EOI must be mandatory or not at all

[Paul Stahura <paul.stahura@xxxxxxxxx>]

Fellow ICANNers,

The purpose of the EOI concept is so that the community can a) know the strings 
- so that we, for example, can help detect undiscovered issues in the proposed 
DAG objection processes, b) know the total number of applications - to know, 
for example, if ICANN's application process personnel resources are sized 
correctly and c) know the number of unique strings - so that we can determine 
issues surrounding root server scaling. If we cannot get this information out 
of the EOI process in a timely manner, then we should not do the EOI process, 
and stick with the original application process outlined in the current version 
of the DAG. Getting the information out of the process in a timely manner means 
we must decide on the EOI process very soon. If it takes three months to 
determine the EOI process (delay being in the economic interests of some 
parties), then that defeats the whole purpose of having it.

I'm going to comment on two issues about the EOI. One is the question of 
whether it should be mandatory, and two, regarding the transfer of EOIs.

1. Regarding the question of having either a voluntary or a mandatory EOI

A cornerstone of the proposed EOI model is that it be compulsory. I believe 
this is the only practical approach to an EOI. Approaches that encourage 
voluntary participation are unfortunately flawed, as there are too many 
incentives for a TLD applicant to not participate in a voluntary EOI, and to 
wait, see the EOI strings of others, then put in an application. A 
non-manditory EOI will not produce the good information we need - which is the 
whole purpose of the EOI in the first place.

An EOI (voluntary or not) creates the following problems for TLD applicants. 
They will have to:

* Obtain and place at risk $55K for an extended period of time
* Prematurely release valuable string and other information to competitors 
(within and without the TLD industry).
* Potentially face premature objections to their string
* Potentially face lobbying of ICANN by competitors

Why would any potential applicant subject himself or herself to these negatives 
if they did not have to?

In order to accomplish the data-gathering goal of the EOI and in order to 
overcome these disincentives, a voluntary program would have to provide strong 
incentives to participate.

One idea is to provide a financial incentive. For example, TLD applicants who 
did not participate in the EOI might face a higher overall fee for their 
subsequent application (e.g. instead of costing $185K for a completed TLD 
application these applicants might face a $225K fee). The challenge with this 
approach is that we do not know what will constitute sufficient incentive for 
voluntary participation. For example, there might be applicants who view the 
additional $40K (per the example above) as an acceptable cost for preserving 
the confidentiality of their business plans. If applicants stay out of the EOI 
it will not serve its information gathering purpose. Any EOI process in which 
ICANN may receive TLD applications without first receiving an EOI for the 
string, means the number of applications may be much greater than the number of 
EOIs. Financial incentives won't work.

Another idea is to stipulate that any string for which there was no EOI cannot 
be applied for during the application phase. In other words, in exchange for 
disclosing the string in an EOI, the applicant (and anyone else) may apply for 
it during the application phase. This does provide a non-monetary incentive to 
make an EOI for those who think no one else will make an EOI for their string, 
but at the same it also incents waiting and not making an EOI because those who 
wait get the benefit of knowledge of other applicants' string information.

ICANN's remit is to provide more fair competition. Not more unfair competition. 
It's not fair competition for one competitor to be forced to disclose their 
valuable string information while allowing another competitor not to disclose 
it We can't allow them both to compete in applying for the same string where 
one party knows the other's string information, but not visa versa. The party 
that submitted an EOI got the benefit (it is allowed to submit an application 
for the string), and paid the cost (disclosure of the string), but their 
competitors got the benefit (being allowed to submit an application for the 
string) without paying the cost (disclosing it). That is not fair competition 
and it won't work.

A voluntary EOI will not produce an accurate count of the number of 
applications that will be received because with a voluntary EOI, no matter what 
the incentives are, monetary or not, a potentially unlimited number of 
applications can be received after the EOI period is closed. The predictive 
value of the EOI phase (the total number of applications, and the number of 
unique strings, in this case) will be lost with a voluntary EOI. The only 
incentive that works is to make the EOI mandatory.

2. Regarding the question of trading or transferring EOIs.

Some are saying that an EOI phase (mandatory or not) will create incentives to 
submit an EOI for the purpose of selling the EOI for a profit, with no 
intention on the part of the EOI applicant to pursue the application. I don't 
agree with them. People will not put in EOIs just to flip them because there 
will be no demand for the EOI string-slots.

Demand for the string-slots can only possibly come from two types of parties - 
from either other EOI applicants, or from parties who did not submit an EOI. 
Lets look at both possibilities.

2.1 No demand from other EOI applicants

An EOI applicant will not pay another for its EOI because its not sure the 
other will pay the application fee. Why pay for something now that you can get 
free later - when the other party does not file the application and pay the 
application fee. This uncertainty and other problems (such as in the case where 
more than two EOIs are received for the same string) will keep other EOI 
players away from buying.

2.2 No demand from parties who did not submit an EOI

A potential buyer who did not put in an EOI either did not want the string in 
the first place, or may want it but did not know the process happened. If that 
buyer did not want the string, then they will not be buying the EOI-string off 
of another party who put in an EOI. So the EOI seller will only be selling the 
string to a potential buyer who did not want the string. From the seller's 
perspective, there is no way to know (without contacting potential buyers) that 
the buyer wants the string AND does not know of the process (since they are not 
contacted), before the process starts. In other words, there is great risk that 
there is actually some buyer out there who wants the string but will not submit 
and EOI because they do not know the process exists... despite the outreach by 
ICANN, and the trademark community, and the heavy marketing by back-end 
registry providers over the last two years. This risk, plus the risk that that 
unknown potential buyer will buy the EOI string for more than the EOI fee means 
that there will be few parties attempting to make money from flipping EOIs, and 
no smart ones.

And anyway, even if I'm wrong and there is demand, so what? What's wrong with 
someone taking a risk (huge risk in my opinion that there will be only one EOI 
for this string and that a buyer clueless to the EOI process will materialize) 
and submitting an EOI, say ".cool", then doing their own private outreach to 
the cool community and finding someone who suddenly wakes up to discover the 
EOI window is closed but somehow has a hankering for the ".cool" TLD? At least 
the EOI string- slot got into the hands of someone who believes in it enough to 
do something with it - which is a good thing.

Conclusions

1 - If there is an EOI, it must be mandatory.

2 - There will not be demand to buy EOIs so making EOI submissions for the sole 
purpose of flipping them is a high-risk venture and will not occur, and even if 
it does happen, there is no harm.

Best Regards,



Paul Stahura