Re: [PATCH 1/2] Refactor digest authentication in url-auth

[Eli Zaretskii]

> From: Yuri Khan <address@hidden>
> Date: Mon, 14 Nov 2016 10:34:18 +0600
> Cc: Jarno Malmari <address@hidden>, Emacs developers <address@hidden>
> 
> >> > GNU coding standards frown on using "path" for anything but PATH-style
> >> > directory lists.  Please use "file name" or "directory name" instead.
> >>
> >> Good to know. Is there a convention to indicate path part in URIs?
> >
> > Not that I know of.  I'd use "file-name part".
> 
> To a network application developer, “path” is a domain[^1] term (RFC
> 3986 § 3.3) and any replacement would be surprising and hinder API
> discoverability. URIs don’t even necessarily map to files, so the
> replacement proposed above is, in addition, misleading.

The terminology in this case doesn't include "path", it includes
"URI" and "abs_path" (see RFC 2617).  The use of "path" in the doc
strings being discussed should therefore be minimal, ideally zero.

The clash with terminology accepted elsewhere is unfortunate, but
nothing new: we have the same issues with "yank" vs "paste", "frame"
vs "window", etc.  GNU Coding Standards are mandatory for us, because
GNU documentation is primarily for users of GNU software, who are
accustomed to our terminology which is used consistently across
projects.  Therefore we should strive to use our terminology whenever
we can, and if that requires some "creative" wording, so be it.

> Please consider “URI path” if you must disambiguate.

If needed, we should use "the path component of a URI" or somesuch.
But ideally the need should not arise in this case at all, at least
not AFAICT.