Consensus on "Git"

Consensus on "Git"

[David Symonds]

Hi folks,

Is there consensus on "Git" being the standard capitalisation, versus
"GIT"? I only really see "git" and "Git" on the mailing list and in
most external documentation and websites (e.g. git-scm.com and
github.com), but git's source tells a different picture:

$ cat *.[ch] | egrep -o '\bG[Ii][Tt]\b' | sort | uniq -c
  36 GIT
   7 Git
$ cat Documentation/* 2> /dev/null | egrep -o '\bG[Ii][Tt]\b' | sort | uniq -c
 284 GIT
 155 Git



Dave.

Re: Consensus on "Git"

[Sverre Rabbelier]

Heya,

On Wed, Nov 11, 2009 at 09:32, David Symonds <dsymonds@gmail.com> wrote:
> Is there consensus on "Git" being the standard capitalisation, versus
> "GIT"? I only really see "git" and "Git" on the mailing list and in
> most external documentation and websites (e.g. git-scm.com and
> github.com), but git's source tells a different picture:

Definitely "Git" or "git", the source code ones are probably from
variable names, something like "GIT_INSTALL_DIR" and whatnot.

-- 
Cheers,

Sverre Rabbelier

Re: Consensus on "Git"

[Junio C Hamano]

David Symonds <dsymonds@gmail.com> writes:

> Is there consensus on "Git" being the standard capitalisation, versus
> "GIT"?

I am not a consensus, but from day 1, Linus talked about "git" (all
lowercase) and "Git" was used only when it came at the beginning of a
sentence (an usual English capitalization rule) or in a section or book
title where all words (except prepositions such as "in", "on") are
capitalized, and I try to mimick it myself for consistency.

As a corollary, unless I spell all the other words in capital to SHOUT, I
never write GIT.

Re: Consensus on "Git"

[Reece Dunn]

2009/11/11 David Symonds <dsymonds@gmail.com>:
> Hi folks,
>
> Is there consensus on "Git" being the standard capitalisation, versus
> "GIT"? I only really see "git" and "Git" on the mailing list and in
> most external documentation and websites (e.g. git-scm.com and
> github.com), but git's source tells a different picture:
>
> $ cat *.[ch] | egrep -o '\bG[Ii][Tt]\b' | sort | uniq -c
>  36 GIT
>   7 Git
> $ cat Documentation/* 2> /dev/null | egrep -o '\bG[Ii][Tt]\b' | sort | uniq -c
>  284 GIT
>  155 Git

All upper case in the sources and documentation will mean that it is
either an environment variable (e.g. GIT_INSTALL_DIR as Sverre has
noted), and for preprocessor constants/definitions.

However, those are not what you are searching for. I.e.

$ grep -P '\bGIT\b' *.[ch]
builtin-apply.c:	 * We have read "GIT binary patch\n"; what follows is a line
builtin-apply.c:		static const char git_binary[] = "GIT binary patch\n";
builtin-cat-file.c: * GIT - The information manager from hell
builtin-check-ref-format.c: * GIT - The information manager from hell
builtin-commit-tree.c: * GIT - The information manager from hell
builtin-diff-files.c: * GIT - The information manager from hell
builtin-init-db.c: * GIT - The information manager from hell
builtin-ls-tree.c: * GIT - The information manager from hell
builtin-mktree.c: * GIT - the stupid content tracker
builtin-read-tree.c: * GIT - The information manager from hell
builtin-update-index.c: * GIT - The information manager from hell
builtin-write-tree.c: * GIT - The information manager from hell
config.c: * GIT - The information manager from hell
date.c: * GIT - The information manager from hell
diff.c:	fprintf(file, "GIT binary patch\n");
diff-delta.c: * Rewritten for GIT by Nicolas Pitre <nico@fluxnic.net>,
(C) 2005-2007
fast-import.c:     # GIT does not permit ':' in ref or tag strings.
fast-import.c:  path    ::= # GIT style file path, e.g. "a/b/c";
fast-import.c:  ref     ::= # GIT ref name, e.g.
"refs/heads/MOZ_GECKO_EXPERIMENT";
fast-import.c:  tag     ::= # GIT tag name, e.g. "FIREFOX_1_5";
fast-import.c:  sha1exp ::= # Any valid GIT SHA1 expression;
fast-import.c:  name  ::= # valid GIT author/committer name;
fast-import.c:  email ::= # valid GIT author/committer email;
fast-import.c:  tz    ::= # GIT style timezone;
fast-import.c:		die("Branch name doesn't conform to GIT standards: %s", name);
hash-object.c: * GIT - The information manager from hell
progress.c: * Simple text-based progress display module for GIT
read-cache.c: * GIT - The information manager from hell
read-cache.c: * cache, ie the parts that aren't tracked by GIT, and only used
sha1_file.c: * GIT - The information manager from hell
sha1_file.c:				     " (try upgrading GIT to a newer version)",
sha1_file.c:		return error("file %s is not a GIT packfile", p->pack_name);
sha1_file.c:			" supported (try upgrading GIT to a newer version)",
trace.c: * GIT - The information manager from hell
usage.c: * GIT - The information manager from hell
var.c: * GIT - The information manager from hell

And selected extracts from:

$ grep -P '\bGIT\b' Documentation/*
Documentation/asciidoc.conf:# Show GIT link as: <command>(<section>);
if section is defined, else just show
Documentation/everyday.txt:Everyday GIT With 20 Commands Or So
Documentation/everyday.txt:My typical GIT day.::
Documentation/git-tools.txt:   providing generally smoother user
experience than the "raw" Core GIT
Documentation/git-tools.txt:   is now in core GIT.
Documentation/git-tools.txt:   pg is a shell script wrapper around GIT
to help the user manage a set of
Documentation/git-tools.txt:   Stacked GIT provides a quilt-like patch
management functionality in the
Documentation/git-tools.txt:   GIT environment. You can easily manage
your patches in the scope of GIT
Documentation/git-tools.txt:   gitk is a simple Tk GUI for browsing
history of GIT repositories easily.
Documentation/git-tools.txt:   GITweb provides full-fledged web
interface for GIT repositories.
Documentation/git-tag.txt:GIT
Documentation/RelNotes-1.6.1.txt:GIT v1.6.1 Release Notes

- Reece

Re: Consensus on "Git"

[Jeff King]

On Wed, Nov 11, 2009 at 01:06:44AM -0800, Junio C Hamano wrote:

> > Is there consensus on "Git" being the standard capitalisation, versus
> > "GIT"?
> 
> I am not a consensus, but from day 1, Linus talked about "git" (all
> lowercase) and "Git" was used only when it came at the beginning of a
> sentence (an usual English capitalization rule) or in a section or book
> title where all words (except prepositions such as "in", "on") are
> capitalized, and I try to mimick it myself for consistency.
> 
> As a corollary, unless I spell all the other words in capital to SHOUT, I
> never write GIT.

Except for the title of every set of release notes, which all start
with:

  GIT vX.Y.Z Release Notes

? :)

For a fun (ab)use of blame, you can see who wrote each of David's
suspect lines with:

   git grep -n "\bGIT\b" |
   while IFS=: read file line junk; do
     git blame -f -L $line,$line $file
   done |
   less

Many of them are variables (e.g., \b gets rid of "_", but we still have
"-", so GIT-VERSION and such are still there). Many others are the title
of the "GIT" section of the manpages. But there are some legitimate
uses, too. Many of them blame to quite a long time ago, though (e.g.,
Documentation/everyday.txt has two uses by you in 2005).

-Peff

Re: Consensus on "Git"

[Johannes Schindelin]

Hi,

On Wed, 11 Nov 2009, David Symonds wrote:

> Is there consensus on "Git" being the standard capitalisation, versus
> "GIT"?

There is definitely no consensus.

Personally, I started to capitalize it as a reaction to business types 
(apparently you have to turn in your sense of humor when you get an MBA), 
who were not happy with the abbreviation "global information tracker" 
(which would merit an all-capitalization a la CVS) [*1*], let alone with 
the connotation as a British slang word.

Yes, in Open Source, we are used to crude, geeky humor, some of us even 
accept blunt mails, but we are embedded in a world that is called 
"reality".

In my experience, in this world it is much easier to tell people that your 
program is called "Git" ("could have been John, either") and be done with 
the questions.

Ciao,
Dscho

Footnote *1*: I tried once to reinterpret GIT as meaning "Git Is 
Terrific", but that did not fly too well. :-)

Re: Consensus on "Git"

[Junio C Hamano]

Jeff King <peff@peff.net> writes:

> Except for the title of every set of release notes, which all start
> with:
>
>   GIT vX.Y.Z Release Notes
>
> ? :)

Yeah these were copied from release to release and I do not pay much
attention to the minute details of how my tools I wrote long time ago
spells things ;-)

Perhaps I should update my scripts to generate them.

Re: Consensus on "Git"

[Jeff King]

On Wed, Nov 11, 2009 at 04:27:31AM -0500, Jeff King wrote:

> For a fun (ab)use of blame, you can see who wrote each of David's
> suspect lines with:
> 
>    git grep -n "\bGIT\b" |
>    while IFS=: read file line junk; do
>      git blame -f -L $line,$line $file
>    done |
>    less
> 
> Many of them are variables (e.g., \b gets rid of "_", but we still have
> "-", so GIT-VERSION and such are still there). Many others are the title
> of the "GIT" section of the manpages. But there are some legitimate
> uses, too. Many of them blame to quite a long time ago, though (e.g.,
> Documentation/everyday.txt has two uses by you in 2005).

BTW, here is a much better grep (still a few false positives, but most
lines are meanginful):

   git grep -n -E "(^|[^A-Z_-])GIT($|[^A-Z_-])" |
   grep -v ':GIT$' |
   while IFS=: read file line junk; do
     git blame -f -L $line,$line $file
   done |
   less

You can really see that most of the uses of GIT are ancient. I wonder if
we should downcase the few that are in user-visible error messages. I
will let somebody else spend their time on such a conversion if they
want, though. I've already spent enough on this. :)

-Peff

Re: Consensus on "Git"

[Teemu Likonen]

On 2009-11-11 10:33 (+0100), Johannes Schindelin wrote:

> Yes, in Open Source, we are used to crude, geeky humor, some of us
> even accept blunt mails, but we are embedded in a world that is called
> "reality".
>
> In my experience, in this world it is much easier to tell people that
> your program is called "Git" ("could have been John, either") and be
> done with the questions.

Yes, that’s how the language works. Proper nouns are capitalized, so in
plain text the name of this program should always be written “Git”. Only
in command examples it is “git”.

Sometimes people mix proper nouns and logos/brands. Proper nouns are
under the rules of the language. For example, I can’t go telling people
that my name must be written “teEMu” because I don’t own Finnish.

In logos the brands’ owner can try to be funny and cool and different
and write anything they want (like “gɨT” or “GīT”).

Re: Consensus on "Git"

[Michael J Gruber]

Teemu Likonen venit, vidit, dixit 11.11.2009 15:58:
> On 2009-11-11 10:33 (+0100), Johannes Schindelin wrote:
> 
>> Yes, in Open Source, we are used to crude, geeky humor, some of us
>> even accept blunt mails, but we are embedded in a world that is called
>> "reality".
>>
>> In my experience, in this world it is much easier to tell people that
>> your program is called "Git" ("could have been John, either") and be
>> done with the questions.
> 
> Yes, that’s how the language works.

And why Dick Cheney was allowed to keep his name ;)

> Proper nouns are capitalized, so in
> plain text the name of this program should always be written “Git”. Only
> in command examples it is “git”.
> 
> Sometimes people mix proper nouns and logos/brands. Proper nouns are
> under the rules of the language. For example, I can’t go telling people
> that my name must be written “teEMu” because I don’t own Finnish.

Given that reasoning, we should really use "Git" whenever the program or
project name is meant. Is that something we can reach concensus on?

> In logos the brands’ owner can try to be funny and cool and different
> and write anything they want (like “gɨT” or “GīT”).

Nice!

Michael