[ppml] Directory Services - section 3.4.3

[Jeff Williams]

Owne and all,

  Good argument if your data/figures are accurate, and I cannot
say that they are not.  However your argument is based on the
now or near term, not the long term.  Things change quickly...

Owen DeLong wrote:

> --On Tuesday, June 14, 2005 10:53 AM +0100 Person who prefers not to be
> attributed wrote:
>
> >> >     >            - Via CDROM to users who complete the bulk data form.
> >
> >> This isn't vast overspecification. It's better described as a minimum
> >> level of service delivery.
> >
> > When you put a statement like that in policy then you
> > are implying that CD-ROM is the only acceptable media.
> > What happens if the data won't fit on one CD-ROM. Will
> > ARIN staff sit there swapping in CD-Rs, and then do
> > it all over again to verify that they are all readable?
> > If we simple say
> >
> No, it's not.  That would be true if there weren't a subsequent statement
> that OTHER MEDIA can be made available at ARIN's discretion.
> However, such a statement is present in the proposed policy, thus trumping
> any such implication.
>
> > -Via recordable media to users who have signed the bulk data AUP
> >
> > Then we are not overspecifying the details. ARIN is still
> > free to offer the data on a DVD-R with 5 copies of the data
> > in different formats for $100 or on CD-R with your choice
> > of a single data format for $2,000 dollars. The extra fee
> > covers the time spent sitting and swapping disks. People can
> > ask ARIN to buy a USB 2.0 or Firewire drive, load the 5 formats
> > of the database and mail it to them. Or people can drive up
> > to the ARIN offices with a PC containing a DLT drive and
> > 100baseT network card to do a high-speed SMB transfer of the
> > data onto tape.
> >
> Right... Then, users are free to expect any random form of media they
> choose.  The policy as written makes it very clear that ARIN will publish
> the data on CDROM and MAY publish it on other media upon request if ARIN
> feels such request is not overly burdensome.  The policy as written
> is better than your proposal, no matter how  many times you repeat your
> proposal.
>
> > These are all details which are not specified in the policy
> > and which are not possible if the policy mentions CDROM. That
> > is why that phrase is OVERSPECIFICATION.
> >
> That simply isn't true.  Under the proposed policy, all of those things
> are possible, but, the policy makes it clear that they are at ARINs
> discretion.
>
> >> Staff
> >> could probably figure out on their own if a CD-ROM made sense, but
> >> there's no harm in saying "else...CD-ROM".
> >
> > Yes, there is harm in saying "else...CD-ROM". By doing that we
> > usurp the ability of ARIN staff to use their own judgement.
> > We are micromanaging. Do you like your boss to tell you in
> > detail how to do your daily tasks? Policy is like instructions
> > from the boss and good policy should not overspecify in the same
> > way that good managers should not micromanage.
> >
> In my opinion, this simply isn't an accurate interpretation of the policy
> as worded.  The policy as worded guarantees that at a minimum, CD-ROM
> will be available.  CDROM is currently the most common form of mass-storage
> media.  As to the data not fitting, let's look at that realistically.
>
> The ARIN region now consists essentially of the united States and Canada.
> To the best of my knowledge, everything else has been shifted either to
> LACNIC or AFRINIC.  Let's overestimate a bit and say that the ARIN region
> contains 1/2 of all IP allocations.  ARIN does not allocate or assign
> smaller than a /22 (other than a statistically meaningless handfull of
> exchange point allocations and the swamp).  More than 3/4ths of ARIN
> allocations are /20 or larger.  There are roughly 16 million /24s, which
> translates to roughly 8 million /24s in the ARIN region.  1/4 of that is
> 2 million /24s which translate to approximately 0.5 million /22s.  The
> remaining 6 million /24s translate into  0.375 million /20s, so, a gross
> overestimation of the maximum size of the database is rougly 0.875 million
> records.
>
> I believe the average whois record is approximately 512 bytes, so, we
> have 875,000 records at 512 bytes is 448,000,000 bytes which is
> roughly 428 Megabytes.  A CDROM is 700 Megabytes.  You can fit almost
> two copies of the worst case database on a CDROM.
>
> The argument of what happens when the database exceeds a CDROM is specious.
>
> Owen
>
> --
> If it wasn't crypto-signed, it probably didn't come from me.
>
>   ------------------------------------------------------------------------
>
>    Part 1.2       Type: application/pgp-signature
>               Encoding: 7bit

Regards,

--
Jeffrey A. Williams
Spokesman for INEGroup LLA. - (Over 134k members/stakeholders strong!)
"Be precise in the use of words and expect precision from others" -
    Pierre Abelard

"If the probability be called P; the injury, L; and the burden, B;
liability depends upon whether B is less than L multiplied by
P: i.e., whether B is less than PL."
United States v. Carroll Towing  (159 F.2d 169 [2d Cir. 1947]
===============================================================
Updated 1/26/04
CSO/DIR. Internet Network Eng. SR. Eng. Network data security
IDNS. div. of Information Network Eng.  INEG. INC.
E-Mail jwkckid1 at ix.netcom.com
 Registered Email addr with the USPS
Contact Number: 214-244-4827