pinyin (and wadegile) request has gotten derailed

Broome, Karen Karen_Broome at
Wed Sep 17 14:29:11 CEST 2008

I think this is a very key point and one on which BCP 47 may not provide a clear answer. Are overlapping tag semantics allowed?

To use an example from an earlier request, if one user needed a specific tag for the 1959 "academy" variant of Belarusian and another user required the Academy variant with no respect to the various revisions, could both tags be assigned?   I think your analogy with respect to Mark Twain is quite relevant and I believe this is why the last few requests have been delayed. Attempting to come up with a perfect mutually exclusive tagging mechanism for variants may ultimately not serve the needs of users.

I think we need to accept that overlapping semantics in variant tags is allowed while still keeping an eye out for tags which can be shared across languages.


Karen Broome

From: ietf-languages-bounces at [ietf-languages-bounces at] On Behalf Of Phillips, Addison [addison at]
Sent: Tuesday, September 16, 2008 4:33 PM
To: CE Whitehead; ietf-languages at
Subject: RE: pinyin (and wadegile) request has gotten derailed

I note that Mark has requested a subtag for explicitly Hanyu Pinyin and not for any other kind of Pinyin. While it is possible for Hanyu (other) pinyin's to be included into a single subtag, I can't tell if that would meet Mark's requirements or not. I would be very astonished and not all that well served if I were to request a subtag 'twain' [to identify the dialect usage of writer Mark Twain] and got instead '1880' [to identify dialectical usages of late-19th-Century American writers] instead. Yes, one encompasses the other. But that doesn't necessarily meet the requirements I am trying to address, and, in this case, which we, as a group, might not fully know.

Mark, as the requester, do you have an opinion or requirements related to this?


Addison Phillips
Globalization Architect -- Lab126

Internationalization is not a feature.
It is an architecture.

> -----Original Message-----
> From: ietf-languages-bounces at [mailto:ietf-languages-
> bounces at] On Behalf Of CE Whitehead
> Sent: Tuesday, September 16, 2008 4:08 PM
> To: ietf-languages at
> Subject: pinyin (and wadegile) request has gotten derailed
> Hi, John, Randy, others:
> Thanks for the information; I think I'm agreed with John here that
> Cantonese Pinyin should not be included in the subtag [pinyin] but
> that Tibetan Pinyin can be--and optionally so can Tongyong Pinyin.
> --C. E. Whitehead
> cewcathar at
> John Cowan cowan at
> Tue Sep 16 15:05:27 CEST 2008
> > Randy Presuhn scripsit:
> >>>  1.  How different is the Tongyong Pinyin romanization (the
> alternate
> >>> Taiwanese romanization) of Mandarin from the Hanyu Pinyin
> romanization
> >>> of Mandarin?
> >>
> >> About as different as Pinyin and Wade-Gile.
> > Quantitatively, no.  Of the 412 Modern Standard Mandarin
> syllables
> > (disregarding tone), Hanyu Pinyin and Tongyong Pinyin differ in
> the
> > spelling of only 81 of them, whereas HYPY and Wade-Giles differ
> in the
> > spelling of 270 of them.  The only things that TYPY and WG have
> in common
> > are that they don't have the features, peculiar to HYPY, of
> separating
> > the alveolopalatals by writing different initials and of eliding
> central
> > vowels in certain triphthong rhymes.
> Thanks for the info!
> >>> 2.  And how different is Tibetan Pinyin from Hanyu Pinyin?
> > By contrast, Tibetan Pinyin is as like HYPY as the nature of
> Tibetan
> > will allow it to be.
> That was my understanding
> >> I see no problem with having to use a meta-content description
> tag
> >> to distinguish Tibetan from Hanyu pinyin until [cmn] and other
> codes
> >> become available)
> > I don't see what 'cmn' has to do with it.  Tibetan is in no way
> part of 'zh'.
> My mistake--I was thinking of Tongyong Pinyin which is a part of
> [zh] but [cmn] won't help there either!
> > The whole point of language tagging is distinguishing languages
> and
> > their important variants.  To do things which would obscure the
> > distinction between two indisputably distinct languages would be
> > counterproductive.
> > The various Soviet Turkic languages are also indisputably
> distinct,
> > but we managed to assign a single tag to cover the Jangalif
> orthography
> > for all of them.
> That's the example I thought of.
> >> (Alas, since Tongyong Pinyin and Hanyu Pinyin are both for
> Mandarin,
> >> ISO 639-3 codes will not help to distinguish these two if they
> need
> >> to be distinguished with something other than a description in a
> >> meta tag.  That's the only problem I can foresee with lumping
> the
> >> two together for now.)
> >
> > This is a strong argument for *not* merging the two.
> The obvious tactic is zh-(Latn)-TW-pinyin vs. zh-(Latn)-CN-pinyin.
> Admittedly, zh-TW has been used to mean zh-Hant in the past, but
> this
> is a truly appropriate use of TW for the orthography specific to
> Taiwan.
> This is problematic though as both Hanyu Pinyin and Tongyong Pinyin
> seem to be used in Taiwan--but it would serve to distinguish the
> two official orthographies.
> Another option I thought of:
> [pinynprc]
> [pinyntwn]
> or [pinyntai]
> but these are totally obscure and non-transparent; however an
> option is t o use [pinyin] for all similar Romanizations used in
> the People's Republic of China and use some other name (such as
> [tongyong]?) for the other variety.
> >> Reading online I understand that Cantonese Pinyin is quite
> distant
> >> (correct me if I'm wrong) from Hanyu Pinyin --
> >
> > Yes.
> > I agree.  This should not be tagged with the wide-scope 'pinyin'
> subtag,
> > although I suppose it sometimes will be.
> Thanks for the reply; agreed!
> --
> > John Cowan   cowan at
> _______________________________________________
> Ietf-languages mailing list
> Ietf-languages at
Ietf-languages mailing list
Ietf-languages at

More information about the Ietf-languages mailing list