Re: [idn] Presentation suggestion about "requirements" doc

["J. William Semich" <bill@mail.nic.nu>]

Dan:

I strongly support your proposed approach (below).

Thanks,

Bill Semich

At 06:41 PM 3/4/01 -0000, D. J. Bernstein wrote:
>I propose replacing ``Requirements'' with ``Considerations,'' and
>rephrasing the document accordingly. For example, instead of saying
>
>   The protocol MUST NOT require that the current DNS cache servers be
>   modified to support IDN.
>
>we can say
>
>   Does this IDN solution require revision and redeployment of DNS
>   caches? How difficult are the software changes? There are a huge
>   number of caches on the Internet, perhaps millions. Upgrading them
>   all will take time and effort, even if the software changes are easy.
>
>and instead of saying
>
>   It MUST make the minimum number of changes to existing protocols on
>   all layers of the stack
>
>we can say
>
>   How many protocols are changed by this IDN solution? How extensive
>   are the changes? For example, what changes are required in the SMTP
>   protocol? There is a cost to changing a protocol, beyond the cost of
>   changing implementations: namely, new protocol specifications must be
>   written and distributed.
>
>Readers should be able to look back at this document and understand what
>we were thinking.
>
>James Seng/Personal writes:
>> 2. If we can't nail down Requirements, it would be fairly difficult for
>> us to move forward with the next step, ie, comparison and protocol.
>
>It will be helpful to have a good framework for rational comparison of
>the IDN proposals. But the current Requirements document doesn't seem to
>have been written for this purpose. It appears to be an attempt to slip
>a bunch of bad decisions past the WG. It's missing the justifications
>that a legitimate document would have, and it imposes several premature
>restrictions upon the directions that the WG can go.
>
>---Dan
>
>