Re: Capitalization conventions for Debian release names
On Fri, 26 Apr 2002, Osamu Aoki wrote:
> My above mentioned observation is the practice by the people who build
> Debian tends to prefer using lower case. I do not know root cause of
> it. My guess is Unix gurus like lower case. Unix is not PDP-11/VMS (or
> DOS) world where capitalization is more popular.
I suspect that when people type "potato" and "woody" they are influenced
by the fact that in the FTP tree those are the lowercased names of
directories to which "stable" and "testing" are symlinks. Using the form
"woody" suggests that you conceive of "woody" as "everything that lives
within the directory /debian/dists/woody". But see following comments...
> For me key word is "convention" in that particular
> organization. It rules over some generic school / news paper / journal
> grammar guideline.
I agree completely. If there is an agreement among Debian documentation
authors/editors to use a particular form, that is the correct form.
So let me rephrase things. I would like to make the following proposal
for a convention regarding release names in Debian documentation,
followed by my reasons for the proposal.
1. When used by itself, the codename of a Debian release should
be capitalized, in ordinary text font. Example:
To upgrade from Potato to Woody, follow these steps . . .
2. When the codenames refer to physical locations in a distribution
tree, lowercase them and use filename syntax. Example:
Currently, <file>unstable</file> is a symbolic link to
<file>woody</file>. Files in <file>woody</file> are . . .
3. Use filename syntax for the distribution meta-names "stable,"
"testing", "unstable", and "sid". ("sid" was originally a proper
name, but now it is a pure synonym for "unstable".)
My main reason is that I think a semantic distinction should be
preserved between "Potato/Woody" and "stable/testing/unstable". The
former are particular states of Debian; once frozen, they do not change
(apart from security patches) The latter names represent three different
statuses of Debian release whose relationship is invariant, even though
their contents constantly change. In short: "Potato" and "stable" are
Anyway, this is not a hugely important issue and I am happy to go with
the general consensus. Or if everyone can't agree, then Osamu's
preference as author should be the rule for "Debian Reference".
David Sewell, Project Editor
The University Press of Virginia
PO Box 400318, Charlottesville, VA 22904-4318 USA
Email: email@example.com Tel: +1 434 924 6066
To UNSUBSCRIBE, email to firstname.lastname@example.org
with a subject of "unsubscribe". Trouble? Contact email@example.com