[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gentoo-dev
Subject:    [gentoo-dev] on case-sensitivity, from the Amiga open mailing list...
From:       <drobbins () gentoo ! org>
Date:       2001-02-22 10:30:03
[Download RAW message or body]

Give this a read :)  All about case sensitivity.

----- Forwarded message from Rudi Chiarito <rudi@amiga.com> -----

To: open@amiga.com
Date: Wed, 21 Feb 2001 17:44:25 -0600
From: Rudi Chiarito <rudi@amiga.com>
Subject: AMIOPEN: Re: System case (in)sensitivity.


On Fri, Feb 16, 2001 at 11:37:38AM -0800, Joe Kiniry wrote:
> 1. The majority of written human languages (spoken, mathematical, and 
> symbolic) do differentiate characters with a notion of "case".  Some also 
> differentiate with other typographic notions; font characteristics in 
> mathematics as the primary example.

They make the distinction for different reasons, though.

Most natural languages actually don't. It's only languages derived from
European scripts (Roman, Greek) that do it. Even for those, it's a fairly
recent development. If you look at the wonderful inscriptions (often a
work of art of their own) on any monument from Classic Greece or Ancient
Rome, you'll see that all of them are in uppercase. The modern Greek
small letters didn't exist before the 9th century, when the Byzantine
minuscule script was introduced. Such script, based on cursive styles
that scribes had developed through the centuries, with time evolved into
the lowercase characters we frequently use in e.g. trigonometry. Same
with the Roman script: it was Charlemagne's reforms around the 9th
century that brought the Carolingian Minuscule script, which is the basis
for today's scripts.

Johann Gutenberg's invention of printing happened at a time when
scholars, in the context of Humanism, had reformed scripts with the
explicit purpose of improving both legibility and aestethics. Thus,
printers like William Caxton in England or Aldus Manutius in Venice, when
creating their own models, relied heavily on what was common in those
days, i.e. Humanist scripts, which descended from Carolingian Minuscule
(something similar occurred with models for Greek text). Printing was a
major factor in making minuscules widely accepted, because one of their
main advantages, compared to older or Gothic scripts, is excellent
legibility even at small sizes and that was a desirable feature for
printers (the smaller the type, the more costs went down).

All of the above to say that it was calligraphy to bring us capital and
small letters. Caesar and Romans of his time did just fine with capital
letters only. We have rules today that we are taught to follow in school,
yes, but they're not essential for communication. When text is sent as a
telegram, in Morse code or as a SMS message (because it's already enough
of a pain to write on a cellular phone a message with the proper letters,
let alone with proper case), we still understand what the sender meant,
even if it's all in uppercase.

Interestingly enough, Japanese, Chinese, Arabic, Hebrew, Hindi, etc. do
not feature any notion of anything resembling cases - Japanese has
furiganas and half-width characters, but I'm not sure if they could be
considered the same. The only language not descending from European
scripts to use something like case is Javanese - they use "their
uppercase" for proper names, with the difference that their capital
letters can occur also in the middle of a proper name, not just as the
beginning as we are accustomed to.

Even with languages using the same script you have different rules.  
English uses capital letters in e.g. "a Greek dentist" and "on December
21", but an Italian would say "un dentista greco" and "il 21 dicembre".
That can only confuse things some more.

Mathematics has a different problem. It just can't have enough symbols to
represent so many entities. When you need to refer to a constant, to a
variable, to an operation, etc. and you want to keep representation
concise _and_ you want to avoid ambiguities, your only solution is to use
a large number of symbols. So, first you start, out of familiarity, with
your own alphabet (e.g. Roman), then you use somebody else's (e.g.
Greek), then you start making variations on what you already have: e.g.
the "for each"  upside-down A quantifier, Roman characters with apices,
in italic or in a different font. I haven't counted them, but I think
Unicode has more than 1000 or 2000 symbols for mathematics.

Mathematics has to use case and other techniques just out of sheer
necessity, because otherwise it'd easily fall into ambiguities, which
would defy one of its main aims. With most natural languages, on the
other hand, ambiguities are allowed - and are quite common indeed - but
they're resolved at one level or another anyway and as such don't pose
serious risks to communication.

> 2. The most widely used computer on the planet that purportedly appeals to 
> beginning computer users is the Macintosh.  Macs *are* case sensitive.

I have used Macs many times but never noticed that. Surprising. I suppose
that's because most Mac users never specify filenames with the keyboard,
they just point and click.

> The UNIX and Mac users typically stated a variation of the following with 
> regards to the second question: "Why wouldn't you?  We write with case, and 
> its never caused me trouble before."  Often such comments were made with a 

We Westerners do, but e.g. the Japanese don't. That's one of the many 
reasons they had many problems getting familiar with the Western concept
of typewriter or word processor.

-- 
 "I think love lyrics have contributed to the general aura of bad mental
  health in America."                                      (Frank Zappa)
 Rudi Chiarito         SGML/XML, user interface, i18n         Amiga Inc.
 rudi@amiga.com                                        http://amiga.com/

--
Subscribe/Unsubscribe: open-request@amiga.com
Amiga FAQ: http://www.amiga.com/faq.html



----- End forwarded message -----

-- 
Daniel Robbins					<drobbins@gentoo.org>
President/CEO					http://www.gentoo.org 
Gentoo Technologies, Inc.			


[prev in list] [next in list] [prev in thread] [next in thread]