Things that need to be addressed
Bill Darte
billd at cait.wustl.edu
Tue Nov 7 09:24:15 EST 2000
ARIN policy needs to be directed toward real problems....I have seen no
justification for this policy except that it wastes addresses.....but the
magnitude of that waste must be assessed before the issue rises to level of
a need for ARIN intervention.... I reiterate my earlier questions...
1. How many addresses ARE being used this way now and
2. what is the slope of the the utilization curve and
3. What are the potential impacts upon that utilization rate in the next 3-5
years???
Scott Marcus' assessment of the AS problem is a crystalline example of the
kind of analysis that should be used and those will motivate action in an
appropriate way.... Discussion in the absence of clear metrics motivates
speculation and endless argument. Granted, coming up with the numbers for
this 'problem' are likely harder to come by, but the attempt must be made in
order to substantiate the need for policy.
My last question is how can we get a handle on the extent of this problem?
Who has this information and will they objectively divulge it?
Bill Darte
AC
> -----Original Message-----
> From: Steve Pierce [mailto:Steve.Lists at HDL.com]
> Sent: Monday, November 06, 2000 6:16 PM
> To: ARIN List
> Subject: RE: Things that need to be addressed
>
>
> you wrote:
> >> ARIN simply cannot base its allocation policies on the actions
> >> of other organizations.
>
> But we must. In fact, we are already forced into making new
> regulations
> about IP address blocks because other organizations do not
> want to give up
> their un-needed or under-utilized IP blocks. So ARIN is most
> definitely
> already basing policies on the actions of other organizations.
>
> You might be taking a purely logical approach rather than
> also considering
> human nature. For the policy to be effective we need to do
> both. When you
> create a policy that makes it difficult for someone to make a
> living, they
> will, by human nature, do whatever it takes to get around that policy.
>
> For example. You set the speed limit at 25MPH on a 4-lane
> divided highway.
> Yet for some reason everyone drives 55. Perhaps because there
> is plenty of
> line of sight vision, smooth pavement and little traffic
> incursions. The
> scientific approach would be to say, our policy is to drive
> 25MPH, so we
> need to step up enforcement to make sure that everyone drives
> 25. But it
> fails to look at human nature. If people feel safe on a road,
> they naturally
> drive faster.
>
> Thankfully traffic laws in most states don't work an
> inflexible policy. The
> 90 percentile rule is applied. That is, if 90 percent of the
> drivers are
> driving 55, then the speed limit is changed to 55. In fact in
> California, if
> you are clocked at a speed that is at or below the 90th
> percentile and also
> below the stated maximum for that type of road, you cannot be given a
> citation. Even if the posted sign says Speed Limit 25 MPH.
>
> So we can make a policy that says virtual hosts cannot request address
> space. But then the hosting companies will come up with
> another reason to
> get around that rule or risk going out of business. If we
> make a rule that
> says virtual hosts can't have address space while their
> legacy competition
> is flush with addresses, then despite the well meaning policy of ARIN,
> people will do whatever it takes to get additional address space.
>
> I am not saying that just because people will try to
> circumvent policy, we
> should just throw our hands up and do nothing. Far from it. I
> am saying that
> since we can predict that people will go around it, we need
> to make sure the
> policy is reasonable enough so as to minimize the temptation
> to circumvent
> the policy. Remember, I am for ip-less addressing as long as
> we can solve
> the problem with address blocking and filters. But I think it
> is naive to
> think ORBS and Net Nanny are interested in ARIN's problem and
> are willing to
> change they way they work to accommodate ARIN.
>
> I already have experience working with ORBS and Net Nanny to
> get specific
> addresses removed from their block lists. In over two years
> not once have
> they removed an address. Even when a law suit was filed in
> one case, they
> claimed it was their free speech right to list whatever
> address they wanted
> to and they weren't forcing people to use or subscribe to the filters.
>
> Yet here in Michigan, Public Schools and Public Libraries are
> required by
> State Law to install net filters. If ARIN is successful in getting a
> commitment from ORBS and Net Nanny to establish reasonable
> guidelines for
> the removal of blocked addresses, then by all means lets move
> forward with
> IP-less address requirements for virtual hosts. But my
> experience has shown
> that these filter companies are not interested in any oversight or
> co-operation. So perhaps we need another plan.
>
> Cheers!
>
> - Steve
>
> Steve Pierce, HDL
> Direct: (734) 482-9682
> mailto:Steve at HDL.com | http://HDL.com
>
>
>
> -----Original Message-----
> From: owner-vwp at arin.net [mailto:owner-vwp at arin.net]On Behalf
> Of Alec H.
> Peterson
> Sent: Monday, November 06, 2000 6:35 PM
> To: Steve Pierce
> Cc: vwp at arin.net
> Subject: Re: Things that need to be addressed
>
>
> Steve Pierce wrote:
> >
> > ARIN introduces the problem because of the restriction of addresses.
> > Blocking and filtering is not a problem if each domain has
> a separate IP
> > address. But with ARIN forcing the change to ip-less web
> hosting, a change
> I
> > fully endorse, ARIN needs to realize the full impact of
> that change. When
> > you go to ip-less domain hosting, the limitation of the
> filter programs
> like
> > ORBS and Net Nanny cause undue hardship to those abiding by
> ARIN's policy.
> > AS much as we would like to have ORBS and NetNanny change
> their ways, ORBS
> > dares anyone to sue them and NetNanny refuses to disclose
> their filtering.
> > Both argue that their filtering software is a free speech.
> But with state
> > and federal governments no mandating filtering software in
> schools and
> > libraries, ARIN is unfortunately forced to having to deal with this
> problem.
> > ARIN could use the head in the sand approach, but all that
> will do is
> force
> > carriers to lie about address space needs.
>
> MAPS/ORBS dares people to sue them over listing practices relating to
> spamming and relaying.
>
> With name-based hosting, MAPS/ORBS could theoretically be black holing
> people who have done NOTHING wrong even by their standards.
> They certainly
> do not want to be listing addresses that are following all of
> their own
> rules.
>
> ARIN simply cannot base its allocation policies on the
> actions of other
> organizations. While many people (including myself) will readily
> acknowledge the issues raised above, they cannot be used as
> reasons in and
> of themselves for us preventing a policy allocaiton. As I
> said before, we
> need to list the issues that will be raised, and see what we can do to
> address them. Simply stating that they exist as a reason for
> not using
> name-based hosting is not enough.
>
> Now, it is entirely possible that MAPS/ORBS and NetNanny
> could tell us to
> pound sand, in which case we need to figure out what else we
> can do about
> these issues. But we need to try.
>
> >
> > The ISP will just lie to create false justification for
> address range
> which
> > defeats the whole purpose of ARIN mandating the move to IP-less
> addressing.
> > There is nothing to stop an ISP from saying, every domain
> has their own
> SSL
> > certificate as you know SSL must be tied to a unique IP
> address. So they
> > then justify 20,000 IP addresses. Are we going to force
> ARIN to determine
> > when a SSL cert is real or faked? It creates an untenable
> situation for
> > ARIN.
>
> Stating that ISPs will just break the rules so there is no
> point in trying
> is again not an acceptable attitude to take when developing a
> policy. If we
> are going to use that logic then it seems silly for ARIN to have any
> policies to begin with. ARIN is not doing this to make
> enemies, ARIN is
> doing this to help make everybodys' lives easier in the long
> run. That is
> why ARIN is working with the people it will affect.
>
> >
> > The problem comes about in the unfair allocation of
> addresses. Bigger
> hosts
> > with class A addresses to burn can easily offer IP address
> to every single
> > domain they host with no fear of running out of addresses.
> So they are not
> > impacted by these filter programs. But smaller hosting
> companies that
> follow
> > ARIN's policy are at a competitive disadvantage.
>
> If these large ISPs who you claim are wasting so much address
> space really
> are doing this, then eventually they will run out. And when
> they do it will
> be up to them to justify their existing allocations.
>
> >
> > ARIN policies should not create an environment where the
> smaller guy can't
> > compete. Especially when it is the smaller guy that is
> following ARIN
> > guidelines and is so carefully conserving IP address space.
> That is the
> guy
> > ARIN should support and encourage, not make it more
> difficult for them to
> > conduct business or force them to lie to stay in business.
> >
> > If ARIN is going to require the move to ip-less addressing
> for web hosts,
> > then ARIN also needs to force holders of Class A address
> space to give
> back
> > Class B and C addresses and those holding Class B should
> not be permitted
> to
> > hold Class C's. At least one world wide carrier has at
> least two if not
> > three Class A addresses yet they also have hundreds of
> Class C addresses.
> > That should not be permitted.
>
> This has been disucssed before. Reclaiming legacy allocations is a
> completely separate discussion, and has absolutely no bearing on what
> happens with this policy. This is an active issue within
> ARIN, and you are
> more than welcome to put forth solutions on how ARIN can reclaim
> under-utilized legacy allocations.
>
> HOWEVER, that is a separate discussion, so please confine
> comments on this
> list to what should be done about the virtual hosting policy.
>
> Alec
>
> --
> Alec H. Peterson - ahp at hilander.com
> Staff Scientist
> CenterGate Research Group - http://www.centergate.com
> "Technology so advanced, even _we_ don't understand it!"
>
More information about the Vwp
mailing list