Logo

Terri Agnew's Personal Meeting Room - Shared screen with speaker view
Cheryl Langdon-Orr
36:25
indeed we do @Jim
Jeff Neuman
36:44
you can chalk that up to the American Holiday and the world's toelerance of that
Julie Bisland
36:57
Welcome to the New gTLD Subsequent Procedures Working Group on Tuesday, 27 August 2019 at 03:00 UTC for 90 minutes.
Cheryl Langdon-Orr
37:38
Shall do @Jeff
Jim Prendergast
41:22
same
Emily Barabas
44:10
Section 2.8.3
Emily Barabas
50:20
“This preliminary recommendation addresses string similarity evaluation and suggests “[p]rohibiting plurals and singulars of the same word within the same language/script.” The preliminary recommendation further went on to require “[u]sing a dictionary to determine the singular and plural version of the string for the specific language.” The requirement to use a dictionary to determine singular/plural form of a word limits the singular/plural determination to a single language, and not script. It would be helpful if the PDP Working Group could provide clarification.”
Justine Chew
54:32
@Jeff, but that may not diffuse end user confusion.
Paul McGrady
55:51
@Jeff, are we talking about confusion at the use level rather than at the string level?
Cheryl Langdon-Orr
56:24
This clarification however by my reading is fcused on the reference tool for look up (in this case a choice of 'Dictionary'needs to be made) and established as a mechanism for determination of the plurality or not.
Paul McGrady
57:59
I'm not sure how an errant/typo email would end up in the right place because of what is in an application.
Justine Chew
58:27
@Jeff, but the end user doesn't necessarily relate to what is stated per applicant's purpose
Paul McGrady
01:00:14
@Jeff, your .new and .news example is a good one of how the "s", in limted cases, creates a different impression
Justine Chew
01:00:38
I understand the challenges though .....
Cheryl Langdon-Orr
01:00:45
Thus the need for a reference tool (dictionary) @Paul
Paul McGrady
01:02:15
Would we have to pick a dictionary for each language?
Cheryl Langdon-Orr
01:02:58
an important 'ímplementation' detail @Paul
Emily Barabas
01:03:26
“Regarding the suggestion to use a dictionary to determine the singular/plural form of a word, it should be noted that a word may be identical in many languages, but generate different plural forms in each of the languages. For example, ‘kitab’ is ‘book’ in Arabic, Persian, Urdu, and many other languages, but the plural form of the word is different in each of the languages. It should also be noted that in many languages, reduplication is used as a means to make words plural where the word is repeated to make its plural form instead of adding a suffix like ‘s’. Additionally, some of these plural forming reduplications also make morpho-phonological changes, which means that the surface form is not just the word repeated by itself but slightly different. Even if the assessment of plural is limited to non-reduplication, it should be noted that pluralization is not limited to just suffixation, but can also include infixes (e.g., in Arabic, KiTaB (sg.) à KuTB (pl.)."
Emily Barabas
01:03:34
“Further, as the context of TLDs is generally not known when in use by the end-user, limiting the confusability assessment to a single language (and language-based dictionary) may provide a limited context for confusability. The PDP Working might want to consider whether the contention set should be expanded to the languages of the users rather than the language identified by the applicant. This, of course, would make the confusability analysis very broad, which should be factored into the consideration.”
Emily Barabas
01:04:05
(excerpt from the ICANN Org comment)
Justine Chew
01:04:09
meaning the same string could appear in more than one language which means there is more than one set of end users
Paul McGrady
01:04:30
This is starting to feel like a Sean Penn film. :)
Justine Chew
01:04:52
Which one @Paul? The Professor and The Madman?
Paul McGrady
01:05:00
@Justine - exactly!
Justine Chew
01:06:19
cross-referencing dictionaries? need linguistic expertise.
Emily Barabas
01:06:45
From the comment: “To inform the PDP Working Group’s continued discussion on this topic, it should be noted that number is only one of a few different ways a word can be inflected in a language. Across languages of the world, there are many other ways a word may be inflected such as gender (e.g., gato and gata in Spanish), person (e.g., mange and manges in French), respect (e.g., manges and mangez in French), tense (e.g., walk and walked), and others. The PDP Working might want to consider whether to expand the confusability assessment to other forms of inflection as well."
Cheryl Langdon-Orr
01:09:30
why not indeed @Sarmad
Paul McGrady
01:10:32
How do we boil all of this down in a way that would provide even a modicum of predictabiilty to an applicant?
Justine Chew
01:11:11
Or to end users.
Cheryl Langdon-Orr
01:11:42
Indeed @Paul amd setting some principles such as in terms of a look up to an authoratiative reference, would IMO aid such predictability
Justine Chew
01:12:56
We are eliminating the SWORD tool!
Paul McGrady
01:13:19
+1 Justine.
Christopher Wilkinson
01:14:24
.CAT is already delegated
Justine Chew
01:14:51
That's why Jeff said ".CATS"
Sarmad Hussain
01:15:07
Would “decide” and “decides” be allowed
Paul McGrady
01:16:27
Verb loophole...
Christopher Wilkinson
01:17:27
I like the idea of side shows (pl) but that begs the question: before who? Perhaps we need a sematinc review panel to advise on difficult cases. @ Paul: full predictability for applicants is not possible in this area.
Justine Chew
01:18:54
@Sarmad, from where would we draw such inflections?
Justine Chew
01:19:03
+1 Jeff
Sarmad Hussain
01:19:14
Inflections are given in a dictionary
Vaibhav Aggarwal, IN
01:19:30
@Christopher : It is a Window of Opportunity for a Legally Binding Agreement between the Two APPLYING parties can be executed if there is meeting of minds.
Justine Chew
01:20:57
Is there a list of prominently accepted dictionaries around the world?
Vaibhav Aggarwal, IN
01:21:06
But since there’s no policy, currently, there is no possibility of legally doing / combining strings - My point is Simple - Why Stop / Curtail the opportunity for the End User ?
Paul McGrady
01:22:05
Can we have another example? You guys are smarter than me. Mouse/mice?
Paul McGrady
01:22:42
Run/Ran?
Paul McGrady
01:22:47
Thanks!
Vaibhav Aggarwal, IN
01:22:48
There are plenty of eg.s and precedence in the 2012 Round as well
Emily Barabas
01:24:26
“An additional consideration is that as strings are frequently labels and not words in a language, the PDP Working Group might also want to consider how the rules in this preliminary recommendation may be applied to labels that are not words in a language. For clarity, the PDP Working Group might want to consider sub-categorizing different forms of string similarity evaluation. For example, defining a separate sub-category of "grammatical similarity" that would include number (singular/plural) and any other grammatical similarity that the PDP Working Group wishes to include such as gender, person, etc., in addition to and separate from the "visual similarity" analysis. “
Paul McGrady
01:25:13
Calm is an adjective and calms is a verb, right? Does that make a difference?
Emily Barabas
01:25:27
“It may also be important to clarify the difference between variant labels and similar labels and that in cases where variant labels are also similar labels, that the variant relationship takes precedence. Variant labels are defined by the community through the Root Zone Label Generation Rules. This would especially need to be addressed for the (less intuitive) cross-script variant labels. For some additional discussion, see Section 9 of the report on Rationale for RZ-LGR recently published for public comment."
Justine Chew
01:25:58
What is an example of labels that are not words in a language?
Paul McGrady
01:26:47
@Justine, we had non-word applications, like .xyz
Jim Prendergast
01:27:00
Shouldn't there be some coordination between this group and others if there is work going on elsehwere? Or we risk two possible standards which is less than ideal.
Justine Chew
01:27:25
@Paul, okay, so how does singular/plural come into play then?
Paul McGrady
01:27:43
.xyzs? :)
Justine Chew
01:27:53
LOL!
Justine Chew
01:28:23
I think Sarmad makes more sense ;)
Cheryl Langdon-Orr
01:28:55
Deffinatly appreciated @Sarmad
Jim Prendergast
01:29:28
https://acronyms.thefreedictionary.com/CARS
Paul McGrady
01:30:40
@Jeff, the .com/coms example is a great one when considering email and other collisions.
Justine Chew
01:31:58
Like how would the applicant address risk of confusion?
Alexander Schubert
01:32:19
Maybe once the AGB is finalized and published ICANN should be offering applicants to submit a string online - and within a reasonable time get a response whether the string is eligble (maybe for a small fee of US $250)? Google for example managed to apply for THREE such ineligble strings. So ICANN would test the string in regard to all provisions on the AGB: geo-names, already applied strings, plurals, censored strings like ".isis", blocked names like .gac, etc.
Vaibhav Aggarwal, IN
01:32:46
@Justine By creating Requirements for the NGTLD
Jeff Neuman
01:33:30
Thanks Paul for the example.
Jim Prendergast
01:34:00
Good point by Paul
Justine Chew
01:34:07
+1 Paul
Alexander Schubert
01:34:15
Agree with Paul.
Christopher Wilkinson
01:34:24
CAR is the usual acronym for the Central African Republic. I would have to look up the ISO3166 alpha three code for the CAR, capital Bangui.
Paul McGrady
01:34:36
.co and .com
Vaibhav Aggarwal, IN
01:35:02
@Paul Ha Ha - There is High Level of Skewed-ness in that
Paul McGrady
01:35:30
@Jeff, thanks!
Cheryl Langdon-Orr
01:39:32
I suspect so @Jeff
Cheryl Langdon-Orr
01:40:37
Take that to the list @Jeff
Paul McGrady
01:41:24
For clarity, we aren't talking about Legal Rights Objection, correct?
Justine Chew
01:41:44
String Confusion Objections
Paul McGrady
01:42:12
Thanks!
Paul McGrady
01:43:09
It was worse than not helpful, because it gave false comfort.
Alexander Schubert
01:45:40
See my previous remark: Why not offering a manual submission of the string to ICANN so they test the string within a few days for a fee?
Paul McGrady
01:46:45
@Alexander - an interesting idea
Justine Chew
01:47:00
@Alex, I think the issue is more to do with test against what?
Alexander Schubert
01:47:24
Same test as later in the evaluation
Cheryl Langdon-Orr
01:50:28
List first then come to the plenary after those deliberations
Alexander Schubert
01:50:29
Good example
Cheryl Langdon-Orr
01:53:59
10 mins left @Jeff
Paul McGrady
01:54:06
Too bad. That will be a fun day.
Jeff Neuman
01:55:33
All - GREAT discussion on the similarity issues. I certainly came out of this discussion with a much better understanding of the issues and have learned a ton!
Paul McGrady
01:56:45
+1 Jeff - I think this was a model call
Julie Hedlund
01:58:23
Yes, that would be placeholder language until we get closer.
Julie Hedlund
01:58:27
for the published schedule
Julie Hedlund
01:58:38
We will update as we get closer
Paul McGrady
01:59:53
When is the small group meeting?
Julie Hedlund
02:00:15
@Paul: I think the discussion will be on the new small group list.
Vaibhav Aggarwal, IN
02:00:17
Good Call. Dropping off now have to head to another Call. Cheers
Justine Chew
02:00:21
@Julie H, could you please add me to the Closed Generics small group? Thanks.
Paul McGrady
02:00:32
Thanks Jeff!
Julie Bisland
02:00:38
NEXT CALL: Thursday, 29 August 2019 at 20:00 UTC for 90 minutes
Paul McGrady
02:00:41
Julie, can you add me to that list?
Julie Hedlund
02:00:46
@Justine, if you could send an email to the list that would be helpful.
Julie Hedlund
02:00:55
The GNSO Secretariat is tracking
Justine Chew
02:00:58
@Julie H, sure, will do.
Julie Hedlund
02:01:04
Thanks11
Heather Forrest
02:01:10
Thanks, Jeff, all.
Paul McGrady
02:01:13
Wil do!
Cheryl Langdon-Orr
02:01:14
Excellent discussion and exchanage of onfromation on todays call... Thanks for your attendance and contibutions... Until our next call this week at 2000UTC on Thursday... Bye for now...
Julie Hedlund
02:01:22
Thanks everyone!