At 08:40 AM 2/16/99 +0200, Dr Eberhard W Lisse wrote:
>On Mon, 15 Feb 1999, William X. Walsh wrote:
>
>>
>> On 16-Feb-99 Dr Eberhard W Lisse wrote:
>> > > Oh, I agree with you entirely, unless, for instance, the chartering
>> > > authority were to take care of that for you by, say, issuing valid PGP
>keys
>> > > before a domain application was even possible.
>> >
>> > I don't follow. How would verification of the sender allow automatic
>> > determination of content?
>
>> Say for example, that some international accrediting group for Penguin
>> Research Scientists issues a PGP encrypted coupon for a domain name in
>> the .penguin TLD to each of their members. They would have to submit
>> this coupon along with their domain application to validate that they
>> "belong" to this group and thus meet the charter.
>
>That's exactly what I am saying. Running a template through a parser is
>trivial. Even I can hack something gross which does glue records.
>
>And whether someone at the registry decides, or someone elsewhere, if 10
>million Penguin huggers want domains, you run into the delay.
Yes, and I can show you a 4-way cluster of quad Alpha 600Mhz SMP boxes,
with Fibre channel connected hardware RAID5, that will handle most things
that you can throw at it. At mhsc-systems.com we call it a 4x4. What's
more, these clusters can be connected in super clusters. Most problems
these days can be solved by throwing hardware, software, or both at it. In
fact the biggest problem for such a meta-cluster is connecting it with
enough bandwidth. If a TLD registry has this much load, then they can
easily afford the hardware/software else they had better consider altering
their non-profit status.
BTW, also consider that charters can be contrived to be more amenable to
automated enforcement techniques. Like I said earlier, "careful crafting of
charters".
>> My problem with all of this is that we are limiting the use of a string
>> to particular purpose. .penguin could also be used for other purposes,
>> and to restrict the use of that string to a particular purpose ONLY
>> basically strikes me badly.
>
>I am purely looking at the implementation. If implementation is not
>possible, we can stop wasting time and insults on the issue.
I don't think implementation is the major issue at the moment.
>> I've yet to see an example of it that isn't wrought with problems.
>> Someone said .doctor be administer by the AMA, but they means that other
>> "doctors" who are not medical doctors are deprived of the use of a
>> string that also identifies them. Now, say we had .doctor, we could
>> delegate md.doctor to the AMA and they could then issue names as they
>> please.
>
>The AMA, by the way, can only speak for its members, a few American
>medical practitioners.
>
>
>el
>
___________________________________________________
Roeland M.J. Meyer -
e-mail: mailto:[EMAIL PROTECTED]
Internet phone: hawk.lvrmr.mhsc.com
Personal web pages: http://staff.mhsc.com/~rmeyer
Company web-site: http://www.mhsc.com
___________________________________________________
KISS ... gotta love it!
