4 ms·
> In twenty years of engineering practice I have never once had to take a difficult limit, and can probably count on one hand the number of times I've had to ta
by roenxi 4y ago
> In twenty years of engineering practice I have never once had to take a difficult limit, and can probably count on one hand the number of times I've had to take a limit of any kind.
In principle, I agree that calculus doesn't need to be mandatory. However this apparently respectable counterargument is misframing the situation. A bit like me saying that a given soldier never placed top 3 in a race, or indeed had to run anywhere in a combat situation, and therefore efforts at army fitness training were useless.
Everyone knows that nearly no engineers use calculus in their day job. Therefore, the purpose and measure of success isn't whether an engineer uses calculus in their day job. It is a great signal of technical problem solving ability, it is really powerful having a large pool of people who are trained in working with rates, areas and margins.
Plus, don't underestimate the impact of forcing people to acknowledge that there is exactly one right answer and that it can be proven. There are people out there who genuinely believe that everything is up for negotiation to a limit that just can't fly in a math course and something needs to screen them out before they get anywhere near anything involving high pressures. People who don't have engineers for parents need to be inducted into the culture from a couple of angles.
- zozbot234 4y ago> Plus, don't underestimate the impact of forcing people to acknowledge that there is exactly one right answer and that it can be proven. There's plenty of classes in a civil engineering degree to instill that particular lesson. Why does it have to be calculus? It's just legacy - it used to be that limits and integrals could only be computed by hand, so any aspiring engineer would have to practice that skill a lot. Now computers can do the tricky algebra on their own (and I'm not even mentioning numerical approaches, which are more broadly useful but can be very tricky in their own way), yet the calculus courses have barely changed.
- roenxi 4y ago> Why does it have to be calculus? It doesn't, but it has to be something. So the proposal isn't 'get rid of calculus' as much as 'replace it with X which will be better at achieving the outcomes we think are important'. Like the original article did with statistics, which is a decent proposal. That being said if my civil engineer hadn't ever studied calculus I, personally, would prefer to find a different civil engineer. It makes sense to give the will-it-stay-up? job to someone unusually well educated and if they haven't studied those sort of mathematical basics it is pretty suspicious.
- zozbot234 4y agoThe mathematical foundations of "will it stay up" questions are a lot more challenging than what you'll find in any calculus course. I'm all for teaching less calculus, and more theory of structural engineering. (Setting statistics aside, which does require calculus but not of the pointless-algebra-tricks kind.)
- aoki 4y ago> statistics aside, which does require calculus but not of the pointless-algebra-tricks kind https://en.wikipedia.org/wiki/Gaussian_integral https://en.wikipedia.org/wiki/Gaussian_integral
- GlenTheMachine 4y ago"Everyone knows that nearly no engineers use calculus in their day job. Therefore, the purpose and measure of success isn't whether an engineer uses calculus in their day job. It is a great signal of technical problem solving ability, it is really powerful having a large pool of people who are trained in working with rates, areas and margins." I'm not saying that calculus is useless in an engineer's education - far from it. I'm arguing that knowing the algebraic tricks required to take the limit of (x^2)sin(2x)/sqrt(x) as x->0 is useless. It's just a display of the mastery of very intricate algebra, exactly the same as all the other displays of mastery of intricate algebra she has had to master and will continue to be required to master. What she actually needs is a solid understanding of the physical meaning of derivatives and integrals, how we arrive at them via numerical approximations, the fact that the velocity of a rigid body can be described as the integral over the forces that body is experiencing, etc. That connection is literally never made in calculus. The connection is of course made in the parallel calculus-based physics class she is taking. Which has its own problems, but is far superior as an introduction to calculus than the actual calculus class is.
- RhysU 4y ago> The connection is of course made in the parallel calculus-based physics class she is taking... The physicists as always (a) are pragmatically out-front but (b) don't worry about whether their tools are internally sound so long as they make useful predictions. Sounds about right. It is important for the aforementioned daughter to know this cultural difference between physicists and mathematicians if she is to be a good engineer. I believe this can only be learned firsthand if it is to be appreciated.
- ivansavz 4y agoIt sounds like your daughter might enjoy reading my MATH+PHYS book. It covers mechanics and calculus in an integrated manner. I don't do all the proofs, so might not be a good fit for a proofs-based, course, but lots of illustrations and intuition. Preview here: https://minireference.com/static/excerpts/noBSmathphys_v5_preview.pdf https://minireference.com/static/excerpts/noBSmathphys_v5_pr... Concept map here: https://minireference.com/static/conceptmaps/math_and_physics_concepts.pdf https://minireference.com/static/conceptmaps/math_and_physic... Also be sure to check out the SymPy, in case you don't know already: https://minireference.com/static/tutorials/sympy_tutorial.pdf https://minireference.com/static/tutorials/sympy_tutorial.pd... very good for checking answers...