Proposal: Language code "de-DE-trad"

Håvard Hjulstad
Mon, 11 Feb 2002 16:48:03 +0100

You obviously have some very good points. It will be extremely difficult to
determine in each case which form is the "default" form. I may very well be
wrong in some of my statements concerning German. I tried, however, to look
at the principles. I think these principles may be applicable for very many

Many/most languages have "default forms". These defaults do change over
time. I think that it in most cases is better that a newer form is the
default, and that we need to specify previous forms as deviations. May be
German isn't a good example, but history gives us a number of examples of
language reforms that have met a lot of resistance, but that have been
implemented amazingly quickly. (I am making no predictions for German!)

I also am uncertain whether "trad" is the best choice of specification (if
the specifiec "non-1996" is needed). I think that dates (years) are better
suited. But may be this also doesn't work for German.

I do not wish to get involved in the German discussion. My input was
intended for the discussion of principles.

Best regards,
Håvard Hjulstad

-----Original Message-----
From: J.W. []
Sent: 11. februar 2002 16:20
Subject: RE: Proposal: Language code "de-DE-trad"

On 11 Feb 02, at 11:00, Håvard Hjulstad wrote:

> I find this discussion most interesting. Language tags are proposed to
> capture the German orthographic revision. I can obviously see the need.
> question is how it should be solved.
> (I am the convener of ISO/TC37/SC2/WG1 and the Project Editor of ISO 639-1
> (FDIS to be circulated this month I have been told), and I am very much
> interested in finding good ways to represent language variation. We are
> going to propose a project to extend the 639 standard series to include
> issues.)
(I omit your example here. I hope this is acceptable)

> We need to look carefully at what is default and what is "marked". It
> be tempting to refer to "de-DE-1996" (or whatever the correct year should
> be) as the default German for current language technology implementation.
> Deviations from the default need to be indicated.

Alas, de-DE-1996 does _not_ provide a good default.

Most german literature, especially the classics, is written in what is
proposed as de-DE-trad, and many contemporary artists oppose the
Also, this revision is opposed by a majority in the population. A poll made
in 6 major newspapers in 200 resulted in 98% voting to take back the
revision, or to ignore it.


The results of similar polls are not that extreme, but most above 75% and
_all_ clearly above 60% in opposing the revision.


Pressure from official sides may or may not change the use in the
population in the long run, but IMHO the next decades the traditional
orthography will be the default.

There is also a political background which is not important, but should not
go unnoted, as it makes the revision a more emotional topic - the 1996
revision bears some resemblances to that propagated by the Nazi
goverment (and taken back at end of World War II). While this may not
be an objective reason to oppose the revision, any emotional
ressentiment rooting here will go deep, and last.

Book "Rechtschreibreform und Nationalsozialismus", ISBN 3-89244-450-1

Also, a part of Germany (Schleswig-Holstein) does officially not accept
the new orthography.

I don't think these are good indications for declaring "de-DE-1996" the

> The fine analysis means that as defaults change, the actual value of an
> unmarked language identifier (like "de", "en", "nb") changes over time. If
> we don't accept that, unmarked language identifiers cannot be used at all.

I agree here.

IMHO, the unmarked language identifier should denote the language in
common use, including and allowing for variations. More specific tags
indicate their user is aware of the differences, and chose intentionally.
The unmarked tag indicates the user does not care, or does not know
about the more specific tags, or cannot tell.
This way, erroneous tagging is reduced and the unmarked identifiers in
use as default.

> I should think that the following approach would probably work in most
> for the development of languages through time: The lastest approved form
> something is considered the default and is unmarked.

This I oppose.

Approved by whom? Goverment, population, IETF?
What about countries where the official language as set by goverment is
different to that spoken by the population?
We're getting on dangerous grounds here, I am afraid.

While I am interested how such problems were resolved in the past here
and with ISO 639, I would prefer if we could avoid political aspects as far
as possible. (Saying that, I have introduced two such aspects myself in
this discussion.)

> Previous forms should
> be marked with the time when it first was approved or taken into use
> (actually: when that form became the default replacing a previous form).
> Håvard

I am afraid this will often be hard to determine, and only with the benefit
of hindsight. What exactly is a default in orthography? Do we need
sociological studies here? How to avoid politics?

I don't know whether the generalized solution you are aiming for is
possible. Determining closer the meaning of the unmarked language
identifier is possible, I hope, but I don't think it can be affixed to a
specific tagged variant.


Johannes Wilkes, Bonn, Germany