User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Icedove/31.0
MIME-Version: 1.0
References: <559AE0A8.3080907@morningstar2.co.uk> <55A3E1DD.8050707@posteo.net>
            <55A3E5E6.2060405@morningstar2.co.uk> <55A4137A.7030807@posteo.net>
            <55A415FF.8030305@nag.co.uk> <55A801A8.5090702@posteo.net>         
            <55A809C4.70009@nag.co.uk> <55AA1554.3050509@posteo.net>           
            <55AA24D9.1050904@latex-project.org> <55AA29B9.9050104@posteo.net> 
            <55AA5596.4090007@telecom-bretagne.eu> <55AA755E.5000609@posteo.net>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Message-ID:  <55AA960A.809@telecom-bretagne.eu>
Date:         Sat, 18 Jul 2015 20:08:10 +0200
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: =?UTF-8?B?w4lsaWUgUm91eA==?= <elie.roux@TELECOM-BRETAGNE.EU>
Subject: Re: LuaTeX support in the LaTeX kernel
To: LATEX-L@LISTSERV.UNI-HEIDELBERG.DE
In-Reply-To:  <55AA755E.5000609@posteo.net>
Precedence: list
Envelope-To: <rainer.schoepf@GMX.NET>
Status: R

> Lua adds a /completely new programming layer/ to typesetting besides the
> TeX macro layer.  And both layers are rather orthogonal to another.  Any
> formats should play nice with the Lua layer.  That is, they shouldn't
> force Lua package authors to target a particular format.

Lua package authors need a way to register a callback, and there can be
only one way per format... Also, suppose that somebody decides to write
a new luatexbase2 with a different interface, that would make things
really impossible to handle, and it couldn't be used in LaTeX nor Plain
without breaking everything, so no luatexbase2 can be written... So
there can be only one interface for packages supporting LaTeX and
Plain... what's the point of not including it in the format? Any other
format can include it too... Apart from philosophical questions which
seem to have 0 impact on the reality of luatexbase/LaTeX development, I
really see no downside in making luatexbase into the format. That was
its first purpose anyway.

> I think there are a few more formats in TL.  But anyway, while different
> formats satisfy different demands, they make up for fragmentation of the
> TeX ecosystem.  This is bad when the same typesetting problems are
> solved multiple times each solution targeting another format.

That's another discussion, but I think you're exposing the reason why
many people now use a monolitic ConTeXt instead of LaTeX where you have
dozens of more or less incompatible packages just to make a table...

> Ouch, I've not been aware of that.  That's certainly ConTeXt's fault and
> should be fixed on their side.

The developpers of ConTeXt are the developpers of LuaTeX, believe me,
they're perfectly aware of what they're doing.

> Come on, you are an experienced open-source programmer.  "Technically
> impossible" is a polemic term for "not yet implemented".  ConTeXt is
> buggy.  That's all.

Well, you should discuss that with Hans, I'm not interested in this
conversation as I've never compiled a single ConTeXt document. Right
now, ConTeXt is designed this way, and format-independant packages are
not possible, period. I've discussed this in length at the time where I
wrote luatexbase, and it took time for me to understand his vision. I
can't say I 100% agree with him, but I'm not in a position where I can
fork luatex and context to make them suit my view of what they should be.

Thank you,
-- 
Elie