🚀 go-pugleaf

In article <5207@shum.huji.ac.il> yair@cs.huji.ac.il writes:
>
[re "dilemna"]
>I think this is a common typo. Since n is next to m on qwerty key keyboards this
>sort of error is very reasonable.  It is also hard to detect beacuse the letters
>look very similar.

Although I am not an expert in this area (there has been some interesting
research on typing and errors in typing), my view is that the proximity
of "m" and "n" on the keyboard would NOT lead to this error. Once
you have hit "m" once, you have no reason to move your finger, and no
way to "miss" the key. It is true that on a _computer_ keyboard,
if your finger landed so it depressed both keys, you might get "mn",
but you would know that you had only pressed down once, and when
you pressed again you would end up with "mnm". If the error were
one in which a single "n" were substituted for a single "m," then
it might be credible as a typo (e.g. "tane" for "tame" etc.).

It is well known, however, that all but the most highly skilled
typographers will substitute their own spellings of words they know
for different spellings of the same word--without being aware
that they are doing so. For example, an American typesetter keying
in a manuscript by a British author will have a hard time making
sure that he writes "colour" every time he encounters this word.
The probability is high that at least a few times he will write
"color." This has been a common problem in Norway, where many
variant spellings of certain words are "permitted" (officially),
where on occasion certain spellings have been "outlawed," and
where authors have varied as to which language reforms they have
adopted. Close examination of published texts reveals that in many
cases an author's spelling has been unwittingly changed by typesetters.

So far the evidence suggests that a certain number of library
cataloguers have simply had the wrong idea about how "dilemma"
is spelled. They may, of coure, have been influenced by a small
number of books in which "dilemna" actually appears on the title page.
>-- 
>					Yair Friedman.
>yair@cs.huji.ac.il
>yair@hujinix.BITNET


-- 
Roger Greenwald
Greenwald@hf.uib.no (address valid only until 10 August 1992)