 'Blbrgrd'


Why would putting one ISO 8859-1 character in mean doubling the post?



From: Peter N. M. Hansteen <peter@datadok.no>   \-(1)+-(1)--(1)--[1]+-[1]--[1]-
[1] Re: Grossly underpaid                            |              |-[1]--[1]
Date: Fri Jun 02 10:38:54 EDT 2000                   |              \-[1]--[1]
Organization: Datadokumentasjon A/S                  \-[1]
Lines: 45
Original-Sender: peter@datadok.no
X-Newsreader: Gnus v5.7/Emacs 20.6

rsteenw@xs4all.nl (Rik Steenwinkel) writes:

> I'll be damned if I know what exactly is going on here, or rather
> _what_ part of the codpage-to-iso-8859-1-mapping is b0rken.

That would be to imply rather too strongly for my taste that only one
part is borken. I beg to differ.

> On my system the original article shows a 'cents' sign {c|} for the
> {o/} (0xF8 in the raw article) , while the {ao} (0xE5) and the {ae}
> (0xE6) show correctly.

That means you're actually quite lucky. I've seen MSvoid files which
in translation from Swedish to Norwegian grew a number of truly silly
character substitutions for [1].

Historians will perhaps in time dig up the basic reason for all this
silliness, including one often whisperingly repeated anecdote of a
celebrated .no "computer linguist". This luminary was contacted by IBM
just to perform a sanity check on the character set they were going to
use on that pesky little micro, the PC. Said he, "oh, we need  and ,
just like the Danes". Enter the era when one of the true joys computing
could be expressed in .no and .dk with the help of  for  and  for .

The next step in wonderment came when one is forced to realize that MS
chose almost, but not quite, iso8859-1 for their 8bit character set in
win*. Forced to deal with the cold brutal world of weird characters
and accents, MS then invented[2] another code page scheme with
numbering apparently pulled out of a magician's hat which,
incidentally, seemed to have just been vacated by a rather bored and
overfed pigeon.


[1] HAH! Now let's count the number of merkins who actually saw something
    approaching reasonable there.

[2] Now I'm hoping somebody will come up with the canonical reference
    for the quote, "Considering the number of wheels Microsoft has
    found reason to invent, one never ceases to be baffled by
    miniscule number whose shape even vaguely resembles a circle".

