Login
You're viewing the discuss.systems public feed.
  • Oct 14, 2026, 5:14 PM

    I've been saying since the 2000s that it's important that every big software stack have at least one angry new software grad to fight back against entrenched complexity in the existing codebase.

    sometimes there is a good reason why we can't "just do X, Y, or Z" but someone should be still raising the question, damn it

    💬 8🔄 18⭐ 0

Replies

  • Oct 14, 2026, 5:25 PM

    @foone

    Yes, asking the question - documenting the answer. I am consistently the 'crazy old guy' for asking where we documented our earlier decisions about a project.

    Sadly, little has changed in 40 years of me asking that same question.

    💬 0🔄 1⭐ 0
  • Oct 14, 2026, 5:35 PM

    @foone this is one of the first things I tell new co-op students. Forcing me to explain why you’re wrong is valuable to both of us, and sometimes you’re right! Or at least less wrong than it looks at first glance.

    💬 0🔄 2⭐ 0
  • Oct 14, 2026, 5:46 PM

    @foone this is a reason I love having paid interns at my employer. They don't know what isn't possible, ask questions of our assumptions and Chesterton's Fences, and generally force us to reconsider decisions - sometimes it's "yeah, sure, we can change that", others it's "no, it's that way for good reasons that still hold". But it forces us to reevaluate our work

    💬 0🔄 4⭐ 0
  • Oct 14, 2026, 6:04 PM

    @foone i love it when i work with a junior and they question my reasoning

    and i love it even more when i can’t provide a good answer, because then we’ll both be learning something new and cool

    💬 0🔄 1⭐ 6
  • Oct 14, 2026, 6:27 PM

    @foone Software is the only discipline where the person asking "why don't we just" probably has a point. So many people in this field are too good at coding to be a good software developer.

    💬 1🔄 0⭐ 0
  • Catcatmisgivings@stranger.social
    Oct 14, 2026, 9:01 PM

    @Nentuaby @foone manufacturing also! My favorite is when operators on the shop floor bring their ideas to the manufacturing engineer

    💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • Oct 14, 2026, 8:42 PM

    @foone @daedalus the act of having to explain “here’s why we do it that way” is the path to realizing “oh wait we don’t have to do it that way anymore” or “wow we never really had a good reason in the first place”

    you gotta question how things are done, and that battle never ends.

    💬 1🔄 3⭐ 0
  • Oct 14, 2026, 9:10 PM

    @darkuncle @foone sometimes people get quite defensive about the status quo because of ego threat, and no agreed way to determine what "good" or "better" even is beyond feelpinions.

    💬 2🔄 0⭐ 0
  • Oct 14, 2026, 10:26 PM

    @daedalus @foone when your identity is wrapped up in what you do, or in being without error (see current admin), it becomes impossible to question decisions without devaluing the person.

    Which is part of the reason I tell especially early career folks:

    what you do != who you are

    💬 0🔄 0⭐ 0
  • 💬 0🔄 1⭐ 0
  • Oct 14, 2026, 10:13 PM

    @foone And, if you've really got the system documented, each of those questions can be answered quickly by seniors just linking to the proper sections of the docs.

    Very few systems are actually well-documented, even when the API doc tool says 100% coverage.

    💬 0🔄 0⭐ 0
  • Oct 14, 2026, 10:38 PM

    @foone Not just software, this applies to pretty well everything else that groups of people do.

    The answer to "why don't we just do X" is usually one of

    (1) a good answer
    (2) something that was a good answer at the time, but the world has changed since then and that problem has gone away
    (3) nobody knows.

    So, if it's (1) there's no need to change anything.

    It's because (2) pretty well always happens eventually that how we do things is (or at least should be) reviewed from time to time.

    And it's to avoid (3) that it's a good idea to ensure you've always got a corporate memory. Which means, for example, not sacking all the old people.

    💬 0🔄 0⭐ 0