|
According to Wikipedia, the phrase "Considered Harmful" was already
a journalistic
cliché
used in headlines prior to the 1968 paper by
Edsger Dijkstra entitled "Go To Statement Considered Harmful".
It's now something of a
cliché/meme
in IT and software circles, and
the original careful, nuanced discussion is usually lost in people's
eagerness to show off their knowledge of it.
https://en.wikipedia.org/wiki/Considered_harmful
|
While I was writing my post on Signal Reflection, I was given cause
to reflect (pun intended ... sorry) on exactly the role and value
of analogies in science, maths, and life in general. I've come to
the conclusion that, like so many thing, analogies are neither good
nor bad. They have their purpose, and each lies somewhere on a
spectrum.
The first purpose of an analogy is to connect a new thing with
something the listener (or reader) already knows. This then can
serve to provide a framework, and a way of thinking about the new
thing based on a familiar context. As such, a good analogy can be
useful.
But for some analogies, they seem to me to be mostly useless. I've
seen an analogy used to describe Diffie-Hellman-Merkle-Williamson
key exchange algorithm in terms of mixing paints, an analogy that
seemed particularly useless. Others, such as sending valuable items
by using a double-locked strongbox might be better, but after the
initial moments of going "OK, I can see how that might work", the
listener/reader is then left with nothing to use.
On the other hand, the analogy of chemical molecules as "ball and
stick", or perhaps even better, "ball and spring", actually lets you
do something. It's possible to use it to calculate approximations
of energy release. It's perhaps closer to being a model than an
analogy, but there again, it's a continuum.
There's a saying:
- All models are wrong, some models are useful.
The same is true of analogies. All are "wrong" in the sense that
they can be pushed too far and either made to seem ridiculous, or
to give wrong intuitions, or both. But some are useful.
When presented with an analogy, it's worth doing a quick assessment
of where in the spectrum it lies.
Send us a comment ...
| 
Suggest a change ( <--
What does this mean?) /
Send me email
Subscribe!
