Re: A ROUGH alternate proposal to CLNS DB for RIPE

["Wilfried Woeber, UniVie/ACOnet" <]

  Hi Sue, Tony,

>  * I'd like some pointers on where to go from here.
>  * I send this document because my questions and alternate
>  * formats provoked no additional comments.
>  *
>I guess from a RIPE DB point of view the next step needs for these new
>objects to be agreed and for a RIPE document to come out of this.
>What say the RIPE DB chairman. Wilfried ?

   	(BTW, Sue and Victor could you please distribute this discussion to
	 the appropriate lists @merit and @rare.nl, 
	 I deliberately shortened the Cc: list in order to save bandwidth)
  
  I've got to different opinions at the same time :-(
  The first one would say
  - we had a lengthy discussion in the RIPE-DB meetings, and we decided to
    implement the stuff according to the circulated proposal after minor
    improvement. The RIPE-NCC found no problem in implementing it.

  Unfortunately it turned out that there is room for considerable
  improvement (thanks to Sue's contribution!), thus my second opinion would
  say
  - as there is not yet wide-spread use of this object we should aim at the
    best solution that is economically feasable (implementation effort and
    quality of data). So let's go that way!
  
  Ok, now the technical part:
  I don't understand much of the game of NSAPs, RDIs, and the like :-( 
  Thus I'd really like the folks who initially proposed the object and those
  who want to use it to review the situation and come up with a good-quality
  proposal. I'd like to have it circulated well before the next RIPE-Meeting
  because I'd like to eventually have final consensus in May. 

  Henk, Victor, and the others who want to use it, please speak up!

>  *                       CLNS routing-domain object for the
>  *                                 RIPE database
>  * 				     Version 1.3B
>  *                                   Feb 1994
>...
>  * 	3) NETs host systems
>  *
>  * 	   The DNS for CLNS system is under development.
>  * 	   As a host.txt file was once used to bootstrap
>  * 	   the internet, so a iso_hosts file must
>  * 	   be generated to help boot strap the CLNP Internet.

  Right now I cannot clearly see the validity of this statement.
  I'd propose to provide an explanation why for example the use of IP-DNS or
  some other directory service is not possible to register hosts that are
  CLNP-capable. Also we would need a specification where, when, and how to
  convert the CLNS-hosts data in the RIPE-DB into an iso-hosts file.
  
  I think it is essential to describe the timeframe and the functional and
  resource limits (if any) of using the RIPE-DB for registering hosts and how
  to migrate the accumulated info afterwards to some form of CLNS-DNS or to 
  whatever.

  Again, you know, when it comes to CLNS I may be terribly wrong :-)
  
  Wilfried.
 --------------------------------------------------------------------------
  Wilfried Woeber                :  e-mail: Wilfried.Woeber@localhost
  Computer Center - ACOnet       :  
  Vienna University              :  Tel: +43 1 4065822 355
  Universitaetsstrasse 7         :  Fax: +43 1 4065822 170
  A-1010 Vienna, Austria, Europe :  NIC: WW144
 --------------------------------------------------------------------------