“It’s a complicated logarithm.”
My head snaps up. What did he just say?
Nah, I think, he couldn’t have confused “logarithm” for “algorithm” in describing the process of trying to accommodate multiple, competing objectives in a public committee environment.
But he did, because he did it again, a minute or so further into the interview.
“It’s a complicated logarithm.”
Now, I seem to remember logarithms being complicated all by themselves, and Wiki confirms this memory.
In mathematics, the logarithm is the inverse operation to exponentiation,
just as division is the inverse of multiplication and vice versa.
Using teaching techniques that I sorely needed as a student, the Khan Foundation explains logarithms and offers “a few” examples in a seven-minute video. I mean, would it take seven minutes to explain and example something that wasn’t inherently complicated?
No. No, it would not.
By contrast, Wiki defines an algorithm thusly.
In mathematics and computer science,
an algorithm is an unambiguous specification
of how to solve a class of problems.
And the Khan Foundation manages to explain “algorithms” in just 5 1/2 minutes in the video accompanying the compellingly titled article: What is an algorithm and why should you care?
I do not know if there is, or even can be, an unambiguous specification for how to solve the general problem of multiple, competing objectives in a public committee environment, but I can agree wholeheartedly with the interviewee on one point.
It’s complicated.
Yet another good reminder that in interviews, as in much of life, the less said the better.
I thought I understood logarithms. I listened to the Khan academy explanations, and I understand more than I did, except for one thing — why in hell would anyone bother writing log2(64) instead of 2 to the 6th (can’t do superscript and subscript in this reply)? And as a corollary, why in hell would anyone want to say that something is a complicated logarithm, about anything? Seems to me he’s suffering from logorrhoea.
Jim T
Jim T – I did not listen to the video. I have to draw the line somewhere. As for the choice of words, it was just a misapprehension, I’m thinking. I can understand saying that the algorithm they were operating under was complicated and this was clearly the intended communication. But an overall simpler approach would have avoided all this. And left me nothing to write about . . .
Once again you have made my brain hurt!
John W
John – My work here is done. Or not . . .
Many years ago I was encouraged to join a service club. Against my better judgement, I did so…nothing against service clubs, I just did not want to be part of one. My tenure ended when an internationally known physicist from the University was introduced as “…a well known astrologer”. Enough, thought I, and left, never to return. Words matter.
Ian – That’s hysterical! I remember talking about a PBS show I’d seen on cosmology to a marketing class (OK, severely off-topic, I admit) and when a student came in late and I mentioned this show, he said he didn’t know anything about make-up.
So, I guess the algorithm will depend on how you define the set of problems. That may be where the profound issues in the “advancement” of technology lie. And thank you for this hilarious seminar. I can draw the logarithmic line “somewhere,” now, too!
Laurna – Yes, defining the set or class of problems reveals a lot – if not about the problems themselves then about the person doing the defining/classification. Dealing with FAQs online, I too often find that my problem is not represented . . . Maybe they should always include one named “User error.”
I find FAQ pages singularly unsatisfying. They usually ask, “Did you find this page/answer useful?” I almost invariably answer “No.” If a single cookie-cutter solution would work for everyone inquiring, I want to know, why don’t they fix the damned problem instead of just publishing an answer — which only a few of those experiencing the problem will ever see? Besides, let’s assume there are two billion users of Microsoft programs worldwide, what are the chances they will all encounter the same problem? Every individual user will get into difficulties in his/her own unique way; a one-size-fits-all FAQ response is hardly likely to work for all those unique situations. Arrrggh. My rant for the day.
Jim T
Jim – I think that writing answers to FAQs (and identifying them in the first place) is very like writing instructions. To do the latter it well, you have to be able to put yourself in the mindset of the person who’s trying to follow your instructions, and tell them every step. No jumping over what seems obvious to you. To do the former, you have to be able to do that, as well as understand how things are likely to go wrong, as perceived by the user. That’s a lot of scope! No wonder it’s not done well.