Re: Lightly used IP addresses

[Chris Grundemann]

On Sat, Aug 14, 2010 at 22:24,   wrote:
> Psst.. Hey.. buddy. Over here... wanna score some gen-yoo-ine Rolex integers, 
> cheap?

Right, because there is no reason to care about the uniqueness of
integers used on the Internet... :/

~Chris

participation in process (Re: Lightly used IP addresses)

[Mikael Abrahamsson]

On Sat, 14 Aug 2010, Chris Grundemann wrote:

I highly encourage everyone who has an opinion on Internet numbering 
policy to do the same.


The same goes for IETF and standards, there one doesn't have to go to 
meetings at all since most work is being done on/via mailing lists openly.


--
Mikael Abrahamssonemail: swm...@swm.pp.se

Re: Lightly used IP addresses

[David Conrad]

Owen,

On Aug 14, 2010, at 8:40 AM, Owen DeLong wrote:
> Let's clarify the definition of abuse in this context. We are not talking 
> about people who use their IPs to abuse the network. We are talking about 
> resource recipients who use their allocations or assignments in contravention 
> to the policies under which they received them (and thus contrary to the RSA 
> which they signed when they received them).

The challenge ARIN (and to a lesser extent, the other RIRs) faces is that in a 
very short time, we're going to have a system in which there will be folks 
barred from entering a market because they signed an RSA while at the same 
time, there will be others who will act without this restriction.

I honestly don't see how this system will be stable and instability breeds all 
sorts of things (some perhaps positive, most probably negative). When resources 
were plentiful this dichotomy could be mostly ignored.  Resources are soon not 
to be plentiful.

It has been depressing to watch participants in ARIN (in particular) suggest 
all will be well if people would just sign away their rights via an LRSA, move 
to IPv6 overnight, abide by increasingly Byzantine rules, accept that folks 
were always under ARIN's policies and they just didn't know it, etc. 
Pragmatically speaking, it seems the most likely to be successful way of 
maintaining stability with the impending resource exhaustion state is to give 
up pretenses of being a regulatory agency and concentrate on the role of being 
a titles registry.  I figure if the existing RIRs don't do it, someone else 
will.

But perhaps I'm missing something since I too gave up on PPML some time back.

Regards,
-drc

Re: Lightly used IP addresses

[Doug Barton]

On 08/14/2010 21:24, valdis.kletni...@vt.edu wrote:

On Sat, 14 Aug 2010 17:03:59 MDT, Chris Grundemann said:

First, in this thread we are not talking about folks who have not paid
ARIN their dues, we are talking about folks who "sell" addresses
despite not being authorized to do so by ARIN - aka abuse/fraud.


Psst.. Hey.. buddy. Over here... wanna score some gen-yoo-ine Rolex integers, 
cheap?


... only if they're prime.

--

Improve the effectiveness of your Internet presence with
a domain name makeover!http://SupersetSolutions.com/

Computers are useless. They can only give you answers.
-- Pablo Picasso

Re: Lightly used IP addresses

[Valdis . Kletnieks]

On Sat, 14 Aug 2010 17:03:59 MDT, Chris Grundemann said:
> First, in this thread we are not talking about folks who have not paid
> ARIN their dues, we are talking about folks who "sell" addresses
> despite not being authorized to do so by ARIN - aka abuse/fraud.

Psst.. Hey.. buddy. Over here... wanna score some gen-yoo-ine Rolex integers, 
cheap?


pgpCx8dNx9RqZ.pgp
Description: PGP signature

Re: Lightly used IP addresses

[John Curran]

On Aug 14, 2010, at 11:30 PM, Patrick Giagnocavo wrote:
> 
> Question:  Why does it cost $11 million or more per year (going to some
> $22 million per year after 2013) to run a couple of databases that are
> Internet-accessible?

Patrick - If this is a reference to ARIN, the budget is approximately $15M
annually, and is not substantially changing any faster than expected for 
normal cost-of-living trends (If $22M is a reference to having both IPv4 
and IPv6 fees, ARIN charges each organization only once for the larger of 
IPv4 or IPv6 registration services fee it makes use of)

Even so, it's a fair question to ask why it costs $15M annual to run ARIN.  
That includes the costs for many tasks which might not be obvious, including 
running the legacy registry system (which handles SWIP email templates), the 
new ARIN Online system (which is quite a bit more elegant), the public WHOIS 
servers, bulk WHOIS and FTP services, IN-ADDR services, the public web sites, 
the polling & election systems, the billing/invoicing systems, and the staging,
development/QA support for same, and the normal office infrastructure for 
things 
like email, mailing lists, replication, business record keeping, and archival.
There's some engineering staff to keep all that  running, registration services 
staff to handle incoming requests, member services for running the meetings, 
elections, and policy process, and outreach thats already been mentioned with 
respect to trade shows and press, but also includes engagement with our friends 
at the ITU, international bodies, and governments.  The full budget is 
available 
in each year's annual report along with the audited financials, and can be found
here:  https://www.arin.net/about_us/corp_docs/annual_rprt.html 

Clearly, the budget can be increased or decreased based on the services desired 
by the community, and this typically discussed on the last day of the ARIN 
Public
Policy & Member meeting (twice yearly) during the Financial Services report.  
In 
between meetings, this topic is probably best suited for the arin-discuss 
mailing 
list as opposed to the nanog list.

FYI,
/John

John Curran 
President and CEO
ARIN

Re: Lightly used IP addresses

[Patrick Giagnocavo]

Randy Bush wrote:
>> >> John - you do not get it...
> >
> > vadim, i assure you curran gets it.  he has been around as long as you
> > and i.  the problem is that he has become a fiduciary of an organization
> > which sees its survival and growth as its principal goal, free business
> > class travel for wannabe policy wonks as secondary, and and the well-
> > being of the internet as tertiary.  they're just another itu, except the
> > clothing expenses are lower and the decision making process pretends to
> > be more open, but isn't.
> >


Question:  Why does it cost $11 million or more per year (going to some
$22 million per year after 2013) to run a couple of databases that are
Internet-accessible?

--Patrick

40 x /18's and an ASN - was Re: Lightly used IP addresses

[Jeffrey Lyon]

The vendor I referred to earlier that does not support IPv6 explained
this in a private meeting, not a sales pitch. We already use their
products extensively. The discussion was more to the tune of "we
developed IPv6 support but stopped including it in the firmware
releases because no one was using it."

I informed them that we would use it so possibly by EOY we can have
IPv6 support (note: I don't know if Telia and BandCon even support
IPv6 yet? Yet another hurdle.)

Jeff


On Sun, Aug 15, 2010 at 7:04 AM, Frank Bulk  wrote:
> This week I was told by my sales person at Red Condor that I'm the only one
> of his customers that is asking for IPv6.  He sounded annoyed and it seemed
> like he was trying to make me feel bad for being the "only oddball" pushing
> the IPv6 feature requirement.  I tried to explain to him that by this time
> next year IANA will likely have handed out all their IPv4 blocks and that I
> didn't have the time spend the first half of 2011 implementing IPv6 across
> my $DAYJOB network, but wanted to spread that work over time.  To his
> credit, it's been on their to-do list for at least 6 months if not a year,
> it's just been pushed back several quarters.
>
> Frank
>
> -Original Message-
> From: Owen DeLong [mailto:o...@delong.com]
> Sent: Saturday, August 14, 2010 10:27 AM
> To: Jeffrey Lyon
> Cc: John Curran; nanog@nanog.org; Ken Chase
> Subject: Re: Lightly used IP addresses
>
>
> On Aug 13, 2010, at 9:12 PM, Jeffrey Lyon wrote:
>
>> John et al,
>>
>> I have read many of your articles about the need to migrate to IPv6
>> and how failure to do so will impact business continuity sometime in
>> the next 1 - 3 years. I've pressed our vendors to support IPv6 (note:
>> keep in mind we're a DDoS mitigation firm, our needs extend beyond
>> routers and switches) and found that it's a chicken and egg situation.
>> Vendors are neglecting to support IPv6 because there is "no demand."
>> I've pointed out your articles and demanded IPv6 support, some are
>> promising results in the next several months. We will see.
>>
> I was at a trade show several months back. I watched a series of people
> walk up to a vendor and each, in turn, asked about IPv6 support. The
> vendor told each, in turn, "You're the only one asking for it."
>
> I walked up to the vendor and took my turn being told "You're the only
> one asking for it." I pointed out that I had seen the other people get
> the same answer. The sales person admitted he was caught red
> handed and explained "We're working on it, but, we don't have a
> definite date and so our marketing department has told us to downplay
> the demand and the importance until we have something more
> definitive."
>
> 
>
> Owen
>
>
>
>



-- 
Jeffrey Lyon, Leadership Team
jeffrey.l...@blacklotus.net | http://www.blacklotus.net
Black Lotus Communications of The IRC Company, Inc.

Follow us on Twitter at http://twitter.com/ddosprotection to find out
about news, promotions, and (gasp!) system outages which are updated
in real time.

Platinum sponsor of HostingCon 2010. Come to Austin, TX on July 19 -
21 to find out how to "protect your booty."

Re: Lightly used IP addresses

[Randy Bush]

> A possible stick for ARIN could be that any AS that advertises space
> for B and any network that uses that rogue AS would not receive
> resource requests/changes from ARIN.  Perhaps too strong of a stick?

maybe you should not be searching for a stick.

RE: Lightly used IP addresses

[Frank Bulk]

This week I was told by my sales person at Red Condor that I'm the only one
of his customers that is asking for IPv6.  He sounded annoyed and it seemed
like he was trying to make me feel bad for being the "only oddball" pushing
the IPv6 feature requirement.  I tried to explain to him that by this time
next year IANA will likely have handed out all their IPv4 blocks and that I
didn't have the time spend the first half of 2011 implementing IPv6 across
my $DAYJOB network, but wanted to spread that work over time.  To his
credit, it's been on their to-do list for at least 6 months if not a year,
it's just been pushed back several quarters.

Frank

-Original Message-
From: Owen DeLong [mailto:o...@delong.com] 
Sent: Saturday, August 14, 2010 10:27 AM
To: Jeffrey Lyon
Cc: John Curran; nanog@nanog.org; Ken Chase
Subject: Re: Lightly used IP addresses


On Aug 13, 2010, at 9:12 PM, Jeffrey Lyon wrote:

> John et al,
> 
> I have read many of your articles about the need to migrate to IPv6
> and how failure to do so will impact business continuity sometime in
> the next 1 - 3 years. I've pressed our vendors to support IPv6 (note:
> keep in mind we're a DDoS mitigation firm, our needs extend beyond
> routers and switches) and found that it's a chicken and egg situation.
> Vendors are neglecting to support IPv6 because there is "no demand."
> I've pointed out your articles and demanded IPv6 support, some are
> promising results in the next several months. We will see.
> 
I was at a trade show several months back. I watched a series of people
walk up to a vendor and each, in turn, asked about IPv6 support. The
vendor told each, in turn, "You're the only one asking for it."

I walked up to the vendor and took my turn being told "You're the only
one asking for it." I pointed out that I had seen the other people get
the same answer. The sales person admitted he was caught red
handed and explained "We're working on it, but, we don't have a
definite date and so our marketing department has told us to downplay
the demand and the importance until we have something more
definitive."



Owen

RE: Lightly used IP addresses

[Frank Bulk]

A possible stick for ARIN could be that any AS that advertises space for B
and any network that uses that rogue AS would not receive resource
requests/changes from ARIN.  Perhaps too strong of a stick?

Frank

-Original Message-
From: Ken Chase [mailto:k...@sizone.org] 
Sent: Friday, August 13, 2010 2:13 PM
To: nanog@nanog.org
Subject: Re: Lightly used IP addresses

On Fri, Aug 13, 2010 at 06:49:35PM +, Nathan Eisenberg said:
  >> Is this upstream going to cut that customer off and
  >> lose the revenue, just to satisfy ARIN's bleating? 
  >
  >Isn't this a little bit like an SSL daemon?  One which refuses to process
a revocation list on the basis of the function of the certificate is
useless.  The revocation list only has authority if the agent asks for and
processes it.  Would you use this SSL daemon, knowing that it had this bug?
  >
  >I would consider a transit provider who subverted an ARIN revocation to
be disreputable, and seek other sources of transit.

Assuming the public even found out about the situation.

For ARIN to make good on this community goodwill, they'd have to

(1) publish the disrepute of the upstream who refuses to stop announcing the
rogue
downstream's prefixes.

Im not sure what step 2+ is going to be there, but I bet ARIN would become
very
unpopular with (1) above amongst its customers reselling bandwidth to other
ARIN
IPv4 block users.

How many large carriers on this list would immediately halt announcing a
downstream-in-good-financial-standing's prefixes just because ARIN say's
they're
delinquent?

I bet most wont even answer this question to the list here - most likely
dont
have an official policy for this situation, and if they did, it's likely not
going to be publically disclosed.

(If any are willing to disclose such publically, I'd love to hear/see the
policy's
details.)

/kc

  >Best Regards,
  >Nathan Eisenberg
  >Atlas Networks, LLC

-- 
Ken Chase - k...@heavycomputing.ca - +1 416 897 6284 - Toronto CANADA
Heavy Computing - Clued bandwidth, colocation and managed linux VPS @151
Front St. W.

Re: Lightly used IP addresses

[Randy Bush]

> First, in this thread we are not talking about folks who have not paid
> ARIN their dues, we are talking about folks who "sell" addresses
> despite not being authorized to do so by ARIN - aka abuse/fraud.

this is less clear-cut than you seem to think it is.  but i suspect we
will see it in court fairly soon.

randy

Re: Lightly used IP addresses

[Randy Bush]

for the embarrassing wannabe example of the month, marla and lee [0] at
the last ietf is just such a shining example.  at the mic, they state
are from the arin ac and board, like it was their day job and they were
speaking fo rarin ploicy.  and they propose to roll back a decade of
progress getting operatonal policy the  out of the ietf.  and
they don't even understand why they got jumped or why thomas's preso was
in the opposite direction and was widely supported.

the arin ceo's response to my suggestion that this be curtailed?

> If you submit it, I will bring it to the Board for consideration.  In
> fairness, I will tell you that I'll also recommend to the that we
> continue to pay for the travel for the Advisory Council, unless and
> until there is no need for a policy development process.

or ask a grown-up who has the stomach to read the arin ppml list (i
could only stomach it so long, and pulled).  it is an embarrassment to
the internet.

randy

--

[0] - sweet, well-meaning folk

Re: Lightly used IP addresses

[Chris Grundemann]

On Fri, Aug 13, 2010 at 21:32, Randy Bush  wrote:
> when the 'community' is defined as those policy wannabes who do the
> flying, take the cruise junkets, ... this is a self-perpetuating
> steaming load that is not gonna change.

Yes, those definitions create a steaming load.

But why is it that the folks actually participating in making policy
are "wannabes" in your definition?

I suggest the true definition of "community" includes at least *all*
of the non-AC-member participants in the ARIN policy process; the
folks who subscribe to the PPML and show up at meetings (or
participate remotely at a greatly reduced cost but nearly equal
voice). There are 15 AC members and around 150 participants at each
meeting... That means that _most_ are *not* being funded by ARIN.

For those who claim the system to not be open, I humbly provide myself
as a test case. I am not one of the "good old boys" of ARIN (if there
is such a thing) and I have never had ARIN pay my way to a meeting (or
for a cruise junket). In fact I am far too young and inexperienced to
possibly qualify as any kind of ruling elite who is handing down
decrees from above. I have however contributed to the formation of
several policies in the ARIN region and to the crafting of several
others currently under discussion, one on a global level amongst all 5
RIRs. I attended a meeting, joined the mailing list and spoke up.
Simple as that. I highly encourage everyone who has an opinion on
Internet numbering policy to do the same.

Cheers,
~Chris


> one start would be for arin to have the guts not to pay travel expenses
> of non-employees/contractors.
>
> randy
>
>

-- 
@ChrisGrundemann
weblog.chrisgrundemann.com
www.burningwiththebush.com
www.coisoc.org

Re: Lightly used IP addresses

[Chris Grundemann]

On Fri, Aug 13, 2010 at 15:25, Ken Chase  wrote:
> On Fri, Aug 13, 2010 at 05:00:04PM -0400, Jared Mauch said:
>  >I know of several large providers that would stop routing such "rogue" 
> space.
>
> Really? They'd take a seriously delinquent (and we're only talking about non
> payment after several months to Arin, not spammers or other 'criminal'
> elements) that's still paying for their transit and cut off their prefix
> announcements? I dont know that that's true for most outfits in these tough
> times. Nixing a $5000 or $1+ MRC revenue stream probably requires some
> hard thought at high levels in most outfits.

First, in this thread we are not talking about folks who have not paid
ARIN their dues, we are talking about folks who "sell" addresses
despite not being authorized to do so by ARIN - aka abuse/fraud.

Either way, if ARIN finds strong enough reason to revoke numbers from
Org A who is ISP X' customer, ARIN will eventually reassign those
numbers. When ISP Y calls ISP X and says "hey, your customer Org A is
advertising my customer Org B's address space." ISP X will check
WHOIS, see that they are telling the truth and filter that block from
Org A. If ISP X does not, they will likely see peering and transit
options shrink rapidly.

So in short - yes, really.
~Chris

>
> /kc
> --
> Ken Chase - k...@heavycomputing.ca - +1 416 897 6284 - Toronto CANADA
> Heavy Computing - Clued bandwidth, colocation and managed linux VPS @151 
> Front St. W.
>

-- 
@ChrisGrundemann
weblog.chrisgrundemann.com
www.burningwiththebush.com
www.coisoc.org

Re: 40 acres and a mule, was Lightly used IP addresses

[Andrew Kirch]

 40 Acres and a Mule were promised to every slave freed in the south by 
General Grant.  It was later rescinded.  600 acres was promised to 
non-landowning general militia soldiers after the Revolutionary war.  
You're only off by ~100 years.


Andrew

On 8/14/2010 1:27 PM, Jimi Thompson wrote:

It was 40 acres and a mule - FYI


On 8/14/10 11:22 AM, "John R. Levine"  wrote:


Convincingly said here on an ISP mailing list. But what about the
folks who were denied address assignments by ARIN policies over the
last 15 years? Denied them based on the fiction that ISPs didn't own
IP addresses, that they were merely holding the addresses in trust for
the public they serve. ...

I dunno.  What was New York's responsibility in the 1790s to guys who
didn't join the army because they had to stay home and take care of their
widowed mother and six younger sisters?

I wouldn't for a moment claim that IPv4 space was a way that was uniformly
fair or wise or close to ideal.  But I don't think you're going to have
much luck imposing fairness and wisdom retroactively on people who've
already got the space.

R's,
John

Re: Lightly used IP addresses

[bmanning]

On Sat, Aug 14, 2010 at 12:32:50PM -0700, David Conrad wrote:
> Bill,
> 
> On Aug 14, 2010, at 8:51 AM, bmann...@vacation.karoshi.com wrote:
> > In the formal ARIN context, there is a distiction between abuse and 
> > fraud.
> > 
> > abuse::  https://www.arin.net/abuse.html
> 
> This is a FAQ for folks who are accusing ARIN of abuse of network. With the 
> possible exception of the last item in that FQA, it has nothing to do with 
> the topic at hand.
> 
> > fraud::  https://www.arin.net/resources/fraud/index.html
> 
> This is the mechanism by which one reports fraud.  
> 
> > It would be helpful in clarifing the discussion if folks used the proper
> > terminology.
> 
> Can you point to where ARIN defines exactly what they consider "abuse" and/or 
> "fraud"?
> 
> Thanks,
> -drc

The AC accepted draft proposal below has a definition of abuse in #b


Draft Policy 2010-11
Required Resource Reviews

Version/Date: 20 July 2010

Policy statement:

Replace the text "under sections 4-6" in section 12, paragraph 7 with
"under paragraphs 12.4 through 12.6"

Add to section 12 the following text:

10. Except as provided below, resource reviews are conducted at the
discretion of the ARIN staff. In any of the circumstances mentioned
below, a resource review must be initiated by ARIN staff:

a. Report or discovery of an acquisition, merger, transfer, trade or
sale in which the infrastructure and customer base of a network move
from one organization to another organization, but, the applicable IP
resources are not transferred. In this case, the organization retaining
the IP resources must be reviewed. The organization receiving the
customers may also be reviewed at the discretion of the ARIN staff.

b. Upon receipt by ARIN of one or more credible reports of fraud or
abuse of an IP address block. Abuse shall be defined as use of the block
in violation of the RSA or other ARIN policies and shall not extend to
include general reports of host conduct which are not within ARIN's scope.


While fraud is outlined here: 
https://www.arin.net/resources/fraud/index.html

Version 1.2 - 18 November 2009

This reporting process is to be used to notify ARIN of suspected Internet 
number resource abuse  including the submission of falsified utilization or 
organization information, unauthorized changes to data in ARIN's WHOIS, 
hijacking of number resources in ARIN's database, or fraudulent transfers.

This reporting process is NOT for reporting illegal or fraudulent Internet 
activity like network abuse, phishing, spam, identity theft, hacking, scams, or 
any other activity unrelated to the scope of ARIN's mission.





so fraud, from ARINs perspective seems to be:

- submitting falsified untilization or org info
- unauthorized changes to the data in ARINs whois
- hijacking number resources in ARINs database
- fraudulent transfers


a kewpie doll for the first one to point out the circular dependencies!  :)

--bill

Re: Lightly used IP addresses

[David Conrad]

Bill,

On Aug 14, 2010, at 8:51 AM, bmann...@vacation.karoshi.com wrote:
>   In the formal ARIN context, there is a distiction between abuse and 
> fraud.
> 
>   abuse::  https://www.arin.net/abuse.html

This is a FAQ for folks who are accusing ARIN of abuse of network. With the 
possible exception of the last item in that FQA, it has nothing to do with the 
topic at hand.

>   fraud::  https://www.arin.net/resources/fraud/index.html

This is the mechanism by which one reports fraud.  

>   It would be helpful in clarifing the discussion if folks used the proper
>   terminology.

Can you point to where ARIN defines exactly what they consider "abuse" and/or 
"fraud"?

Thanks,
-drc

Re: Lightly used IP addresses

[Owen DeLong]

I think you mistake my meaning. I don't regard RA and SLAAC as a problem. I 
regard their limited capabilities as a minor issue. I regard the IETF religion 
that insists on preventing DHCPv6 from having a complete set of capabilities 
for some form of RA protectionism to be the largest problem. That was my 
meaning for RA religion.

Owen


Sent from my iPad

On Aug 14, 2010, at 10:30 AM, Joel Jaeggli  wrote:

> 
> 
> On Aug 14, 2010, at 8:05, Owen DeLong  wrote:
>> On Aug 13, 2010, at 8:01 PM, Randy Bush wrote:
>> 
>> The lack of end-site multihoming (more specifically the lack of PI for
>> end-sites) was created by the IETF and resolved by the RIRs.
>> The beginning of resolving this was ARIN proposal 2002-3.
>> 
>> The RA religion still hasn't been solved.
> 
> Neither for that matter has the dhcp religion. Autoconfiguration and 
> bootstrapping were not solved problems for ipv4  inn 1994 and in some 
> respects still aren't. The mind boggles that we consider the ipv4 situation 
> so much better than the v6 case...
> 
>> Owen
>> 
>> 
>> 

Re: 40 acres and a mule, was Lightly used IP addresses

[Joel Jaeggli]

On Aug 14, 2010, at 10:27, Jimi Thompson  wrote:

> It was 40 acres and a mule - FYI

No 40 acres was 1/4 of 1/4 of a section. That's 's Sherman's field order (1865) 
not the homestead act (which was 160). Or the circa 1790 activity referred to 
in this thread. 

Joel's iPad
> 
> 
> On 8/14/10 11:22 AM, "John R. Levine"  wrote:
> 
>>> Convincingly said here on an ISP mailing list. But what about the
>>> folks who were denied address assignments by ARIN policies over the
>>> last 15 years? Denied them based on the fiction that ISPs didn't own
>>> IP addresses, that they were merely holding the addresses in trust for
>>> the public they serve. ...
>> 
>> I dunno.  What was New York's responsibility in the 1790s to guys who
>> didn't join the army because they had to stay home and take care of their
>> widowed mother and six younger sisters?
>> 
>> I wouldn't for a moment claim that IPv4 space was a way that was uniformly
>> fair or wise or close to ideal.  But I don't think you're going to have
>> much luck imposing fairness and wisdom retroactively on people who've
>> already got the space.
>> 
>> R's,
>> John
>> 
> 
> 
> 

Re: 40 acres and a mule, was Lightly used IP addresses

[Scott Brim]

On 08/14/2010 13:27 EDT, Jimi Thompson wrote:
> It was 40 acres and a mule - FYI

That was Civil War, for freed slaves.  Here in NY, war of independence
veterans were given at least 100 acres each.

See http://en.wikipedia.org/wiki/Central_New_York_Military_Tract

Re: Lightly used IP addresses

[Owen DeLong]

On Aug 14, 2010, at 8:47 AM, Bret Clark wrote:

> On 08/14/2010 11:27 AM, Owen DeLong wrote:
>> I was at a trade show several months back. I watched a series of people
>> walk up to a vendor and each, in turn, asked about IPv6 support. The
>> vendor told each, in turn, "You're the only one asking for it."
>> 
>> I walked up to the vendor and took my turn being told "You're the only
>> one asking for it." I pointed out that I had seen the other people get
>> the same answer. The sales person admitted he was caught red
>> handed and explained "We're working on it, but, we don't have a
>> definite date and so our marketing department has told us to downplay
>> the demand and the importance until we have something more
>> definitive."
>>   
> What company was that? I find it rather odd that any marketing group in any 
> company would tell a sales team to downplay a possible future migration path; 
> especially in the case of IP6 which isn't a possible future migration 
> strategy, but IS a future migration strategy. That's one company I don't want 
> to do business with if that's what they are telling their sales team...shows 
> lack of a road map and a total lack of any understanding of this industry!

I won't name names as that company has since changed their
tune and there is nothing to be gained by publicly embarrassing
them.

Owen

Re: Lightly used IP addresses

[Joel Jaeggli]

On Aug 14, 2010, at 8:05, Owen DeLong  wrote:
> On Aug 13, 2010, at 8:01 PM, Randy Bush wrote:
> 
> The lack of end-site multihoming (more specifically the lack of PI for
> end-sites) was created by the IETF and resolved by the RIRs.
> The beginning of resolving this was ARIN proposal 2002-3.
> 
> The RA religion still hasn't been solved.

 Neither for that matter has the dhcp religion. Autoconfiguration and 
bootstrapping were not solved problems for ipv4  inn 1994 and in some respects 
still aren't. The mind boggles that we consider the ipv4 situation so much 
better than the v6 case...

> Owen
> 
> 
> 

40 acres and a mule, was Lightly used IP addresses

[Jimi Thompson]

It was 40 acres and a mule - FYI


On 8/14/10 11:22 AM, "John R. Levine"  wrote:

>> Convincingly said here on an ISP mailing list. But what about the
>> folks who were denied address assignments by ARIN policies over the
>> last 15 years? Denied them based on the fiction that ISPs didn't own
>> IP addresses, that they were merely holding the addresses in trust for
>> the public they serve. ...
> 
> I dunno.  What was New York's responsibility in the 1790s to guys who
> didn't join the army because they had to stay home and take care of their
> widowed mother and six younger sisters?
> 
> I wouldn't for a moment claim that IPv4 space was a way that was uniformly
> fair or wise or close to ideal.  But I don't think you're going to have
> much luck imposing fairness and wisdom retroactively on people who've
> already got the space.
> 
> R's,
> John
> 

Re: 600 acres and a mule, was Lightly used IP addresses

[William Herrin]

On Sat, Aug 14, 2010 at 12:22 PM, John R. Levine  wrote:
> I wouldn't for a moment claim that IPv4 space was a way that was uniformly
> fair or wise or close to ideal.  But I don't think you're going to have much
> luck imposing fairness and wisdom retroactively on people who've already got
> the space.

John.

As things stand, IP addresses you've gained control of in the past
decade and a half you gained under a contract in which you explicitly
agreed that they not only didn't belong to you but that your continued
control was subject to the general public's pleasure as expressed
through regularly revised ARIN public policy. You won't tear that up
like an Indian treaty without first overcoming a certain amount of
push back from that disenfranchised public.

If you want IPv4 addressing to enter a legal regime similar to real
estate, I suggest that among other things you figure out how the taxes
should work and write some guidance for the pols before they start
figuring it out for themselves. If you don't construct the public
policy from the bottom up, you can count on someone else building it
from the top down and no newly defined form of property with a
quantifiable value is likely to escape taxation. Bear in mind that as
with real property, tax regimes which encourage a concentration of
ownership by a few wealthy owners will tend to be viewed with
suspicion and disdain.

Regards,
Bill Herrin


-- 
William D. Herrin  her...@dirtside.com  b...@herrin.us
3005 Crane Dr. .. Web: 
Falls Church, VA 22042-3004

Re: 600 acres and a mule, was Lightly used IP addresses

[John R. Levine]

Convincingly said here on an ISP mailing list. But what about the
folks who were denied address assignments by ARIN policies over the
last 15 years? Denied them based on the fiction that ISPs didn't own
IP addresses, that they were merely holding the addresses in trust for
the public they serve. ...


I dunno.  What was New York's responsibility in the 1790s to guys who 
didn't join the army because they had to stay home and take care of their 
widowed mother and six younger sisters?


I wouldn't for a moment claim that IPv4 space was a way that was uniformly 
fair or wise or close to ideal.  But I don't think you're going to have 
much luck imposing fairness and wisdom retroactively on people who've 
already got the space.


R's,
John

Re: Lightly used IP addresses

[bmanning]

On Sat, Aug 14, 2010 at 08:40:28AM -0700, Owen DeLong wrote:
> 
> On Aug 13, 2010, at 9:33 PM, Franck Martin wrote:
> 
> > Funny! 
> > 
> > On one hand people talk about ARIN providing IP allocation at nearly zero 
> > cost and on the other hand talking that ARIN goes after companies that use 
> > their allocation for abuse (which has a non trivial cost and potential 
> > expensive lawsuits)...
> > 
> > Do you know what you want?
> 
> Let's clarify the definition of abuse in this context. We are not talking 
> about people who use their IPs to abuse the network. We are talking about 
> resource recipients who use their allocations or assignments in contravention 
> to the policies under which they received them (and thus contrary to the RSA 
> which they signed when they received them).
> 
> Owen


In the formal ARIN context, there is a distiction between abuse and 
fraud.

abuse::  https://www.arin.net/abuse.html

fraud::  https://www.arin.net/resources/fraud/index.html

It would be helpful in clarifing the discussion if folks used the proper
terminology.


--bill

Re: Lightly used IP addresses

[Bret Clark]

On 08/14/2010 11:27 AM, Owen DeLong wrote:

I was at a trade show several months back. I watched a series of people
walk up to a vendor and each, in turn, asked about IPv6 support. The
vendor told each, in turn, "You're the only one asking for it."

I walked up to the vendor and took my turn being told "You're the only
one asking for it." I pointed out that I had seen the other people get
the same answer. The sales person admitted he was caught red
handed and explained "We're working on it, but, we don't have a
definite date and so our marketing department has told us to downplay
the demand and the importance until we have something more
definitive."
   
What company was that? I find it rather odd that any marketing group in 
any company would tell a sales team to downplay a possible future 
migration path; especially in the case of IP6 which isn't a possible 
future migration strategy, but IS a future migration strategy. That's 
one company I don't want to do business with if that's what they are 
telling their sales team...shows lack of a road map and a total lack of 
any understanding of this industry!

Re: Lightly used IP addresses

[Owen DeLong]

On Aug 13, 2010, at 9:33 PM, Franck Martin wrote:

> Funny! 
> 
> On one hand people talk about ARIN providing IP allocation at nearly zero 
> cost and on the other hand talking that ARIN goes after companies that use 
> their allocation for abuse (which has a non trivial cost and potential 
> expensive lawsuits)...
> 
> Do you know what you want?

Let's clarify the definition of abuse in this context. We are not talking about 
people who use their IPs to abuse the network. We are talking about resource 
recipients who use their allocations or assignments in contravention to the 
policies under which they received them (and thus contrary to the RSA which 
they signed when they received them).

Not that I don't think going after network abuse is worth while, it absolutely 
is, but, that's not within the current scope of ARIN policy. The community 
would need to come to consensus on a definition of abuse and the desire for 
ARIN to take on such a role before it would be possible.

For now, ARIN's role is limited to the administration of the address space in 
the public trust. That includes taking action to resolve situations where 
addresses are being used in a manner contrary to the ARIN policies developed by 
the community.

Owen

Re: Lightly used IP addresses

[Owen DeLong]

On Aug 13, 2010, at 9:12 PM, Jeffrey Lyon wrote:

> John et al,
> 
> I have read many of your articles about the need to migrate to IPv6
> and how failure to do so will impact business continuity sometime in
> the next 1 - 3 years. I've pressed our vendors to support IPv6 (note:
> keep in mind we're a DDoS mitigation firm, our needs extend beyond
> routers and switches) and found that it's a chicken and egg situation.
> Vendors are neglecting to support IPv6 because there is "no demand."
> I've pointed out your articles and demanded IPv6 support, some are
> promising results in the next several months. We will see.
> 
I was at a trade show several months back. I watched a series of people
walk up to a vendor and each, in turn, asked about IPv6 support. The
vendor told each, in turn, "You're the only one asking for it."

I walked up to the vendor and took my turn being told "You're the only
one asking for it." I pointed out that I had seen the other people get
the same answer. The sales person admitted he was caught red
handed and explained "We're working on it, but, we don't have a
definite date and so our marketing department has told us to downplay
the demand and the importance until we have something more
definitive."

> Meanwhile, there are hosting companies, dedicated server companies,
> etc. with /17 and /18 allocations who are either forging justification
> or wildly abusing the use of that space outside of the declared need.

Then those cases should be submitted to the fraud/abuse reporting
process so they can be investigated and resolved.


Owen

Re: Lightly used IP addresses

[Owen DeLong]

On Aug 13, 2010, at 8:01 PM, Randy Bush wrote:

>> Yet most of the bad ideas in the past 15 years have actually come from
>> the IETF (TLA's, no end site multihoming, RA religion), some of which
>> have actually been "fixed" by the RIR's.
> 
> no, they were fixed within the ietf.  that's my blood you are taking
> about, and i know where and by whom it was spent.
> 
I'm not sure what is meant by TLAs in this context, so, I'll leave that alone.

The lack of end-site multihoming (more specifically the lack of PI for
end-sites) was created by the IETF and resolved by the RIRs.
The beginning of resolving this was ARIN proposal 2002-3.

The RA religion still hasn't been solved.

Owen

Re: Lightly used IP addresses

[Patrick W. Gilmore]

Watching people snark on mailing lists is occasionally entertaining.  Watching 
them snark on the wrong mailing lists is usually less entertaining.  Watching 
them snark on the wrong mailing list for 100+ posts when the things they are 
snarking about were voted on by themselves is getting a little silly.

Watching them snark about the people they are snarking -to- trying to get them 
to participate in the process they are snarking -about- is pathetic.


If you don't like the way ARIN does things, change them.  I don't like people 
going to the IETF and trying to get the IETF to do things the operators should 
be doing.  I talked to the AC & BoD members before I voted, and none of them 
mentioned this to me.  I feel like snarking about that is valid, since I put in 
time & effort, but was still caught by surprise.  But instead of snarking, I'm 
working to change that.

How much time & effort was spent (wasted?) reading mailing lists that could 
have been used to put forth proposals to ARIN (or the other RIRs)?  Which is 
more likely to get what you want?

Oh, and about ARIN wasting money: Do you really think a 10% or even 50% 
reduction in ARIN fees will make -any- difference to the companies paying those 
fees?  OTOH, I do believe a 50% reduction in ARIN fees will result in far less 
outreach, which means less community participation, which I feel is suboptimal. 
 If you disagree, propose a change, get me & people who feel as I do outvoted, 
and things will change.  What's more, I will not snark about the fact I got 
outvoted on NANOG.

Or you can post to NANOG and see nothing change.  Up to you.

-- 
TTFN,
patrick

Re: Question of privacy with reassigned resources

[bmanning]

On Sat, Aug 14, 2010 at 12:23:02PM +0100, Michael Dillon wrote:
> > Shall I go on? Regardless of what you may think about whether those
> > injured folks should be entitled to the information, the fact is that
> > they are entitled to it under ARIN policy developed based on public
> > consensus. Which means you injure them by denying it.
> 
> Enough with the amateur lawyering!
> 
> A minor inconvenience is NOT injury under the law.

Pot, Kettle, Black.
Thanks for that legal advice Michael.

--bill

Re: Question of privacy with reassigned resources

[Michael Dillon]

> Shall I go on? Regardless of what you may think about whether those
> injured folks should be entitled to the information, the fact is that
> they are entitled to it under ARIN policy developed based on public
> consensus. Which means you injure them by denying it.

Enough with the amateur lawyering!

A minor inconvenience is NOT injury under the law.
And in fact, if the organization is failing to disclose the customer information
in order to direct all queries to their own address, where they can be
handled by technically competent people, then the judge would laugh you out
of court.

It is common practice for ISPs to redact whois entries to only include
the customer's city and state for the address. This is not fraud and
has been going on for years.

Every ISP should check their whois entries and make sure that any
entries for apartments and people's homes are redacted to say
nothing more that PRIVATE RESIDENCE.

Re: 600 acres and a mule, was Lightly used IP addresses

[William Herrin]

On Sat, Aug 14, 2010 at 12:00 AM, John Levine  wrote:
>> And in complete fairness - why should folks who received vast tracts
>> of addresses for little or no cost under a justified-need regime now
>> have free reign to monetize their sale?
>
> All of the real estate in my part of New York traces back to Military
> Tract so and so.  In the 1790s the state allocated two million acres
> on a justified-need basis, with the rule being that if you were a
> Revolutionary War veteran, that justified you getting 600 acres, free.
> You had to provide your own mule.
>
> Once the lots were all handed out, in subsequent transfers, nobody
> justified anything.

John,

Convincingly said here on an ISP mailing list. But what about the
folks who were denied address assignments by ARIN policies over the
last 15 years? Denied them based on the fiction that ISPs didn't own
IP addresses, that they were merely holding the addresses in trust for
the public they serve. With the IPv4 free pool almost totally
allocated, what's our residual responsibility to the folks that we
(and by we I mean you) deliberately prevented from getting addresses
while forking over /9's to Verizon?

It might have been wiser to handle IPv4 addresses in a property-like
regime. It might yet be smarter to handle IPv6 addresses that way. But
we didn't handle IPv4 addresses that way and now that the free pool is
nearly empty there would be significant fairness issues with any plan
to allow current registrants to treat their IPv4 address holdings as
if they were fee simple real estate. Like those 600 acres and the
mule.


> If you want to sell and I want to buy, that's
> that and the county records it for a nominal clerical fee.
> You will doubtless find similar histories all over the country.

Taxes. Nearly everywhere real estate is owned. A non-trivial
percentage of the land's sale value in real estate taxes year after
year after year. Makes it tough to horde or under-utilize land. When
paired with other progressive tax regimes that discourage leasing in
favor of leveraged purchases, tends to balance early ownership
inequities over time. Even Hawaii's ownership by the Doles, Baldwins
and Robinsons is slowly falling to the relentless hammering of taxes.

Ready to forge ahead with ownership, the rights of ownership and the
great privilege that is property taxes in the IPv4 addressing realm?
Be careful what you wish for.

Regards,
Bill Herrin



-- 
William D. Herrin  her...@dirtside.com  b...@herrin.us
3005 Crane Dr. .. Web: 
Falls Church, VA 22042-3004