[ga] Re: On Its Way: One of the Biggest Changes to the Internet

[Stephane Bortzmeyer <bortzmeyer@xxxxxx>]

On Fri, Oct 12, 2007 at 12:45:52AM -0700,
 Karl Auerbach <karl@xxxxxxxxxxxx> wrote 
 a message of 71 lines which said:

> IDN names in and of themselves do pose a risk of increased buffer
> overrun issues - code will be seeing longer names than has
> frequently been the case.

I doubt it. The maximum length for names in the DNS has been fixed a
long time ago and is well-known, 255 bytes (63 bytes for one
label). IDN did not change that. I would be really surprised to see an
application broken enough to reserve less room for the names it
handles!

> Another issue caused by length of IDN names is that query
> interactions may have to shift from UDP to TCP

This one is not IDN-specific. Every DNS-using protocol invented in the
last fifteen years (DNSSEC, IPv6, SPF) use larger responses, in
average.

And the solution is also old and well-known: EDNS0, specified in
august 1999 and widely implemented (BIND has it for many, many years).

> Whereas today I can look at a traceroute printout and kinda grok the
> meaning of the host names on the path, 

You mean you don't use traceroute -n and whois? Shame on you :-)

> in an IDN world, it will look like gibberish.

This was a privilege only for english-speaking engineers. With IDN,
the playing field is more level.