3 ms·

A bigger problem is that there are fewer and fewer living engineers that actually understand things well enough from first principles to be able to reconstruct

• hilbert42 6y ago
  "...there are fewer and fewer living engineers that actually understand things well enough from first principles to be able to reconstruct a technological society" I think that's either true or partially true due to (vertical) specialization that's so prevalent these days and also for other reasons some of which are cultural. You qualify your comment by saying that 'there are fewer and fewer living engineers'. Here, your inclusion of the word 'living' is interesting. By that, you are not only implying that once engineers could do so (and their modern day counterparts cannot) but also that modern engineers aren't taught first principles—at least not to the extent that they once were. This is a bold statement, why do you think this true?
  • thefz 6y ago
    > By that, you are not only implying that once engineers could do so (and their modern day counterparts cannot) but also that modern engineers aren't taught first principles—at least not to the extent that they once were. I think what he means is that complexity today has scaled so much that a new graduate would have a hard time to be able to do and remember and understand everything. We are so vertically specialized that I doubt an electrical engineer could, as an example, build a basic computer fast enough and use it to code a functioning basic web server from scratch. At the start of 1900, not so much.
    • hilbert42 6y ago
      Yeah, right. I was trying to draw him out on that point because I reckon it's an important one that we often brush under the carpet (as there's no easy/practical solution). It seems to me that the sheer volume of information today poses special problems for training and I reckon we've not yet found the ideal structure to educate people in ways so that they are taught a hierarchy of information that decreases in relevance as it moves out and away from the core material. The problem is that every teacher (and university syllabus) will have a different view about what is relevant and what is not. Essentially, we don't have a theory about what's relevant in other fields of endeavor and how to connect or relate that information to the core material let alone implement teaching it in practice. I've been around a while now so I've reasonable experience in my work—and I try to keep abreast of my subjects—but still I often shock myself when I come across information that's close (or relevant) to the core material that I ought to damn-well know but I do not. Somehow, I've missed out learning it. What's even more troubling is that after coming to grips with this 'new' material my view (and understanding) of the core material either alters and or is improved (methinks shades of the Dunning-Kruger effect have been at work here—and that's a real worry). Precious few of us have the intellect, perfect recall and ability to drill down and instinctively search far and wide across other fields then vacuum up every skerrick of information about a subject that's relevant—which the amazing John von Neumann was capable of doing effortlessly—so I don't have a sensible suggestion about what we ought to do to solve the problem. Perhaps eventually AI will come to the rescue. Let me give you an illustration of this. A few years back I was bemoaning the fact to the head of a university physics department that very few people who I was employing to work in my department had a broad cross-section of general knowledge of relevant/related subjects. He immediately recognized the problem and retorted with the following example. During his evaluation for a PhD, a candidate whose study was on LEDs (Light Emitting Diodes) was asked a general question† about how a junction transistor worked and he was completely flummoxed; essentially, he didn't have a clue! Frankly, this revelation amazed me; I was almost speechless. I claim no special knowledge of LEDs and junction transistors over that which most others in my field would know but do I know enough semiconductor theory to know that he should have been able to field a reasonable answer. I simply do not understand how he managed to get though four years of undergraduate physics and electrical engineering (electronics theory) let alone his PhD research and not have some knowledge of the matter. The fact that at a well established highly respected university that he even managed to get to the stage where he was being evaluated for a PhD amplifies the fact that we've a serious problem with educating specialists. Tragically, that facetious saying that 'these days, PhDs know more and more about less and less', may have some ring of truth to it. (Yes, I know, there are many exceptions and there are many brilliant people still out there—much cleverer than I am—but you get my gist.) † I recall a similar instance with a university exam paper where third year physics students were asked a very short 'reality-check' question: 'Derive F=ma' and a significant number either gave the wrong answer or omitted answering the question altogether. Clearly, teaching students to see both the woods and the trees remains a significant problem.