On 03/12/17 10:58, occam wrote:
Post by occam
Has anyone considered the possibility that 'Google Translate' is taking
the piss? I have never been convinced of its claim that it can handle
the extensive list of 'language pairs' shown on the site. What we are
seeing is empirical proof that it cannot. (I am curious to know which
language Richard H.'s translation is from?)
English, I'm afraid. I didn't use Google Translate. I used Richard
Translate. This is not software, but wetware.
It works like this. Given an arbitrary word or phrase in a foreign
language, such as:
arvno anfio cenka grapolr
(and yes, I just mashed the keyboard to get that phrase), I now
translate each word in turn. If it happens to remind me of a foreign
word that I /do/ know, then obviously that's the correct translation. If
it doesn't, I pick the nearest English word or words that convey
approximately the same sound and which might conceivably be
In this case:
"arvno anfio cenka grapolr"
"Have no hand fee, O conker grappler."
In one of my exams for Latin O-level, I latched onto the word "uxor" and
- using much the same technique as outlined above - I went on to
describe the joy of the wedding, the dancing of the guests, and so on. I
discovered later (after the exam, of course) that the passage was
actually about a chap waving goodbye to his family as the ship departs
that is carrying him away into exile without them. Oops.
Astonishingly, I passed. (Despite the above, rather than because of it.)
Email: rjh at cpax dot org dot uk
"Usenet is a strange place" - dmr 29 July 1999
Sig line 4 vacant - apply within