[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [wg-c] Recap from past threads...
>This is an excellent point. But, it is easier to rename than to
>renumber, by a bunch. I don't think SLD owners would hesitate to move to
>another TLD, on another registry, if there were such a beast.
Renumbering is tough for admins/individuals - renaming is tough on the rest
of the Internet. I just had to renumber DNS for 15,000 subscribers, and
despite a 100% smooth process, it went reasonably well because we were able
to do almost 100% of it internally. No easy task, but manageable. On the
other hand, I couldn't even begin describing the process of renaming to
these clients, never mind executing on it.
>> - TLDs can be held as IP until such time that the contract between the
>> registry is pulled by ICANN for non-performance or insolvency.
>This point has some serious legal chuck-holes.
How so? The intent of this summation (IIRC) was to raise the bar to ensure
that dabblers are very aware of the extreme consequences and gravity of
running a trademarked gTLD. For instance, IO might want to risk the IP on
.web, (or not), but what about .yahoo? In my mind, it is a relatively
elegant way of managing the additions to ensure a stable base of TLDs...
>> - that the number of TLDs a registry can operate must, at
>> least at first,
>> be limited in number but not so limited as to disallow the
>> incumbent NSI's
>> participation in the process.
>Not quite. I wasn't considering NSI participation at all. Why? They are
>already a TLD registry. In fact, they have three of them now. I rather
>doubt that they would fail to pass a solvency test and the SLA I had in
>mind uses NSI as the baseline model, and enhances from there.
Keep in mind this was a broad summary. The idea was that in order to have
maximum acceptance of these ideas, that the number of gTLDs that a registry
would be limited to managing would have to be at least set high enough so
as to ensure that NSI did not inadvertently fall afoul of the new
guidelines (it was recommended that the max be set to three tlds
initially - NSI currently manages at least five).
>Since then, these debates have convinced me that the emphasis needs to
>be on registries, not TLDs. That our fuzziness stems from confusing
>these different issues (TLD vs Registry). I don't believe that there is
>too much disagreement with this view.
I'm not sure about that. In my mind, the relationship between a registry
and a TLD is fleeting and incidental. We need to focus on one and be aware
of the impact on the other - linking them is pretty extreme.