1
0
mirror of https://github.com/nbeaver/why-linux-is-better.git synced 2025-08-26 07:14:23 +02:00

Quotes from book.

This commit is contained in:
Nathaniel Beaver
2016-04-22 15:18:11 -05:00
parent f34cabbb43
commit a09dcb084d

View File

@@ -363,9 +363,67 @@ Filename case-insensitivity.
Linux uses case-sensitive filenames Linux uses case-sensitive filenames
because Unix used case-sensitive filenames. because Unix used case-sensitive filenames.
Unix was case-sensitive because Multics was case-sensitive. Unix was case-sensitive because Multics was case-sensitive.
Multics was case-sensitive because of ASCII. [#Multics_case_sensitive]_ Multics was case-sensitive because the ASCII standard
included a lowercase alphabet. [#Multics_case_sensitive]_
This has some intuitive appeal; Why was ASCII case-sensitive?
It nearly wasn't.
Early telegraphy codes did not distinguish upper and lowercase
because it would have slowed transmission speeds prohibitively.
Encodings with different bit patterns for uppercase and lowercase
had been proposed as early as 1959, [#Bemer_1959]_
but is was not widely implemented.
The IBM 7030 "Stretch" supercomputer used an 8-bit encoding
that included interleaved uppercase and lowercase alphabets,
and it was used at Los Alamos in 1961.
[#Stretch_supercomputer]_
The ASCII committee concluded that 6-bit encodings (64 bit patterns)
were insufficient to include both control characters and special characters
in addition to the required 26 alphabetics and 10 numerics,
so they decided to use a 7-bit code.
The consideration of a 6-bit, 64-character graphic subset was important
to the standards committee. If the ultimate decision was that columns 6
and 7 would be for graphics, then columns 2 through 7 would contain
Space, 94 graphics, and Delete. But, even with the code providing 94
graphics, a major assumption of the standards committee was that data
processing applications would, for the foreseeable future, be satisfied with
a monocase alphabet (that is, a 64- or less graphic subset) as they had in
the past---that 64-character printers would predominate. So it was import-
tant to be able to derive a 64-character, monocase alphabet, graphic
subset from the code by simple, not complex, logic.
--- Charles E. Mackenzie, "Coded character sets: history and development" (1980), p.228
In fact, the some of the committee members
wanted to reserve the remaining to use the remaining space for control characters.
The conclusion of the preceding paragraph is based on the assump-
tion that two alphabets, small letters and capital letters, would be in-
cluded in the 7-bit code and that decision had not yet been made. If the
decision was ultimately made that columns 6 and 7 would would contain
controls, then small letters would not be included in the 7-bit code. ``*``
``*`` If the committee did decide for controls in columns 6 and 7, it is still likely that
they would have wanted an alphabet of small letters to be provided. Presumably,
the small letter alphabet would then have been provided by a caseshift approach.
--- Ibid, p.232
Thought the comittee first formed in 1961,
it wasn't until late 1963 that they finally agreed to include a lowercase alphabet.
At the first meeting of ISO/TC97/SC@ in 1963 October 29-31, a resolu-
tion was passed that the lower-case alphabet should be assigned to
columns 6 and 7.
--- Ibid, p. 246
Why is it useful for filenames to include upper and lowercase?
It has some intuitive appeal;
it is useful to be able to distinguish between, say, it is useful to be able to distinguish between, say,
the abbreviation for United State ("US") the abbreviation for United State ("US")
and the first-person plural objective pronoun ("us"). and the first-person plural objective pronoun ("us").
@@ -506,6 +564,21 @@ For example, the Linux port of the `Unity engine`_ has `issues with case-sensiti
http://www.multicians.org/pl1.html http://www.multicians.org/pl1.html
.. [#Bemer_1959]
Simple pattern of correspondence should exist between codes assigned to
upper and lower case alphabetic characters.
--- R. W. Bemer
From page 20 of "A proposal for a generalized card code for 256 characters",
Communications of the ACM, Volume 2 Issue 9, Sept. 1959.
http://dx.doi.org/10.1145/368424.368435
.. [#Stretch_supercomputer]
From "Coded character sets: history and development" by Charles E. Mackenzie, 1980.
.. [#against_case_sensitivity] .. [#against_case_sensitivity]
Mac Windows users have to have filenames read to them over the phone by Mac Windows users have to have filenames read to them over the phone by