On 7/22/14, 11:35 AM, Pete Resnick wrote:
John and I talked today about how to solve our impasse with the PRECIS framework document. I think we have reached a mutually tolerable (note that I didn't say "agreeable") path forward:
Thanks to you both for putting your time and energy into finding a solution.
The key issue for John is the differences between IDNA and PRECIS rules.
Clarifying question: (a) the borrowing of IDNA rules in the PRECIS framework, including possible synchronization issues or (b) the additional rules that PRECIS defines above and beyond the borrowed rules, and the interaction between the PRECIS-specific rules and the existing IDNA rules?
I think that the PRECIS registry must not include the IDNA rules. Instead, the PRECIS framework document needs to point to the IDNA rules by reference - we included them in the document only for ease of reading, but if it would be less confusing to simply reference them then I'm fine with that!
If a change to Unicode occurs, the Designated Expert(s) for IDNA and PRECIS (which will likely be the same person(s)) will have to separately figure out the impact of the change to IDNA and PRECIS and update the rules (and tables) for each accordingly. Without guidance in the document as to what those differences are, the job of updating becomes significantly harder.
Yes, let's avoid that fate.
So, the thing we can do (short of restructuring the document) that will address John's concern is to add some prose that describes how each of the PRECIS Classes differs from the IDNA rules so that a Designated Expert who is looking at any Unicode change can figure out what impact it has on each of the rule sets in one pass and can update them accordingly. That seems to me a perfectly reasonable approach, and will unstick this document in a way that John can live with and will give other protocols, which are now waiting, *something* to refer to. Thoughts? Comments?
That does seem like a tolerable path forward. IMHO ideally we'd have one registry of Unicode-based rules, but this is the next-best thing.
BTW I am not averse to restructuring the PRECIS framework document if necessary.
And I am still committed to addressing the other feedback that John provided. All of it was extremely helpful, but it takes focused time to address such feedback and I've had very little of that lately.
Peter _______________________________________________ precis mailing list [email protected] https://www.ietf.org/mailman/listinfo/precis
