I don't think your reply is so weak, especially when it quotes the DAG3 section 3.4.1. 
Really what he seems to be arguing is that an incumbent TLD operator can control the "meaning" of a TLD, not just string confusion. 
When Chuck quotes trademark treaties you might remind him that trademark law applies to a wide variety of identifiers, not just domain names, and that TLDs have no colors, are rarely "pronounced" as words or normal names are pronounced - so a lot of the TM confusing similarity stuff just doesn't apply. 

> -----Original Message-----
> From: Non-Commercial User Constituency [mailto:NCUC-
> [log in to unmask]] On Behalf Of Avri Doria
> Sent: Friday, April 16, 2010 10:27 PM
> To: [log in to unmask]
> Subject: Re: [NCUC-DISCUSS] are confusing similar gTLDs not confusing if
> the same registry has them all?
> 
> Hi,
> 
> >>  Or is that battle already lost?
> >
> > I do not believe it is lost.  The DAG from m reading still upholds the
> visual criteria for confusing similarity.
> >
> > It is just that Chuck and others repeat that it includes meaning so
> often and so absolutley that it is gaining currency and while I don't
> find it in my reading of DAGv3, they claim it is a viable option for
> objection by a current string holder.
> 
> But Chuck is going to fight tooth and nail for the position.
> 
> The following is his argument.  i will be working on a response in the
> next days although i already sent a preliminary response which i will
> append after his message.
> 
> 
> >
> > Let me first start with your minority statement that accompanied the
> > final GNSS Report:
> > http://gnso.icann.org/issues/new-gtlds/pdp-dec05-fr-parta-
> 08aug07.htm#_T
> > oc48210874.  If you believed that the GNSO report defined confusing
> > similarity as visual only and that that was a position of the GNSO
> > Council, then it would not have been necessary for you to express the
> > following reservations about that recommendation:
> >
> > "In the first instance I believe that this is essentially a technical
> > issue that should have been resolved with reference to typography,
> > homologues, orthographic neighbourhood, transliteration and other
> > technically defined attributes of a name that would make it
> > unacceptable. There is a large body of scientific and technical
> > knowledge and description in this field that we could have drawn on."
> >
> > "By using terms that rely on the legal language of trademark law, I
> > believe we have created an implicit redundancy between recommendations
> 2
> > and 3. I.e., I believe both 2 and 3 can be used to protect trademarks
> > and other intellectual property rights, and while 3 has specific
> > limitations, 2 remains open to full and varied interpretation."
> >
> > "As we begin to consider IDNs, I am concerned that the interpretations
> > of confusingly similar may be used to eliminate many potential TLDs
> > based on translation. That is, when a translation may have the same or
> > similar meaning to an existing TLD, that the new name may be
> eliminated
> > because it is considered confusing to users who know both."
> >
> > Now let me focus on the language in the final report.  Please see the
> > Recommendation 2 Discussion in the report, the first item under the
> > subtitle "TERM OF REFERENCE -- SELECTION CRITERIA".  I first call your
> > attention to the first item: "i) This recommendation has support from
> > all the GNSO Constituencies. Ms Doria accepted the recommendation with
> > the concern expressed below[39]."  As you know, your concerns are the
> > ones pasted above.
> >
> > Item iii confirms that the issue discussion that ". . . the issues
> found
> > below have been discussed at length, both within the Committee and
> > amongst the Implementation Team."  The discussion goes on in item iv
> to
> > say, "The Committee used a wide variety of existing law[42],
> > international treaty agreements and covenants to arrive at a common
> > understanding that strings should not be confusingly similar either to
> > existing top-level domains like .com and .net or to existing
> > trademarks[43]."  I won't quote all of them here because there are a
> lot
> > of them but let me quote a few that are particularly relevant to our
> > discussion now.
> >
> > Item vii says, ". . .  the 1883 Paris Convention on the Protection of
> > Industrial Property[48]. It describes the notion of confusion and
> > describes creating confusion as "to create confusion by any means
> > whatever"".
> >
> > Item x says, ". . . the European Union Trade Mark Office provides
> > guidance on how to interpret confusion. "...confusion may be visual,
> > phonetic or conceptual. A mere aural similarity may create a
> likelihood
> > of confusion."
> >
> > Item xi says, ". . . Likelihood of association is not an alternative
> to
> > likelihood of confusion, "but serves to define its scope". Mere
> > association, in the sense that the later mark brings the earlier mark
> to
> > mind is insufficient to find a likelihood of confusion, unless the
> > average consumer, in bringing the earlier mark to mind, is led to
> expect
> > the goods or services of both marks to be under the control of one
> > single trade source. "The risk that the public might believe that the
> > goods/services in question come from the same undertaking or, as the
> > case may be, from economically-linked undertakings, constitutes a
> > likelihood of confusion...". (found at
> > http://www.patent.gov.uk/tm/t-decisionmaking/t-law/t-law-manual.htm)"
> >
> > There is of course much more in this discussion than I quoted above
> but
> > I don't there is anything that says that visual similarity is the only
> > area of possible confusion.  If you can find anything in the report
> that
> > says that, please point it out.
> >
> > As you know, the ICANN Staff implementation team attempted in their
> > early steps to limit confusing similarity to visual similarity only
> and
> > I challenged them on that several times, pointing them to the
> discussion
> > section mentioned above. Consequently, in DAG3 we find the following:
> > "The visual similarity check that occurs during Initial Evaluation is
> > intended to augment the objection and dispute resolution process (see
> > Module 3, Dispute Resolution Procedures) that addresses all types of
> > similarity. (From DAG3 section 2.1.1.1)"  The standards for a string
> > confusion dispute are found in DAG3 section 3.4.1: "A DRSP panel
> hearing
> > a string confusion objection will consider whether the applied-for
> gTLD
> > string is likely to result in string confusion. String confusion
> exists
> > where a string so nearly resembles another that it is likely to
> deceive
> > or cause confusion. For a likelihood of confusion to exist, it must be
> > probable, not merely possible that confusion will arise in the mind of
> > the average, reasonable Internet user. Mere association, in the sense
> > that the string brings another string to mind, is insufficient to find
> a
> > likelihood of confusion."  Note there is no restriction to visual
> > similarity only.
> 
> 
> my weak response:
> 
> ---
> 
> It is late and I will write in detail with chapter verse and
> interpretation later.  You will possibly be surprised that i use many of
> the same lines you have quoted to make my point.
> 
> A few quick points now
> 
> - As i said the fact that i expressed a concern that something might
> happen is not an acknowledgment  that this thing has been permitted.   I
> was predicting the possibility of discussions such as we are having.
> 
> - Yes, we discussed many contributions to the point we reached. But none
> of those discussions firmly established a single standard beyond visual.
> The fact that the background material from the EU includes conceptual
> does not mean that this was accepted as part of the standard by the
> council.  Though of course it might be relevant in a court case, I don't
> know.
> 
> - I do agree that the staff has fuzzed up the borders a wee bit on
> confusing similarity and yes the DAGv3 does define it a way that allows
> confusion and that if it stays this way it will eventually  be a court
> that decides what it really means.  DAGv1 was much better in this
> respect.
> 
> - you quoted
> 
> DAG3 section 3.4.1: "A DRSP panel hearing
> a string confusion objection will consider whether the applied-for gTLD
> string is likely to result in string confusion. String confusion exists
> where a string so nearly resembles another that it is likely to deceive
> or cause confusion. For a likelihood of confusion to exist, it must be
> probable, not merely possible that confusion will arise in the mind of
> the average, reasonable Internet user. Mere association, in the sense
> that the string brings another string to mind, is insufficient to find a
> likelihood of confusion."
> 
> This talks about 'resemblance' which is a term that specifically refers
> to how things look.
> 'Mere association' on the other hand is the best one ever get out of
> translation.
> 
> What this has done, is made the issue of transliteration more difficult
> to argue and this because establishing whether the average Internet user
> is one who can look at hebrew letters and ascii letter and visually see
> the same thing could be a little challenging.  One would also have to be
> able to prove that there was an exception in the transliteration.
> 
> More later.
> 
> ----