RE: IETF-51 WhoisFix BoF Announcement

>
>In the Whois BoF at IETF 49, three major uses of Whois were identified:
>
>     o Registries, registrars, etc. for domain names
>     o RIRs for IP allocations, ASNs, etc.
>     o RRs for routing policy data
>
>(Other people can probably categorize these more eloquently, but 
>hopefully the idea is clear.)  There are certainly other 
>applications, but this list feels right in terms of common uses on 
>the public Internet.
>
>It seems reasonable that the WhoisFix effort should attempt to 
>address all three of these uses, so certainly we should be thinking 
>about the RRs.
>
>I'll venture another legitimate use, but won't speculate about its 
>dominance relative to router config builds:  transfers of domain 
>names between ICANN-accredited registrars.

Wouldn't this last item be dealing with how to register at the database?

At least for me, whois is a query protocol. How you get the data into 
the database is not directly connected to whois. Some people use web 
forms, others provide e-mail interaction, ...

The handle uniqueness issue is also not whois proper, even though 
databases that store nichandles today are mostly accesible via whois.
However, even if some of them where to be moved to LDAP or other 
system, I would still like to see a common agreed format for a NIC 
handle such that I, or a program, would be able to tell where to look 
for the authoritative info for that handle. I look forward to the 
IETF standardising that in some way.

Other wishes I have seen going around lately are:

- dealing with non US-ASCII char sets (hints or clues to clients)
- referrals or other kind of help for proxy writers

If it turns out we can easily agree on a small set of simple 
extensions that fit (even if not perfectly) a big number of people 
then fine. Otherwise, please leave the poor thing alone.

Joao

--