Response to: Questions to the Community on Accountability and Transparency within ICANN

[Kieren McCarthy <kierenmccarthy@xxxxxxxxx>]

    Questions for the ICANN Community

My name is Kieren McCarthy and I have been a long-time observer of ICANN and
its processes both from the outside as a journalist and blogger, and from
the inside - until relatively recently I was ICANN’s general manager of
public participation.



I was also a major contributor to each of the three previous efforts aimed
at addressing transparency and accountability within ICANN. I was a contact
point, staff resource, as well as author and copy editor to the One World
Trust report in 2007; the Accountability and Transparency Frameworks and
Principles in 2008; and the Improving Institutional Confidence consultation
in 2008 and 2009.









Q1. Do you think ICANN is accountable to all stakeholders? Can you identify
a specific example(s) when ICANN did not act in an accountable manner? If
so, please provide specific information as to the circumstances and indicate
why you believe ICANN’s actions were not taken in an accountable manner.





A1. In broad terms, ICANN is accountable to all stakeholders. There are
certainly lots of structures and procedures in place aimed at making the
organization accountable to all stakeholders.



The problem is, and remains, that while there are lots of aspects of ICANN
that you can point to as it being “accountable”, the reality is that
decisions are made without anyone being aware of their logic used to arrive
at them. And when they are made, there is an enormous cultural bias against
revisiting them – even when there is significant protest at that decision.



This situation – where decisions can be made in private and where those
decisions cannot be challenged – has created a culture where groups of
people have become very adept at manipulating the formal structures and
procedures to their own ends. And this process is why the community
continues to complain that the organization is not accountable, despite a
significant façade of accountability.



As to specific examples, here are some below. Although I have to question
whether it is worthwhile providing then when it appears that the ATRT team
has already accepted the examples put forward by the Berkman Center, namely:



·       Introduction of new gTLDs (EoI, IRT, GAC and Vertical Integration)

·       Dot-xxx

·       DNS-CERT



Nonetheless, here are some quick examples in case the ATRT wishes to look
further:





·       A review of the Registrar Accreditation Agreement, caused by the
collapse a registrar, RegisterFly. This was initially done out in the open,
with public discussion and suggestion, with lists of ideas for changes
compiled and discussed, and with ICANN staff diligently working through
community suggestions and outlining the pros and cons of each and putting
forward what they felt were the best changes.



This process was running well until it came to the point where the changes
were becoming concrete. At this point, the process suddenly became opaque,
at the registrars’ insistence. What emerged was a greatly cutdown and
changed version of what had been created in public.



ICANN followed its accountability procedures – revisions were posted to
public comment and that comment was duly noted. But there was no mechanism
to go back to the previous version. The process drove forward and would only
accept change to what was in the current version.



The end result of this is that it is possible for a group to have
significant impact on a process in private, and then those changes – because
they have been made – are largely unchallengeable.



It was no mistake that as soon as the revised RAA was announced that the
registrars conceded that there would a further round of review of the RAA –
something that so far has been fiercely resisted.



·       The President’s Strategy Committee and its Improving Institutional
Confidence consultation.



Again, this process started out very open and public. Views were invited,
papers were produced, public comment was summarized and introduced and then
put back out again.



But, again, as soon as the process starting drawing toward concrete
conclusions, the process became opaque and a new paper emerged that saw
significant changes to what had previously been produced.



Again, all the ICANN procedures were followed – public comment periods,
public sessions at meetings  - but the reality was that significant changes
had been made in private with little or no explanation as to why and there
was no mechanism or discussion about the changes themselves. All discussion
became focused on what was in the current paper.



Eventually all of this work – which initially included a wide range of small
measures – was boiled down to two large measures. The next stage saw bylaw
revisions drawn up - which it could be argued didn’t reflect the many months
of prior discussion – and then put out again for public comment.



And then, with the changing of the JPA to the AoC, the measures appear to
have been completely forgotten, with no explanation as to why, despite
several public requests. The fact that no answer at all has emerged as to
what is happening with these measures is a demonstration of a lack of
accountability at the end of a process, as well as the various failures of
accountability during the process.





·       The Berkman Center has chosen well with some of its suggested
topics, namely the Expressions of Internet (EoI) proposal and the
Implementation Recommendation Team (IRT).



Both are examples of extremely badly run processes, done in the face of
quite precise criticism and, on occasion, pleadings from all sides to stop
and improve the process. Both failed miserably, despite best efforts on all
sides.



It should also be strongly noted that those running both processes were
convinced that they were doing the right thing in the right way and there
was no malice or conspiracy involved. The reason it is important to note
that fact is because it is at the heart of the accountability gap. Those
running the process – typically ICANN staff – feel in their hearts they are
doing the right thing, and because of that they feel shielded against
criticism of their work, with the result that useful, constructive criticism
is lost.



The ICANN community has a tendency toward conspiracy and so when a poorly
structured system is continued with, they will tend to question to personal
motives of individuals and point to behind-the-scenes manipulation. Since
this is rarely the case, the overt personal criticism only causes those
running the process to shut themselves down against all criticism.



The result is that only those inside the process are listened to, and so a
balance of views is lost and the end result confirms the fears of those
outside the process – that they aren’t being listened to.



My personal philosophy of why this vicious circle continues to persist is
because ICANN lacks almost entirely a culture of learning from previous
mistakes. Never has it had a review of the effectiveness of a particular
process at the end of that process – despite this going against every best
practice guidebook in the world.



As a result, ICANN doesn’t have any internal checkboxes or red flags that
would indicate something is going along the wrong lines. And, ironically,
the mistakes of the past are frequently repeated because follow from
well-worn paths of travel.



For example, I argued strongly with the EoI, while it was progressing, would
fail because the staff insisted that the process could not be voluntary
(despite quite a few ICANN veterans arguing that it should be). The reality
is that due to ICANN’s makeup, there will always be resistance to anything
that is mandatory.



But rather than accept that fact and try to find ways of creating a
voluntary system that alleviated people’s concerns (thereby ensuring a
successful process), staff in this case stubbornly pushed ahead with a
mandatory system (the logical reasoning of which in the public documents was
weak). The result was the collapse of the whole process.



Likewise with the IRT process – where trademark holders were allowed to draw
up their own rules to decide how the trademark issues with new gTLDs would
be decided. The result was a closed process (fiercely contested by the
community), a hand-picked group of people (all of whom shared the same basic
philosophy) and an end result that ICANN staff tried to force through.
Again, it failed while also causing a lot of ill-feeling along the way.



·       I have other examples but I doubt they will be considered due to
time constraints and the fact that the Berkman Center has already picked out
certain topics, so I will go on to the other questions.









Q2. Do ICANN’s accountability mechanisms, including the Ombudsman, the Board
reconsideration procedure and the Independent Review Panel provide
meaningful accountability and, if not, how could they be improved?



A2. No they do not. If anything they are damaging because they give the
illusion of accountability while not providing any.



The Ombudsman’s authority was very specifically and purposefully limited by
ICANN’s staff and does not comply with standard Ombudsman rules. In reality
he has extremely limited powers to make or force change.



At every step the Board has the ability to ignore the Ombudsman’s
recommendations and, unfortunately, it has done precisely that whenever he
has promoted something that the Board disagrees with. The Board failed to
get back to the Ombudsman about one of his reports for many months, despite
his pointed and public requests for them to do so.



The Board also recently ignored (or punted into the long grass) a series of
important recommendations regarding civility within ICANN.



The answer with the Ombudsman is simple: provide him with the additional
powers that other Ombudsmen take for granted. There is an international
Ombudsman association – I am sure if you asked that association to draw up
the general rules surrounding Ombudsman and then presented them to the ICANN
Board with a very strong recommendation that they not be changed by one word
and implemented fully, then you would have a fully functional Ombudsman
role.



Re: the Board Reconsideration Committee. Unfortunately this has also been a
failure and has only served to reinforce an unhealthy attitude in the Board
that they make the right decisions and that attempts to question them stem
from a hidden agenda.



Most recently the Board was asked to review the fact that its own minutes
were not being produced in the time period that it had set itself. The
result was that the Board simply changed its own rules, publishing bylaw
changes (produced in private) and only publishing the finished bylaws for
public comment.



The few comments to that comment period were ignored. In fact, it’s not
clear whether those bylaws changes have actually been introduced. Whichever
way you cut it, the Board has a long culture and tradition of looking very
unfavourably on any questioning of its decisions and as a result the Board
reconsideration procedure is actually a damaging process of negative
accountability.



The Independent Review Panel is also a very poor accountability mechanism.
It has been used only once and that proved excessively time-consuming and
costly. It took two years and cost millions of dollars.



And, incredibly, when the end result was critical of the ICANN Board’s
decision, its first response was to stress that the IRT Declaration was
non-binding i.e. the Board felt it was able to ignore the entire end result
if it didn’t like it.



What should also be stressed is that ICANN’s staff spent much of the time in
the IRP process arguing how the process itself should run. It lost all its
arguments except one – that the end result should be non-binding. And if you
read the Declaration, the Panelists make it quite clear that the non-binding
nature had been purposefully and deliberately written in by ICANN staff and
that it clashed with the normal rules and procedures for that sort of
arbitration.



In the same way that the Ombudsman authority was deliberately reduced,
stepping away form accepted norms, the IRP process also had its authority
deliberately reduced, stepping away from accepted norms.



What is all the more remarkable – and which ICANN staff and Board are hoping
no one will notice – is that the Board has deliberately not accepted the
conclusions reached by the IRP Declaration with respect to how the process
itself should be run. In fact, the CEO specifically spoke out against one of
the recommendations at a recent public Board meeting.



The end result is that anyone wishing to use the IRP subsequently would need
to run through the same arguments as provided by the first applicant on the
process itself, before even starting to discuss the issue in question.



There is a clear pattern of deliberate and calculated efforts on the part of
ICANN staff and Board to avoid as far as possible any accountability as
introduced by the IRP process. In that sense, it is also a very poor
accountability mechanism.



It may be worth noting at this point that ATRT member Becky Burr – who was
very recently asked to leave the ATRT because she represented the first
applicant of the IRP process – was, I believe, a key drafter of the IRP
rules. It may be instructive to ask her what she knows of deliberate changes
made to the IRP rules that reduced the process’ overall accountability.



I suspect you will find she is unable to disclose any information as she is
bound by a confidentiality contract. You may wish to consider whether the
ATRT should be in a position to ask that ICANN waive such confidentiality in
order to do its work effectively.











Q3. Do you think ICANN’s processes and decision making is transparent? Can
you identify a specific example(s) when ICANN did not act in a transparent
manner. If so, please provide specific information as to the circumstances
and indicate why you believe ICANN’s actions were not taken in a transparent
manner. Are ICANN’s transparency mechanisms robust and how could they be
improved?



A3. I have largely answered this question in my Q1 answer. But, broadly, yes
the process and decision-making is transparent in the sense that you can see
what is going on.



The problem is that parts of the process are not transparent – I outline two
examples in my A1 above. The biggest problem is a cultural one – ICANN staff
maintain a default of not providing information.



As such, unless you very closely follow a process, you will find that things
progress and reach near conclusion without you being aware of it. Those
conclusions are put out to public comment before being adopted but by that
stage it is too late to make anything but cosmetic changes, and people who
have closely followed the process and invested hours of their time in the
results are naturally very resistant to last-minute changes.



The other problem is the way in which deliberations are carried out and made
public. Typically there are long and often tedious conference calls at all
times of day. The end result of these discussions are then provided in long
and often very dense documents. These documents are then posted for public
comment with a single announcement made on the front page of the ICANN
website (and occasional mentions on the many ICANN mailing lists). And then,
presented with a big document, people are asked to comment.



Time and again this process has been heavily criticised as being extremely
ineffective at encouraging contributions, and providing useful feedback. And
yet it persists with little or no effort made to review that process.



A clear example of a very strong resistance to any form of chance in policy
processes that are known to be ineffective is the issue of executive
summaries. For at least five years the community has been calling for every
report to come with an executive summary – standard practice in most
industries.



Eventually these calls led to a document reviewed at no less than three
ICANN meetings and formally approved by the Board that argued specifically
for executive summaries of documents and even provided a template for
producing one.



More than six months after direct Board action, the majority of documents
still do not posses an effective Executive Summary. In many cases the
document starts with something called an executive summary but it serves
more as a summary of the entire document and is laden with information about
the process followed rather than the conclusions reached and the
implications of those processes.



The public comment period is also a highly ineffective one-way system for
receiving input that the community has asked for years be improved. The
result has been a formal and lengthy review that looks unlikely to produce
effective recommendations. And even if good recommendations did emerge from
that process, there remains no mechanism by which anyone can be held
accountable for introducing them or measuring their success or not.









Q4. What is your general assessment of ICANN's commitment to the interests
of global Internet users? Can you provide a specific example(s) when ICANN
did not act in the interests of global Internet users? If so, please provide
specific information as to the circumstances and indicate why you believe
ICANN’s actions were not taken in a manner consistent with the interests of
global Internet users.





A4. I think ICANN intends to look after the interest of global Internet
users but in reality those users do not have much of a voice in the
processes and so any decision made in their interests is largely guesswork.



It is not as simple as pointing to decisions made and saying they were *not*
in global Internet users’ interests, but rather *had* Internet users been a
more valid part of the exercise then very different results would have
emerged.



I am a big fan of looking at specific examples but in this case I would
argue that the ATRT needs to look at where Internet users are in a position
to affect decisions, and what efforts ICANN puts in in order to ensure their
views are heard.



The problem of course is that gathering the views of “global Internet users”
is extremely difficult – there are huge numbers of them; the vast majority
will have to coaxed or dragged into an ICANN policy process; and they will
come with a very wide range of views.



The result is that typically those who make the decisions – ICANN Board
members – are left to decide what it is that global Internet users would
want. And that, understandably, ends up being what the individual Board
member as an Internet user themself wants.



The ALAC is the home for Internet users and it has made great strides in
recent years for gathering broader Internet user view and presenting them in
a useable format. However, so far, the Board has failed to show much respect
of concern for ALAC views, leading in some cases to public requests for the
Board to at least formally recognise they have received advice from the
ALAC.



Cheryl Langdon-Orr is chair of the ALAC and also a member of the ATRT. I
have no doubt that she has much more useful information on this issue that I
could ever provide.











Q5. What is your assessment of the ICANN Board of Directors’ governance with
respect to the following factors:



· ongoing evaluation of Board performance,

· the Board selection process,

· the extent to which Board composition meets ICANN’s present and future
needs , and

· whether an appeal mechanism for Board decisions is needed?





A5. I think the independent review of the Board produced by the Boston
Consulting Group has proved itself time and again as an accurate and
forward-looking summary of problems and solutions to the ICANN Board.



Unfortunately, in a key accountability issue, the Board decided that it
should be in the position to review the review of itself. Unsurprisingly the
Board decided that it worked much better than the BCG felt it did and
disregarded the most important reforms.



The ATRT may wish to review the BCG report and review whether it was helpful
or appropriate for the Board to decide what to do with itself. And ask why
it was that it did not introduce reforms that, in my and many others’
opinions, would make the Board function much better.











Q6. What is your assessment of the role of the GAC and its interaction with
the Board? How do you view the role of the GAC within the overall ICANN
process?



· What is your assessment of the interaction between the GAC and the Board?



· Should the GAC be viewed as the body best placed to advise the Board on
what constitutes the "public interest" regarding the coordination of the
DNS?







A6.  I am actually quite a big fan of the GAC. Even though it is slow and
its meetings are often tortuously dull. And even though only five GAC
members ever seem to say anything, the GAC has consistently produced some of
the best advice and input into ICANN processes.



It has also done more than almost every other group to understand and change
its processes to fit in with the multistakeholder model. It is often fair
and it is often clear. Recently it asked to be included earlier on in policy
processes because it recognised that it was having a detrimental impact if
it provided its view too far down the line.



As to the elevated position it enjoys within the ICANN structure – where the
Board is obliged to consider its advice on public policy issues – I think,
broadly, this is a good, functioning system.



The problem lies not with the GAC, in my view, but with the ICANN Board who
seems to have a schizophrenic attitude toward governments – at times being
unfairly aggressive, and at times cowering and submissive.



The best example of this relationship not working at the moment is the issue
of dot-xxx – where the ICANN Board appears to be using the GAC as a shield
because it doesn’t wish to approve or reject the application itself.











Q7. Are additional steps needed to ensure effective consideration by ICANN
of GAC input on the public policy aspects of the technical coordination of
the DNS? If so, what specific steps would you recommend?



A7. It would be helpful if the Board had the courage to make its
deliberations public.









Q8. What is your assessment of the processes by which ICANN receives public
input? What is your assessment on how ICANN receives input of
English-speaking and non-English speaking communities? ? Can you identify a
specific example(s) when ICANN did not adequately receive public input from
English or non-English speakers? If so, please provide specific information
as to the circumstances and indicate why you believe ICANN’s actions were
taken without adequate public input.



A8. I think the whole process of receiving input is fundamentally flawed,
and I spent nearly three years trying to improve it as ICANN's general
manager of public participation. Unfortunately there is resistance on all
sides to change – particularly when that change brings in more people and
diminishes people’s individual power to influence results.



ICANN as an organization also fails enormously in providing sufficient
resources to improve the process of public input. There is no culture of
checking public input against benchmarks, and very little value put on
public input – it is seen more as an annoyance than a crucial check and
balance. There are also far too many public comment periods, and as result
they are not taken sufficiently seriously by either the staff or the
community.



My former role – General Manager of Public Participation – is charged in the
bylaws as having authority to make changes but the reality is that the role
has no resources or authority internally to make changes. And when faced
with stiff resistance, the only way to introduce change is to provide
someone with the authority to make changes.



There is also a strong American-English bias where the value of providing
material in other languages is completely lost on most.



I produced a leaving report when I left ICANN that summarized my thoughts
and provided clear recommendations. I have seen little indication of that
report being given serious consideration. I would recommend it to the ATRT.



You can find it at:
http://www.icann.org/en/participate/gmpp-leaving-report-25nov09-en.pdf











Q9. Does ICANN provide adequate explanation of decisions taken and the
rationale thereof? Can you identify a specific example(s) when ICANN did not
provide adequate explanation of decisions taken and the rationale thereof?
If so, please provide specific information as to the circumstances and
indicate why you believe ICANN’s actions were taken without adequate
explanation of decisions taken and the accompanying rationale.





A9. No it doesn’t. In terms of examples, it would be easier to simply decide
on other factors what issue you wish to look at and then try to see if you
can find the rationale for making decisions connected to that issue i.e. the
explanation for why decisions are reached is lacking in almost every aspect
of ICANN’s work.



Even when some rationale is provided, to my eyes, that rationale has been
very weak and does not stand up to even basic scrutiny.



That said, there are occasions at the Board level – particularly wrt the
chairman Peter Dengate Thrush - where the rationale for a particular
decision is very clearly and lucidly presented. Even if you disagree with
the end result, you can at least appreciate the logic used to arrive at the
final decision.









Q10. What is your assessment of the extent to which ICANN’s decisions are
embraced, supported and accepted by the public and the Internet community?
Can you identify a specific example(s) when ICANN decisions were not
embraced, supported and accepted by the public and the Internet community?
If so, please provide specific information as to the circumstances and
indicate why you believe ICANN’s actions were taken without adequate support
and acceptance by the public and the Internet community.





A10. Broadly what ICANN decides is accepted. Sometimes because no one wants
to go through the process again; sometimes because a careful balance has
been struck and everyone accepts that it is a workable compromise.



The Board tries not to make decisions that will not be accepted or will
rebound (which conversely means that it is extremely resistant to efforts to
get it to reconsider decisions). The worst example of where this may not
work in the interests of global Internet users is in the case of Whois,
which has been dragged out for nearly 10 years.











Q11. What is your assessment of the policy development process in ICANN with
regard to:



· facilitating enhanced cross-community deliberations, and

· effective and timely policy development



Can you identify a specific example(s) when the policy making process in
ICANN did not facilitate cross-community deliberations or result in
effective and timely policy development? If so, please provide specific
information as to the circumstances and indicate why you believe the policy
making process in ICANN did not facilitate cross-community deliberations or
result in effective and timely policy development.





A11. There are many examples where a failure to have cross-deliberation has
resulted in subsequent argument and wasted time.



But broadly these occasions are growing smaller over time as the different
groups grow increasingly willing to work with one another and start to
understand one another’s cultures.



I do feel obliged to point out however that despite determined efforts,
sometimes by the chairs of the different ACs and SOs themselves, to increase
cross-community discussions, they have on occasion been deliberately
undermined by members of the community and of the staff who have felt,
largely for personal reasons, that such coordination is not helpful. Those
efforts are also in the decline, I am pleased to say.



One very simple mechanism which I believe would help ICANN enormous would be
to give control of some slots at ICANN meetings to community members, where
they are free to arrange and discuss meetings as they desire. This, I think,
would help bring issues to the forefront faster plus allow for innovations
to be made in what are often staid and ineffective formats.









To conclude: I would like to thank the ATRT for considering this submission.
I would be happy to discuss any element in greater depth if requested.



I do remain concerned that the ATRT itself is far from an effective model of
accountability and transparency, and that it is falling into a common ICANN
trap of spending too much time discussing matters other than the actual work
it is there to do.



I would also like to strongly encourage the ATRT to spend much, much more of
its time discussing how to engage the community effectively – and
questioning whether it has received as much engagement as it feels it should
have. And, if it feels it hasn’t, what the reasons may be for why it hasn’t
engaged the community as much as it could have. There is always a reason.



Very last point: use simple input mechanisms – people like to click boxes
and look at images. Long treatises – like this response of mine – only
create more work and more discussion.