Re: [Tzdist] Fwd: [tzdist] #32 (service): managing historical data

[Lester Caine <[e-mail]>]

On 10/12/14 17:32, Tim Parenti wrote:
> Where I /do/ see a potential race hazard is in defining exactly what
> value "changedsince" takes on.  Is it the time at which a tzdist
> publisher receives the new data, or begins distributing it, even if it
> was technically available to them hours or days beforehand?  Or should
> this date be fixed at the source (e.g., tz release timestamps) and
> propagate somehow throughout the tzdist system?

It has been stated already that "changedsince" is managed by the server
and so even for the same publisher a different server will give a
different value. So that can't be used to identify WHICH "changedsince"
matrial was used by the information source since it is more than likely
using a different source? IF "changedsince" was the published date for a
version ... but it's being able to access an older version that is the
problem here.

Of cause if we WERE defining a single set of data with a single update
sequence then things would be a little easier, and personally I will
rely on TZ and it's well defined versions for my own source of data to
publish. But no doubt other sources will keep their own slightly
different views of things.

-- 
Lester Caine - G8HFL
-----------------------------
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk