[GLOBAL-V6] Re: [GLOBAL-V6] Re: [GLOBAL-V6] Re: Válasz: Re: Válasz : [ G

  • To: Thomas Narten <narten at us dot ibm dot com>
  • Subject: [GLOBAL-V6] Re: [GLOBAL-V6] Re: [GLOBAL-V6] Re: Válasz: Re: Válasz : [ GLOBAL-V6 ] RE: How to reduce the junk applications?
  • From: Gert Doering <gert@Space.Net>
  • Date: Fri, 15 Feb 2002 00:28:03 +0100
  • Cc: Gert Doering <gert@Space.Net>, global-v6 at lists dot apnic dot net
  • In-reply-to: <200202142142.g1ELg1F10925 at rotala.raleigh dot ibm dot com>; from narten at us dot ibm dot com on Thu, Feb 14, 2002 at 04:42:01PM -0500
  • References: <20020212115606.V56841@Space.Net> <200202142142.g1ELg1F10925@rotala.raleigh.ibm.com>
  • Sender: owner-global-v6@lists.apnic.net
  • User-agent: Mutt/1.2.5i
    • Hi,
      
      On Thu, Feb 14, 2002 at 04:42:01PM -0500, Thomas Narten wrote:
      > Gert Doering <gert@Space.Net> writes:
      > 
      > > But for a metric to be useful, you have to be able to measure something,
      > > and use the result to judge something.   And IPv6 address usage is NOT
      > > useful to judge anything (mostly due to the /48 rule).
      > 
      > The draft document has the following words relating to utilization:
      
      I know what the draft says here, but I still claim that "number of used 
      /48s is not a useful metric to judge a network's size (or worthyness of 
      anything)".  Neither is "number of used IPv6 addresses".
      
      As a consequence, allocation criteria based on "number of /48s" are 
      not useful.
      
      Criteria are there to "keep people out".  Whom?  And who do we *want*?
      
      Do those criteria facilitate this?
      
      If the answer to that is "no", the criteria are wrong.
      
      [..]
      > Are you saying that this sort of utilization is "not useful"? 
      
      It is not useful to judge whether someone should get an allocation or not.  
      
      It *is* useful to figure out whether he needs a bigger block than 
      "default", or an additional bit later on.  This part is easy, because it's
      only math.
      
      The part about "who is 'in' and 'out'" isn't.
      
      > If so, what do you propose measuring instead?
      
      My proposal, modified by Arano-san, is "give everybody an allocation 
      that signs a paper stating he will re-distribute this to other entities"
      (and take it away if there is no hint in the proper registration database
      about him actually doing it).
      
      Of course one could lower the minimum number of /48s to something that
      will fit the "very few sites, but very big ones" networks - but then the
      whole number starts getting ridiculous, because anybody can bring together
      some friends, declare them "sites", hand them /48s, register them, and
      be done with it.
      
      Gert Doering
              -- NetMaster
      -- 
      Total number of prefixes smaller than registry allocations:   71887 (71770)
      
      SpaceNet AG                 Mail: netmaster@Space.Net
      Joseph-Dollinger-Bogen 14   Tel : +49-89-32356-0
      80807 Muenchen              Fax : +49-89-32356-299
      
      -
      - This list (global-v6) is handled by majordomo at lists dot apnic dot net