Norbert Bollow wrote:
>
> Andrew Winkler <ckp@micro.lucent.com> wrote in response to a message
> from me:
> > > I disagree. The user says "Do you speak German?", not "Do you speak informal
> > > German?". Majordomo replies in formal or informal language, as configured by
> > > the majordomo-owner.
>
> > How is that better than saying "Ich spreche Deutsch", but also
> > elaborating by saying "for now I'll use the formal Swiss dialect, but
>
> Do you speak Swiss German (or at least German)? It appears from your
> comments that you don't...
My German is not as fluent as my English, Spanish or French, but not
as bad as my Russian, Hungarian, Greek, Aramaic, Akkadian, ... :-)
>
> 1) There is no such thing as a 'formal Swiss dialect'. Rather, you speak a
> Swiss dialect and while doing so, you choose to use formal or informal
> language.
Or, more compactly, "formal form of Swiss dialect", or even more
compactly "formal Swiss dialect"
>
> 2) When two strangers meet in a multicultural context, the first thing
> they do is try to negotiate a language for further communication. This
> means aking questions like "Do you speak German?". It is necessary to
> implement something like this in a program like Majordomo.
>
Exactly the sort of protocol I was suggesting. You ask the question in
some reasonable default, providing the user the option of specifying
a more precisely appropriate form. When the day is done, speech
production will allow and require such things as "address me formally,
in the Northeastern dialect of American English, using the diction and
pronunciation of an upper middle class white female with a college
education." Communication will happen between any two English speakers,
but every single one of the factors in that list, togher with a whole
host of more subtle factors, like timbre and intonation and emotional
color, has a substantial impact on the reaction of the hearer to the
"voice", and on the understanding/misunderstanding signal-to-noise
ratio.
A good protocol will look ahead to these facets that will clearly be
important within the next two years.
> 3) Choice of 'formal' vs 'informal' variations of a language or dialect is
> a separate issue, and much more complicated because it is very culture-
> dependent. For example, when a new neighbor moved in, I introcuded
> myself by saying "ich bin dae Norbert" which communicated three things:
> Namely, a) my name is Norbert, b) let's use informal language, and
> c) it would be nice if we were friends as well as neighbors.
> It was appropriate for me to suggest use of informal language, because
> our new neighbour is younger than I am. This is only a very simple
> example for showing that there may be complicated patterns of social
> interaction woven into negotiation of choice of 'formal' vs 'informal'
> language. It really makes no sense to try to implement anything like
> this into Majordomo. Let the choice of 'formal' vs 'informal' be
> configurable by the Majordomo-owner and that's it. We have enough
> creeping featurism in Majordomo already.
I'm not suggesting creating the whole protocol right now. I am
suggesting that the protocol should be designed for easy extension to
handle the (already well known and well understood) subtleties.
>
> -- NB.
Andy
Follow-Ups:
References:
|
|
