They also don't concisely prove that my aunt is not called Susan. (She's not, fyi). The only thing that ever could concisely prove it is if we were to objectively count the occurrences of such incidents that has ever happened anywhere. Ever. I.e. impossible to do so.
What we can take from it is that due to their being more discussion in areas that do not concern performance that it is fair to assume that performance (to the nth degree along the lines of those in this thread) is not as wide spread a concern as some are making out.
The ten year thing, at face value, is just akin to "ages" to me. (I didn't spot that post before as I was looking out for 'ten' and not 'decade'
) Which I agree with.
Since the publication of linked lists and the like (and the more powerful compilers, VMs/CLRs etc. that coincidentally came around the same time as them) the concern for performance has dropped massively. Never gone - I don't claim
nobody cares, obviously some do (source: this thread
) but the days of designing your own sorting or searching algorithm and the like (as claimed by D.P.) are pretty gone for an overwhelming majority of developers. More and more so as tech has advanced, resulting in the differences in results made to smaller and smaller gains, on top of all that. Much like nearly all Java and C# developers don't ever use malloc().
But this is all aside from the point. The OP asks about Maths. I'd argue a lot of the examples are also fixed by common bloody sense. If someone needs to sort a list of millions upon millions of things, I first want to know why they have all that info in memory and not in a managed environment like a DBMS, where the algorithm used is abstracted away from the developer of the application, anyway. Next if I see them sorting that same list in the same way every time they need something from it, they need a slap to wake up. etc.
