Re: [Tzdist] TZ data publisher registry?

Tim Parenti <tim@timtimeonline.com> Sat, 24 January 2015 23:10 UTC

MIME-Version: 1.0
In-Reply-To: <54C2B673.4080602@lsces.co.uk>
References: <54C295CE.8040309@andrew.cmu.edu> <54C2B187.5030202@cs.ucla.edu> <54C2B673.4080602@lsces.co.uk>
Date: Sat, 24 Jan 2015 18:10:01 -0500
Message-ID: <CAFpi07xJWh9Avzz2ynv2T3kf44gnPvmR8BRiat_XPC9h71JwOA@mail.gmail.com>
From: Tim Parenti <tim@timtimeonline.com>
To: Lester Caine <lester@lsces.co.uk>
Content-Type: multipart/alternative; boundary=001a11c133703c7f8d050d6e0179
Archived-At: <http://mailarchive.ietf.org/arch/msg/tzdist/IX4k18SYttOB9_1rghY9FO0MFyU>
Cc: Time Zone Data Distribution Service <tzdist@ietf.org>
Subject: Re: [Tzdist] TZ data publisher registry?
Precedence: list

On 23 January 2015 at 16:00, Lester Caine <lester@lsces.co.uk> wrote:

> Except ... how does one flag if backzone is included or it's a truncated
> view of the data?
>

I think if we create a registry, we would need to implement namespacing in
some way.  So, if "tz" is used to refer to the IANA/Olson database, for
example, something like "tz-extended" could be used to refer to the data
including the backzone file, etc.

If we go that route, I think it only really makes sense for us to keep a
central registry at the top level; individual tzdist servers may wish to
slice and dice a given dataset in any way they please (various truncation
schemes being one such way) and should be allowed to file those subsets
under the corresponding top level publisher token.

--
Tim Parenti

[Tzdist] TZ data publisher registry? Ken Murchison
Re: [Tzdist] TZ data publisher registry? Paul Eggert
Re: [Tzdist] TZ data publisher registry? Lester Caine
Re: [Tzdist] TZ data publisher registry? Tim Parenti