[governance] Are Internet users powerless or empowered, and how?

[George Sadowsky]

Karl has raised a number of useful observations, some of which I 
agree with and some of which I don't.  My comments are interspersed 
below.

I think that this has been an interesting discussion, but I am seeing 
decreasing returns to continuing it.  Others may feel differently.


>George Sadowsky wrote:
>
>>However, I do think that the way you phrase it, i.e. "the body that 
>>extracts over half a billion dollars (US$ out of the pockets of 
>>domain name buyers every year," goes in the wrong direction.  It's 
>>correct that ICANN is involved in price setting, but per domain 
>>name the cost is closer to $6.
>
>My estimate is based on a computed registry cost (which I'll get to 
>in a few paragraphs), an ICANN granted registry fee of about $7 (not 
>to mention the ICANN piece of every registration), and about 
>75,000,000 names (largely in .com).
>
>There is no doubt that domain name buyers are paying in excess of 
>$500,000,000 per year just in the ICANN granted registry fee.  The 
>question is how much does it actually cost to provide those registry 
>services?
>
>A price-setting regulatory body ought to know how much it costs to 
>provide the regulated service being provided.  Unfortunately the 
>body of internet governance that sets domain name registry prices 
>(and its own fee as well) seems to never have bothered to inquire as 
>to the actual cost.
>
>Perhaps it is obvious to that body, but it is certainly not obvious to me.
>
>The $6 appears to be nothing but a fiat amount - it appears to be 
>based on no evidence, no information, no audit - no nothing.  There 
>is no evidence that ICANN has never tried to establish a cost basis. 
>And now it is going up to $7, not $6, with an additional bi-yearly 
>7% rise, again without any supporting accounting, much less audited 
>accounting.
>
>I've estimated the cost at about $0.03 per name per year.  Perhaps 
>that is too low, perhaps not.  But what evidence is there to 
>contradict my calculations?  I'd love to hear concrete, auditable, 
>quantitative information that leads me (and us) to a real answer 
>that we can believe and use rather than debate.

Cost, of course, depends upon the number, nature, and level of 
services that go with the registration.

>
>The analysis of others also indicates that $6 much, much higher than 
>the real cost.
>
>Suppose that I'm off by a factor of 33x.  That still means that 
>ICANN is pumping/taxing the internet to the tune of about 
>$400,000,000 (USD) on .com alone every year and splitting the 
>proceeds between itself and Verisign.
>
>Alternatively we can use Tucow's bid at running .com at $2 - on 
>which basis the money pump is a mere $330,000,000 every year (on 
>.com alone) and rising with ICANN's 7% solution.
>
>Even at these lesser amounts, the sums are still quite significant.

I'm not as familiar with the history as I would like to be, but I 
believe that you are right, that $6 was at first a guess at what it 
would take to support registry and registar operations.  that guess 
must have been made a long time ago, based upon some untested 
assumptions regarding volume, cost projections, and service levels.

>
>Thus we see an ICANN, because it is not accountable to the community 
>of internet users, that has become excessively accommodating to the 
>registry industry - gifting it with huge revenue streams and profit 
>margins that are measured in the 1000% to 35,000% range.

Without commenting on your profit margin estimates, I would be 
curious to know if any studies or good business plans exist that 
indicate that a much lower cost would be commercially viable.  The 
business has changed significantly since the fist price was set, 
including mirror servers, query rates and now data escrow issues. 
Perhaps that would be a useful study for an economist in industrial 
organization who also has a technical bent.  In fact, the entire 
domain name industry is ripe for study.  Ross Rader gave me the name 
of a writer at the Wall Street Journal who he thought was writing 
such a study, but I've receiived no response from him.

>
>When the body of internet governance not only guarantees registries 
>a profit, but a profit margin measured in tens of thousands of 
>percents, is it still internet governance?  Or is it something else?
>
>When I was on the board at ICANN I found an across-the-board (pun 
>intended) reluctance to look at any sort of hard numbers of 
>anything, even ICANN's own expenses.  Indeed, when I went to look 
>for myself I found my way barred and I had to bring legal action 
>simply so that I could do what board members around the world are 
>empowered to do - look at the financial records.

When you were on the Board, the dynamics of ICANN, and of its 
interactions with the burgeoning industry were considerably 
different.   I can understand that it is natural for you to look 
through the lens of that period, but isn't it possible that it's not 
the right lens now?  I can think of a number of ways to test that 
hypothesis.  Can you?

>
>In other words, I am suggesting that there may be an institutional 
>aversion to asking too many questions about where and how money 
>flows.
>
>One of my concerns about ICANN's nominating committee process is 
>that it tends to produce people who are worthies but are of an 
>accommodating nature, not of the ilk will demand to see hard proof 
>of an assertion.

Well, I would disagree somewhat here, in part because of my 
familiarity with the process as Chair of the ICANN NomCom for the 
last three years.  I know that you said "tends to produce" and 
certainly some people nominated have an accommodating nature, but we 
have also nominated people who are professionally aggressive almost 
to a fault.  They may have directed their energies to other problems, 
of course.  And you would not want a Board of people, most of whom 
were overly demanding in different directions (we may disagree here), 
because it would be difficult for such a Board to get things done. 
There are, of course, cultural differences that cause people from 
some cultures to be more accommodating  ---  essentially practicing a 
different style of management  ---  and that often produces a less 
than perfect result.  When I worked at the United Nations, I observed 
many misunderstandings and disagreements  largely due to different 
cultural norms and conflicting cultural assumptions.

I don't see the ICANN Board as complacent.  I agree that it would be 
a bad thing; just look at corporate board behavior  in the US and we 
could both identify cases in which the board-CEO collusion has led to 
disaster for stockholders, as well as having significant financial 
repercussions across the economy (Tyco, Enron, MCI, ...)

>
>As such it is not surprising that ICANN has simply accepted a domain 
>name registry price policy that began with an arbitrary number - a 
>number that was simply created out of thin air a few years ago - and 
>increments it by a percentage that was also created out of thin air.
>
>Had ICANN had a working election process it may have found its board 
>populated by more people willing to require hard facts before 
>granting rich price terms, paid for not by ICANN but, instead, out 
>of the pockets of the users of the internet.

There are several problems with elections, and I don't want to get 
into a discussion of them now, but the one show-stopper that I see is 
defining the electorate

>
>>I agree with you that WHOIS continues to be a problem, complicated 
>>by competing interests but also by non-interoperable national legal 
>>codes, over which we have relatively no control (at least in the 
>>short run).  I'd like to see that sorted out also, but I don't see 
>>any voting scheme able to solve that problem without creating other 
>>problems of equal or greater magnitude.
>
>You are right that voting systems alone will not solve Whois.
>
>But allowing internet users light a fire under ICANN's board, a fire 
>created through the accountability provided by elections, then I 
>submit that ICANN would not have repeatedly waivered when the 
>intellectual property industry said "boo", as it did just a few 
>weeks ago in Los Angeles.

I do not like the WHOIS result either.  In my view, Ross Rader's 
presentation of the alternative made a great deal of sense.  But it 
seems to me that this is a problem within the GNSO, which could be 
considered a general problem in the structure of ICANN.  If GNSO 
makes policy and decides not to make a recommendations, and the ICANN 
Board reviews the appropriateness of the process, how would an 
elected Board be able to come to a different result? (Just asking, 
not criticizing)

>
>>I understand that you have a severe dislike of the current UDRP. 
>>Is there a comprehensive alternative you would like to suggest that 
>>is significantly better?  If you have already suggested it, what 
>>has been its reception and why?
>
>The UDRP starts with a fundamental error: It acts as a sword to 
>vindicate rights in a name only if those rights are based on 
>trademark.
>
>In other words, if I own a trademark "foo" then I can use the UDRP 
>to challenge others who use "foo".  I might win, I might loose, but 
>at least I have the UDRP as a tool.
>
>On the other hand, if I am named "foo" or my god is named "foo" or 
>my university is named "foo" - all of which are legal, valid, and 
>legitimate non-trademark uses of that name - and I feel that my 
>rights are violated by someone else's use of "foo", then I can not 
>call upon the UDRP, the UDRP is not a tool that I can invoke simply 
>because my rights in the name are not trademark based.
>
>In other words, the first thing to fix in the UDRP is to make 
>require only that the plantiff have rights in a name, not that those 
>rights are trademark rights.
>
>Secondly, the UDRP replaces the existing legal system.  The legal 
>system is complex and expensive because it bends over backwards to 
>be fair. The UDRP is attractive to intellectual property owners and 
>lawyers (like me, on both counts) because it is fast and cheap.  But 
>that speed and low cost come at a price - the loss of fairness. 
>Among the ways the UDRP is unfair is the way that those who make 
>choices are paid, it tends to make them friendly to the plaintiff.
>
>Thirdly, because the UDRP is a private law that supersedes nations 
>it tends to squash cultural differences.  I'm certain that in the 
>Sudan right now nobody is wondering about the trademark names 
>associated with a certain teddy bear that has been in the news. 
>That situation demonstrates how different are the cultural feelings 
>about names that the UDRP covers with a single worldwide, commercial 
>trade name based system.

Thanks for the exposition.  I feel a bit out of my depth here, 
because I have no legal training.  But it seems to me that a 
significant piece of this issue is related to the concept of 
copyright in general, and its application in certain media. 
Furthermore, is it not correct that the UDRP is an optional mediation 
device, and that challenge within a legal system (not clear whose 
system or what decides it), outside of the Internet space, is still 
possible?

>
>>What do you think of my suggestion to concentrate on the great 
>>majority of Internet users, mostly those without domain names, and 
>>do two things.  First, define their real needs to the best of our 
>>ability.  Second, and only after we've done the first, discuss what 
>>forms of structure, conduct and governance would best meet those 
>>needs, nows and in the future?
>
>Yes is useful to remember that the internet is much larger than 
>those who spend money on domain names.  And that is precisely why I 
>find the "stakeholder" conception so pernicious - it tends to 
>identify the degree of interest ("stake") and thus the degree of 
>authority in bodies of internet governance with the amount of money 
>that the putative "stakeholder" spends or makes.
>
>So yes, we ought to remember the vast masses who are unheard and 
>who's money in the net is not clearly identifiable and not, on an 
>individual basis, very large.
>
>On the other hand, when we have a fairly clear cut issue - such as 
>domain name registry fees unrelated to the actual cost of providing 
>the domain name registry service - and a well identifiable body of 
>people being harmed (those who buy domain names and also, as we 
>should not forget, those who find them too expensive and this forego 
>buying a domain name), and an amount of money that would be 
>significant even by Rockefeller standards, then that is an issue we 
>ought to face.


I agree that DN registry fees are a clearer issue.  But I'd rather 
attack the larger and IMHO the considerably more important issue, 
even if it is ill-defined, because i think that convergence and 
appropriate action in this space will yield, to be somewhat trite, 
the greater good for the greater number.  However, I'd still love to 
see the domain name industry study done so that there would be 
durable factual evidence to help guide the evolution of that industry 
in an efficient and effective manner.

>
>		--karl--


George
____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, send any message to:
     governance-unsubscribe at lists.cpsr.org

For all list information and functions, see:
     http://lists.cpsr.org/lists/info/governance