[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: New Internet Draft on registering IDNs

At 2:40 PM -0300 3/27/03, vinton g. cerf wrote:
the alternative of allowing the registrant to adjust the subset that is registered from the "bundle" has the nice property that it is the registrant's choice (and responsibility) for the resulting ease of use or lack thereof.

Indeed. Novice registrants could start with "let me handle the most likely name and block the rest", and then migrate to fuller use when they are ready. Within a few years, when registrant-level bundle handlers are common, users will probably mostly go towards all-resolving.

In thinking about this more, there are some downsides as well:

- Much more of a hassle for the registry because now they have to manage the individual elements of the bundle manually instead of automatically

- Much less predictable to DNS users of the zone

I am not sure what to say about the economic position to be taken and perhaps this could be best left to the TLD operator.

Right. RFCs usually don't talk about economics.

Excessive "greed" if you will pardon the use of the word, might prove to be a poor business choice, so there might be some balance between a single price regardless of the size of the selected bundle subset and a price equal to registering N distinct SLDs.

I think any registry reading this document (or the JET document, or others) will be able to quickly figure out the costs associated with the work they are taking on. We don't need to list it for them any more than BGP documents talk about the financial aspects of routing choices.

Intuition is hard to rely on here since the properties of different languages and chosen rules for "equivalence" will lead to quite a variety of different cases, I would think.

Which is exactly the reason this document is more generic than the JET document.

--Paul Hoffman, Director
--Internet Mail Consortium