FINAL VERSION SUBMITTED (IF RATIFIED)
The final version to be submitted, if the draft is ratified, will be placed here by upon completion of the vote.
FINAL DRAFT VERSION TO BE VOTED UPON BY THE ALAC
The final draft version to be voted upon by the ALAC will be placed here before the vote is to begin.
ALAC Statement on Proposal for Latin Script Root Zone Label Generation Rules
The At-Large Advisory Committee (ALAC) welcomes the opportunity to comment on the proposed Latin Script Root Zone Label Generation Rules. The ALAC appreciates the efforts of the Latin Generation Panel (LGP) but questions some of the instructions received by and assumptions made by the panel. These “quirks” have led to a proposal that might not be the best for multilingual expression or the mitigation of DNS abuse.
Principle of Least Astonishment
The principle of least astonishment proposes that a component of a system should behave in a way that most users will expect it to behave. The now expert nature of the Latin GP suggests that it is no longer, if ever, representative of “most users.” Therefore, establishing a high bar for possible confusion is counter intuitive. If even a single Panel member considers that two code points should be a variant, that conclusion should be applied to the Panel findings. Otherwise, the Panel is implying that non-experts will be more successful in spotting differences than some of its experts. This is an entirely unreasonable expectation of a typical end user and making this assumption further paves the way for misdirection and consequent DNS abuse.
Underlining
The Latin GP has concluded that underlining of domain names (as happens routinely) does not obscure diacritics below the line. While some software products put a blank space around such diacritics, it is not at all clear that end users are cognizant of this practice. Furthermore, many software products, including ICANN’s own Public Comment system, do not use these spaces. Accordingly, the ALAC recommends revising that conclusion.
Capital Letters
The ALAC notes, in the Latin Generation Panel’s report, that the Internationalized Domain Name (IDN) Project has directed the Generation Panels (GPs) to ignore upper case letters when identifying variants. The apparent rationale for this direction is that domain names are restricted to lower case. Unfortunately, a half century of experience has taught end users that, in a domain name, upper and lower case are entirely interchangeable. Thus, a domain name of www.example.com and of www.example.COM will take the user to the same place. The users neither know nor care that this is an artifact of their browser, rather than a feature of the DNS.
As a result, if the user is presented with a domain name using Cyrillic lower case for the TLD, such as www.example.сом, they will naturally interpret it to be the usual .com, just with the TLD capitalized. What they will NOT do, despite in expectation of the IDN Project, is look at the TLD and notice the third letter doesn’t look like a Latin lower case M and conclude this is not a .com domain name. The potential for DNS abuse is obvious.
Accordingly, the ALAC encourages the LGP to revisit the instructions, attributed to the IDN Project, and consider treating capital letters as a special case.
Repertoire
The objective of the IDN project is to make domain names available in the languages of all non-native speakers of English. To restrict the repertoire for the Latin script to less than half of the living languages which use that script is contrary to that goal. The ALAC believes that the LGP should go back and include all of the languages which use the Latin script. (At a minimum, languages which have more native speakers than the smallest of the “official languages” which are already included, should be added.) For example, Hawaiian has perhaps 25,000 native speakers, but is included because it is an official language of the State of Hawaii, despite the stated threshold of 1 million speakers.
Conclusion
It is the considered position of the ALAC that, while great strides have been made by the Latin Generation Panel, more must be done to enable more users, with sufficient security, to use the DNS. Again, we appreciate the opportunity to comment on the draft proposal and look forward to further discussions on this topic.
DRAFT SUBMITTED FOR DISCUSSION
The first draft submitted will be placed here before the call for comments begins. The Draft should be preceded by the name of the person submitting the draft and the date/time. If, during the discussion, the draft is revised, the older version(S) should be left in place and the new version along with a header line identifying the drafter and date/time should be placed above the older version(s), separated by a Horizontal Rule (available + Insert More Content control).
ALAC Statement on Proposal for Latin Script Root Zone Label Generation Rules
The At-Large Advisory Committee (ALAC) welcomes the opportunity to comment on the proposed Latin Script Root Zone Label Generation Rules. The ALAC appreciates the efforts of the Latin Generation Panel (LGP) but questions some of the instructions received by and assumptions made by the panel. These “quirks” have led to a proposal that might not be the best for multilingual expression or the mitigation of DNS abuse.
Principle of Least Astonishment
The principle of least astonishment proposes that a component of a system should behave in a way that most users will expect it to behave. The now expert nature of the Latin GP suggests that it is no longer, if ever, representative of “most users.” Therefore, establishing a high bar for possible confusion is counter intuitive. If even a single Panel member considers that two code points should be a variant, that conclusion should be applied to the Panel findings. Otherwise, the Panel is implying that non-experts will be more successful in spotting differences than some of its experts. This is an entirely unreasonable expectation of a typical end user and making this assumption further paves the way for misdirection and consequent DNS abuse.
Underlining
The Latin GP has concluded that underlining of domain names (as happens routinely) does not obscure diacritics below the line. While some software products put a blank space around such diacritics, it is not at all clear that end users are cognizant of this practice. Furthermore, many software products, including ICANN’s own Public Comment system, do not use these spaces. Accordingly, the ALAC recommends revising that conclusion.
Capital Letters
The ALAC notes, in the Latin Generation Panel’s report, that the Internationalized Domain Name (IDN) Project has directed the Generation Panels (GPs) to ignore upper case letters when identifying variants. The apparent rationale for this direction is that domain names are restricted to lower case. Unfortunately, a half century of experience has taught end users that, in a domain name, upper and lower case are entirely interchangeable. Thus, a domain name of www.example.com and of www.example.COM will take the user to the same place. The users neither know nor care that this is an artifact of their browser, rather than a feature of the DNS.
As a result, if the user is presented with a domain name using Cyrillic lower case for the TLD, such as www.example.сом, they will naturally interpret it to be the usual .com, just with the TLD capitalized. What they will NOT do, despite in expectation of the IDN Project, is look at the TLD and notice the third letter doesn’t look like a Latin lower case M and conclude this is not a .com domain name. The potential for DNS abuse is obvious.
Accordingly, the ALAC encourages the LGP to revisit the instructions, attributed to the IDN Project, and consider treating capital letters as a special case.
Repertoire
The objective of the IDN project is to make domain names available in the languages of all non-native speakers of English. To restrict the repertoire for the Latin script to less than half of the living languages which use that script is contrary to that goal. The ALAC believes that the LGP should go back and include all of the languages which use the Latin script. (At a minimum, languages which have more native speakers than the smallest of the “official languages” which are already included, should be added.) For example, Hawaiian has perhaps 25,000 native speakers, but is included because it is an official language of the State of Hawaii, despite the stated threshold of 1 million speakers.
Conclusion
It is the considered position of the ALAC that, while great strides have been made by the Latin Generation Panel, more must be done to enable more users, with sufficient security, to use the DNS. Again, we appreciate the opportunity to comment on the draft proposal and look forward to further discussions on this topic.
3 Comments
Gopal Tadepalli
Pitinan Kooarmornpatana
I am sorry, the Google Doc is not allowing me to comment. Edit Suggestions for consideration are given below.
First para, second line: The ALAC appreciates the efforts of the Latin Generation Panel (LGP) to address the internet language barrier but questions....
First para, third line: ......mitigation of DNS abuse in the rapidly advancing internet technologies and devices.
Gopal Tadepalli
Evin Erdogdu
Very sorry for getting back to this page only now. However, I am sure that, ICANN Secretariat has a way of reaching out to the concerned "Process Owner" for this page as appropriate. As you would appreciate, we have to cope up with the "Computer Supported Collaborative Work (CSCW)" such as the one happening with ICANN Community Wiki.
Good References:
Quick Point to Check:
Sorting (or collating for the technical term) is a complicated matter, already within one language but even more so if there is a mixed bag of languages. There is no context associated with a domain name string. There is no need to know in which language it is. It is just a string with characters (even between all languages using only ASCII the sorting order is not the same). Many European languages are descendants of Latin.
Searching any string and sorting in DNS queries is a common requirement and it is usually a huge number of a huge number of DNS queries or queried hostnames. Please see if this pot is relevant to this document titled "Proposal for Latin Script Root Zone Label Generation Rules" being drafted.
Evin Erdogdu
Thank you, Dr. Gopal, for providing your comments on the ALAC statement.