At 16:06 18/02/2005, John C Klensin wrote:
--On Friday, 18 February, 2005 13:19 +0100 "JFC (Jefsey) Morfin" <[EMAIL PROTECTED]> wrote:
> At 03:52 18/02/2005, Martin Duerst wrote: >> The rest will be taken care by registrants, they'll figure out >> sooner or later what names are easy to type and what names are >> not easy to type. It has worked quite well for the ASCII-only >> DNS. > > Welcome to "phish and click" !!!
Which, as Martin points out, has worked quite well for the ASCII-only DNS.
And will work much better for IDNA which offers far more phishing and tricks opportunities.
Welcome to the IETF understanding of internationalization/multiligualization/vernacularization !!!
And you are the one who understand it the best....
Never asked: did you work with Doug Engelbart too?
jfc
PS. By the way where did you take that IDNA was not using an ASCII-only DNS?
That international/multilingual/vernacular layers are much simpler when using ASCII domain name does not imply that they do not exist and that you are permitted layer violations.
Or may be you mean that a far better way was to respect the DNS basic rules, not to introduce as left to right contradicting hierachy (IDN/DN) and management mess? IMHO we do not need to worry: this is what a grassroots process is going to impose on the long/medium?/short?? range if they want to avoid a balkanization of the root.
Let get real. I do prefer "name.tag--eu.ch" being read "name.ech" by my browser/userbox tables than a multilingual root being linked to the NTIA root through an "*" entry. Do you know what is the size of a virus table maintained every morning on my PC when compared with a TLD language transcoding table? 260 TLDs, possibly 500 scripts = 2000 ITLDs for a 10 chars entry average: 20000 chars.Good ASN.1 may reduce it. Compressed probably 10 K. I am sure every PAD Administrators will include it and the root file for free in their delivery. Sent as patches it probably takes 100 bytes a day.
