assert technical authority. The minimal amount of 'wit' required was access to a smart phone, wiki, or some blog, and you had complete juniors arguing with seniors about architecture, frameworks. AI is taking that to the extreme.

      Putt's Law's relevance here is that prior to the past 2 decades of enabling tech and knowledge bases 'the clueless manager' had the metric of "older more senior more likely to be correct", and clueless juniors didn't have blogs or wikis or frameworks that required a handful of shell commands to install, and spinup a 'demo'. AI has made that even worse.