Re: ASCII convention

Peter T. Daniels wrote:

Those three paragraphs are irrelevant to the question of what to use in
ASCII.


(So why did you quote them?)

I didn't notice any restriction to ASCII in the question, though the question was somewhat obscure and mentioned ASCII at the end. Of course, in sci.lang it's usually best to stick to ASCII, but I think we can discuss the usage in general here.

If, however, you refer to typography where special fonts (containing
angle brackets) are not available, and if you're clever enough to use a
Mac, the Single Guillemets provide a very reasonable substitute

I don't see why you could not use single guillemets in other environments. They are Windows Latin 1 characters.

On the other hand, they are quotation marks, and it is reasonable to assume that both human readers and computer programs will treat them according to their definitions as single quotation marks as used in some languages. Using them just because they "look right" is very unsafe (and I'm not talking about character encoding here). They are typically quite small as compared with angle brackets, and any font change may increase the difference.

If you use substitutes, it's best to use the same substitutes as everyone and his brother, not invent your own.

> Unfortunately, they
turn into Icelandic in Windows: instead of <xyz> write Ðxyzð.

They turn to something unpredictable if you send data in a proprietary encoding without specifying the encoding, or specifying them incorrectly. They might turn to something unpresentable (which is a different thing) if you specify the encoding but the recipient's software is unable to interpret the encoding. As a rule, when sending anything from a Mac, it is best to ensure that the data is correctly convented to some widely recognized format (such as ISO-8859-1, as your E-mail client seems to do, or UTF-8, if ISO-8859-1 is insufficient).

Quotation marks already have other functions that are likely to occur in
the same text as the mentioning of letters.

Quotation marks are indeed overloaded with many meanings, even when used by the rules of a language. (We use them in direct quotations, but also for many other purposes, e.g. to indicate that we are using a word in an extraordinary meaning, or as "irony quotes", or something.) That's one reason why they - including single and double guillemets - should not be overloaded further by using them as delimiters.
.

Relevant Pages

  • Re: Problem using BinaryReader
    ... > specifying the encoding, desktop will use ASCII and device will use Unicode. ... and BinaryReader), but even when I set them both to ASCII (I even tried other ...
    (microsoft.public.dotnet.framework.compactframework)
  • =?iso-8859-1?q?Re:_About_the_Apparent_Resurrection_of_Thoth_and_Mr_B.C._=83?=
    ... Thoth seems to be best for Mac binaries. ... You are confusing compression with encoding. ... one needs to understand the difference between "binary" and ASCII. ... These are all ASCII encoding formats. ...
    (comp.sys.mac.apps)
  • Re: Padding Bits
    ... hardware platform that uses padding bits. ... etc.), they have committed to using a specific character encoding, ... examples in the book "C Unleashed" in ASCII encoding, ...
    (comp.lang.c)
  • Re: Unicode to ASCII string conversion
    ... files can be ANSI, ASCII, UTF7, UTF8, EBCDIC, UTF16 and many other ... > to string conversion and explicitly no bytearray, ... >> If you need an ASCII file, then use a ASCII encoding. ...
    (microsoft.public.dotnet.languages.vb)
  • Re: Special Characters (Unicode, Ascii) in Python and MySQL
    ... Special characters and punctuation all seem not to be stored and ... perhaps the encoding of the database itself should be different? ... 'ascii' codec can't decode byte 0xe2 in position ... and pasted from Microsoft Word. ...
    (comp.lang.python)
Loading