At 12:46 PM 6/19/00 -0700, William X. Walsh wrote:
>DJB> s/your/his and the STANDARDS community's/
>
>s/STANDARD community/overly rigid/ :)
Standards are MEANT to be rigid. That's why they're called "Standards" and
not "rules of thumb".
>DJB> Agreed. You're more than free to design whatever incompatible systems
>you
>DJB> want. However, OpenSRS should be releasing code that conforms to the
>DJB> established standards. If you - on your own - want to write something
>that
>DJB> is technically broken and call it 'a-ok', that's your own look-out.
>
>The issue here is what the OpenSRS server will accept. The code
>should be the broadest possible implementation of what the server will
>accept, and if someone wants to modify it to be more restrictive than
>what the server will accept, they are free to. If the server will
>accept it, the code should reflect that.
No, exactly the opposite. That data will be used by more than just SRS, but
also by the Internet community at large. For maximum compatibility, it
should be resricted to ONLY standards-compliant data.
>Now the final question, will OpenSRS let the server accept those
>addresses? I believe that someone already said yes, but I may be
>wrong.
If they do, then the server should be corrected. A domain-registrar should
not be playing "fast and loose" with standards. If you want to play fast
and loose, that's your business, but the data that a domain registrar is
presenting to the Internet should ALWAYS abide by that Internet's
established standards, and that means following RFC822 to the letter.
D
This archive was generated by hypermail 2.1.3 : Tue Oct 19 2004 - 23:35:38 EDT