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

[Lester Caine <[e-mail]>]

On 12/12/14 21:22, Daniel Migault wrote:
> I am incline to understand that tagging tzid with version solves this
> issue has almost reached consensus. Now we should move forward to close
> this issue and agree on whether the version is optional with a default
> value and where the version is indicated.

The sticking point seems to be an acceptance that versions should be a
required since in addition to providing a solution to managing
historical data it also removes the race hazard in handling current
material. Once the mechanism is documented is there any reason to leave
an inherent race problem in the main distribution process? It may be
optional that a server provides the full range of versions however I
still feel this is only permissible if the protocol is documented
explaining the inherent race problem in the current sync process.

-----------------
I have a problem with the proposed conference call next week since I
will be driving down to London at that time, ready for a flight out to
Malta on the 19th. I think that we do have an understanding on why
managed versions are required for consistent operation and the agreement
is just required as to IF the race problem is something that is to be
avoided in the first draft or will have to come later? I think we had
already got agreement that anything done in day one has to reliably
support material published using tzdist, for which version is a
required. So it can't easily be optional in other processes?

-- 
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