MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
References: <49A453A1.9040000@morningstar2.co.uk>
            <FB18C832-EA5E-42F0-9BF7-6853DB79B805@gmail.com>
            <49A4F031.50405@morningstar2.co.uk>
            <49A4FE4F.6040304@morningstar2.co.uk>
            <27990a880902250600v21963220v319195a504107625@mail.gmail.com>
            <18853.41758.364009.703166@morse.mittelbach-online.de>
            <49A5A982.9050508@morningstar2.co.uk>
Message-ID:  <18853.44886.353831.975849@morse.mittelbach-online.de>
Date:         Wed, 25 Feb 2009 21:51:34 +0100
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
Sender: Mailing list for the LaTeX3 project <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
From: Frank Mittelbach <frank.mittelbach@LATEX-PROJECT.ORG>
Subject: Re: Missing expl3 primitives
To: LATEX-L@LISTSERV.UNI-HEIDELBERG.DE
In-Reply-To:  <49A5A982.9050508@morningstar2.co.uk>
Precedence: list
Status: R

Joseph Wright writes:
 > One "philosophical" question: are the xpackages allowed to use :D
 > functions?  My (possibly faulty) understanding was that expl3 is the
 > language definition, and only the functions defined there are allowed
 > anywhere else in LaTeX3.  This implies to me that the xpackages can't
 > use :D functions, but that something in expl3 has to make the
 > functionality available.

my philosophical answer ... that depends :-)

understand the xpackages as experimental modules. Some of them will end up
becoming core kernel modules, e.g., something like a galley concept has to
live in the kernel really as it cuts across everything. Others are experiments
in what would become extention modules.

Now for kernel modules providing higher-level interfaces to :D stuff is ok,
for others we may find that even though we want to keep them as extention
modules, ie something that could be replaced by something else, that we need
to distill some functionality into the kernel as supporting functionality.

in other words, the xpackages are not outside the kernel per se, but just
first bigger attempts to implement functionality that either doesn't belong
into the kernel (like xinitial) or isn't yet ready in any way to be considered
part of the kernel but eventuallly some sort or form of it will (eg xor,
galley, ...)

frank