Comment Close
Date
Statement
Name 

Status

Assignee(s) and
RALO(s)

Call for
Comments
Call for
Comments
Close 
Vote
Announcement 
Vote OpenVote
Reminder
Vote CloseDate of SubmissionStaff Contact and EmailStatement Number
11.03.2014Translation and Transliteration of Contact Information Working Group SO-AC Input RequestADOPTED
11Y, 0N, 0A

Rinalia Abdul Rahim (APRALO)

13.02.201410.03.201411.03.2014
00:01 UTC 
11.03.2014
00:01 UTC 
14.03.2013

16.03.2014 00:01UTC

17.03.2014Glen de Saint Géry
gnso.secretariat@gnso.icann.org
AL-ALAC-ST-0314-04-00-EN

Dear Olivier,

As you may be aware, the GNSO Council recently initiated a Policy Development Process (PDP) on the Translation and Transliteration of Contact Information; the relevant Issue Report can be found here. A more detailed background is available online on the Working Group’s Wiki where you can also consult the Charter. As part of its efforts to obtain broad input from the ICANN Community at an early stage of its deliberations, the Working Group would very much appreciate receiving your views.

Any provision of input or information you or members of your respective communities may have (either on the charter questions or any other issue that may help inform the deliberations) would be very welcome. Please send these to the GNSO Secretariat (gnso.secretariat@gnso.icann.org) who will forward these to the Working Group. If possible, the WG would greatly appreciate if it could receive your input by Tuesday 11 March 2014 at the latest. Please note, if you cannot submit your input by that date, but your group would like to contribute, please let us know when we can expect to receive your contribution so we can plan accordingly.

Your input will be very much appreciated.

With best regards,

Chris Dillon (Co-Chair)

Rudi Vansnick (Co-Chair)

Input Request

Translation and Transliteration of Contact Information Charter Questions

  1. Whether it is desirable to translate contact information to a single common language or transliterate contact information to a single common script
  2. What exactly the benefits to the community are of translating and/or transliterating contact information, especially in light of the costs that may be connected to translation and/or transliteration?
  3. Should translation and/or transliteration of contact information be mandatory for all gTLDs?
  4. Should translation and/or transliteration of contact information be mandatory for all registrants or only those based in certain countries and/or using specific non-ASCII scripts
  5. What impact will translation/transliteration of contact information have on the WHOIS validation as set out under the 2013 Registrar Accreditation Agreement?
  6. When should any new policy relating to translation and transliteration of contact information come into effect?
  7. Who should decide who should bear the burden translating contact information to a single common language or transliterating contact information to a single common script? This question relates to the concern expressed by the Internationalized Registration Data Working Group (IRD-WG) in its report that there are costs associated with providing translation and transliteration of contact information. For example, if a policy development process (PDP) determined that the registrar must translate or transliterate contact information, this policy would place a cost burden on the registrar.
  8. Who does your SG/C believe should bear the cost, bearing in mind, however, the limits in scope set in the Initial Report on this issue?

Please click here to download a copy of the PDF below.

FINAL VERSION TO BE SUBMITTED IF RATIFIED

Please click here to download a copy of the PDF below.

 

FINAL DRAFT VERSION TO BE VOTED UPON BY THE ALAC

ALAC Statement on Translation and Transliteration of Contact Information

The ALAC provides the following responses to the Translation and Transliteration of Contact Information Working Group Charter questions: 

(Note:  The use of the word “transformation” of contact information pertains to the “translation and / or transliteration” of contact information.)

 

(1) Whether it is desirable to translate contact information to a single common language or transliterate contact information to a single common script

In principle, translating contact information to a single language privileges the users of that language over other users, just as the transliteration to a single script privileges users of that particular script over others.  The choice of one single language or script does not reflect or support a global Internet community, which is diverse in language and script use. 

Ideally, registration contact information should be available in languages and scripts that reflect the global diversity of Internet users.  We recognize that the current WHOIS implementation is based on a limited 7-bit ASCII that cannot properly render many languages and scripts.  We are also aware that once the IETF completes its work on the Web Extensible Internet Registration Data Service (WEIRDS), a WHOIS database that can handle multiple scripts and character sets will be available. 

Irrespective of whether contact information is transformed or not, we strongly believe that verifiable and verified contact information must be accessible to those with a right and need to access it via the WHOIS database.  The availability of contact information that is validated for accuracy and usability promotes consumer trust in the Domain Name System and is beneficial to the At-Large community.

Until a fully internationalized WHOIS is available, ensuring that registrations created using non-ASCII-based scripts have meaningful and usable contact information in the current ASCII WHOIS can be achieved in various ways:

One option is to transform the contact information into a single language or specific set of languages using scripts that are representable within the constraints of the current WHOIS.  Another option is to have the ASCII-based WHOIS record point to a non-ASCII based record, which would be maintained by the Registry and in parallel with the ASCII WHOIS.

When considering the transformation of contact information (i.e., whether to translate or transliterate), transformation to a level where it is understandable to users of registration data may require a combination of both translation and transliteration.  For example, when transforming contact information, proper names or nouns (i.e., unique names of persons, places, events and things) should not be translated and should instead be transliterated.  Translation and / or transliteration may thus be appropriate for specific parts of WHOIS as identified in the Issue Report.

 

(2) What exactly the benefits to the community are of translating and/or transliterating contact information, especially in light of the costs that may be connected to translation and/or transliteration?

In the absence of transformation of contact information, non-ASCII represented scripts cannot be placed in the WHOIS record, which contravenes the Registrar Accreditation Agreement.

Transformed contact information that is verified to be accurate and useable allows users who can understand the information to see who are the domain name registration holders and how to contact/reach them. 

Contact information is used by a range of users for various purposes, which include regulatory/contract enforcement, domain name monitoring and research, domain name purchase/sale, individual Internet use, technical issue resolution, legal action, abuse mitigation and malicious Internet activities (see Expert Working Group on gTLD Directory Services Report). 

In cases where harmful Internet activities are perpetrated, the availability of contact information that is verified to be accurate and useable facilitates those taking action to protect end users.

 

(3) Should translation and/or transliteration of contact information be mandatory for all gTLDs?

The transformation of contact information should be mandatory for gTLDs that allow registration of domains using non-representable scripts in the current ASCII WHOIS for contact information.

 

(4) Should translation and/or transliteration of contact information be mandatory for all registrants or only those based in certain countries and/or using specific non-ASCII scripts?

Transformation of contact information for the purposes of a limited ASCII WHOIS would be applicable for registrants that use non-ASCII scripts in providing their registration information.

In principle registrants should not be expected or required to transform their own contact information.  It is unreasonable to assume that registrants will be able to enter contact information in scripts or languages other than their local script and language.  Nevertheless, it is possible that some registrants may have the ability to do so. 

If transformation of contact information is automated, in order to reduce the risk of data deformation/distortion, the option of allowing registrants to provide ASCII representations voluntarily at the time of entering contact information can be explored.

 

(5) What impact will translation/transliteration of contact information have on the WHOIS validation as set out under the 2013 Registrar Accreditation Agreement?

No Comment 

 

(6) When should any new policy relating to translation and transliteration of contact information come into effect?

No Comment.

 

(7) Who should decide who should bear the burden translating contact information to a single common language or transliterating contact information to a single common script? This question relates to the concern expressed by the Internationalized Registration Data Working Group (IRD-WG) in its report that there are costs associated with providing translation and transliteration of contact information. For example, if a policy development process (PDP) determined that the registrar must translate or transliterate contact information, this policy would place a cost burden on the registrar.

The decision on who should bear the burden of transforming contact information should be informed by the views of and impact on all affected parties, which would include the provider of the information (registrants), those involved in collecting and maintaining the information (i.e., registrar, reseller, registry) and the range of users that include end users and regulatory authorities/law enforcement agencies. 

 

(8) Who does your SG/C believe should bear the cost, bearing in mind, however, the limits in scope set in the Initial Report on this issue?

In determining who should bear the cost, it would be important to consider whether the transformation of contact information is made for general use or specialized use. 

If the transformation is required for general use, the cost should be borne by the process/entities that collect the information.  If the transformation is required for specialized use, the parties requiring the specialized service should bear the cost of contact information transformation. 

An example of general use is contact information that serves a public purpose and is made available to the general public without charging a fee (e.g., WHOIS).  Specialized use is a paid service where the data requester can choose the language of required data.  Groups that may require specialized use of contact information in particular languages may include Law Enforcement Agencies, the Intellectual Property Community, the Network Security Community, etc.   The cost of specialized use can be negotiated between the data provider and the data accessing entity and the level of accuracy required of transformed data would have bearing on the cost.

 

END

FIRST DRAFT SUBMITTED


The At-Large community provides the following responses to the Translation and Transliteration of Contact Information Working Group Charter questions: 

(Note:  The use of the word “transformation” of contact information pertains to the “translation and / or transliteration” of contact information.)

 

(1) Whether it is desirable to translate contact information to a single common language or transliterate contact information to a single common script

It is important to note that translating contact information to a single common language privileges the users of that language over other users, just as the transliteration to a single common script privileges users of that particular script over others.  The choice of one single common language or script does not reflect or support a global Internet community, which is diverse in language and script use. 

Our preference is to have registration contact information in languages and scripts that reflect the diversity of Internet users. 

We believe that it is unreasonable to assume that registrants will be able to enter contact information in scripts or languages other than their local script and language.  It is possible that some registrants may have the ability to do so, but in principle registrants should not be expected or required to transform their own contact information.

In terms of the choice of whether to translate or transliterate contact information, the transformation of contact information to a level where it is understandable to users of registration data may require a combination of both.  For example, when transforming contact information, proper names or nouns (i.e., unique names of persons, places, events and things) should not be translated and should instead be transliterated.

 

 (2) What exactly the benefits to the community are of translating and/or transliterating contact information, especially in light of the costs that may be connected to translation and/or transliteration?

The Internet has a global user base.  Transformed contact information that is validated to be accurate and useable allows users worldwide to see who are the domain name registration holders and how to contact/reach them.

Contact information is used for various purposes by a range of users.  The purposes include regulatory/contract enforcement, domain name monitoring and research, domain name purchase/sale, individual Internet use, technical issue resolution, legal action, abuse mitigation and malicious Internet activities (see Expert Working Group on gTLD Directory Services Report). 

In cases where harmful Internet activities are perpetrated, the availability of transformed contact information that is accurate and useable facilitates those taking action to protect end users. 

Moreover, transformed contact information that is validated for accuracy and usability promotes consumer trust in the Domain Name System and is beneficial to the At-Large community.

 

(3) Should translation and/or transliteration of contact information be mandatory for all gTLDs?

Based on the principle that gTLDs serve global Internet users, transformation of contact information should be mandatory for all gTLDs.

 

(4) Should translation and/or transliteration of contact information be mandatory for all registrants or only those based in certain countries and/or using specific non-ASCII scripts

If transformation of contact information is to be implemented as a matter of policy, then it should be mandatory for all registrants for the purpose of serving global Internet users. 

Registrants based in countries and / or using specific non-ASCII scripts should not be discriminated against in the implementation of contact data transformation.  Contact information for registrants in ASCII script should also be equally transformed for the benefit of non-ASCII script users.

 

(5) What impact will translation/transliteration of contact information have on the WHOIS validation as set out under the 2013 Registrar Accreditation Agreement?

The principle of equivalent validation should be apply (i.e., the validation level of transformed data should be equivalent to the validation level of the original internationalized registration data/contact information).

 

(6) When should any new policy relating to translation and transliteration of contact information come into effect?

New policy on transformed contact information should only come into effect after the policy on internationalized registration data has come into effect.

 

(7) Who should decide who should bear the burden translating contact information to a single common language or transliterating contact information to a single common script? This question relates to the concern expressed by the Internationalized Registration Data Working Group (IRD-WG) in its report that there are costs associated with providing translation and transliteration of contact information. For example, if a policy development process (PDP) determined that the registrar must translate or transliterate contact information, this policy would place a cost burden on the registrar.

The decision on who should bear the burden of transforming contact information should be informed by the views of and impact on all affected parties, which would include the provider of the information (registrants), those involved in collecting and maintaining the information (i.e., registrar, reseller, registry) and the range of users that include end users and regulatory authorities/law enforcement agencies. 

The PDP decision-making rules and procedures for Consensus Policy should apply with input from ICANN Advisory Committees (i.e., At-Large Advisory Committee, Governmental Advisory Committee and Security and Stability Advisory Committee).

 

(8) Who does your SG/C believe should bear the cost, bearing in mind, however, the limits in scope set in the Initial Report on this issue?

In determining who should bear the cost, it would be important to consider whether the transformation of contact information is made for general use or specialized use. 

If the transformation is required for general use, the cost should be borne by the process/entities that collect the information.  If the transformation is required for specialized use, the parties requiring the specialized service should bear the cost of contact information transformation. 

If transformation has general use and is crucial for particular specialized use, then a sharing of costs between those collecting the information and those requiring specialized use of the information could be considered.

 

END

  • No labels

13 Comments

  1. I note that in converting the e-mail at the top of this page to PDF format, all of the hyperlinks (except the e-mail ones) were lost, maing this document not very useful.

    Perhaps we can just cut/past the e-mail into the wiki so the links are available.

     

     

  2. Dear At-Large colleagues,

     

    The first draft of the statement has been developed as a basis for discussion.  Please provide comments.

     

    Best regards,

     

    Rinalia

  3. I have a lot of problems with this statement. I will go through it part by part, but overall, I think that we need a strong statement on overall intent. Something like:

    Regardless of specifics, the ALAC believes that verifiable and verified contact information must be accessible to those with a right and need to access it. Given that the current WhoisS implementation is based on 7-bit ASCII which cannot properly render many languages and scripts, some technique must be used to ensure that registrations that are created using non-ASCII-based scripts have meaningful and usable contact information. Translation and/or transliteration are one way of accomplishing this.

    T/T, however, is not the only way that this end-point can be accomplished.Once we have a Whois database that can handle multiple scripts and character sets, T/T would not be required unless there is a need to have all contact information in a single language or a specific set of languages. But this is a need over and above the one identified above.

    An intermediate step to a fully internationalized Whois might be to have the ASCII-based Whois record point to a non-ASCII based record (maintained by the Registry and in parallel with the ASCII Whois).

    Specific comments:

    1. I am not sure what the first part of this reply is saying. Perhaps it is somthing like I have above, but if so it needs to be clearer. I fully agree with the last paragraph that the answer cannot be translate OR transliterate. Each may be appropriate for specific parts of Whois, as identified in the Issue Report.
    2. I think that the simple part of the answer is missing. In the absence of any T/T, non-ASCII represented scripts cannot be placed in the Whois record, and that contravenes the RAA.
    3. Mandatory for all gTLDs that allow registration of domains using non-representable scripts for contact information.
    4. I do not understand "Contact information for registrants in ASCII script should also be equally transformed for the benefit of non-ASCII script users". Are you suggesting that English (and of their ASCII-based languages) be translated into a range of other languages? So (for instance) Russian or Chinese speakers coulds see the contact information for alangreenberg.org in their own languages and scripts?  
    5. -
    6. I think that this answer side-steps the question. Once we have a replacement Whois which can handle internationalized data, The motivation and need for T/T is (in my opinion) FAR less. To wait for that implies that until we reach that grand new day, names registered by entering non-ASCII data as contact will either not be represented in the Whois database (in contravention of the RAA), or we are leaving it up to the registrar and registry to set their own standards (today's situation).
    7. This seems to be saying that we have no opinion now, but the WG should listen to the WG members and public comments. Is this presuming that in the absence of this statement, they would do something else. In particular, does "The PDP decision-making rules and procedures for Consensus Policy should apply with input from ICANN Advisory Committees (i.e., At-Large Advisory Committee, Governmental Advisory Committee and Security and Stability Advisory Committee)." imply that this is not the case now?
    8. I am confused by what the specialized and non-specialized uses are.
    1. Hi, Alan.

      Excellent input.  Will incorporate text in next version.

      Responses to specific comments:

      Specific comments:

      1. (The first part is a statement on principle that is essential for non-Latin, non English users - no compromises on this one)
      2. (OK with incorporating this input)
      3. (My response to your question 4 below is that I am indeed suggesting that English be translated to a range of other languages to serve global users.  That is only fair.  We are moving towards internationalization and the technical community is working on a solution for this to happen.  Therefore, the response to 3 is that it applies to all gTLDs)
      4. (You understood me perfectly.  The answer is absolutely yes)
      5. -
      6. (OK, I am amenable to not waiting.  When do you suggest we start?  Now/Immediately?)
      7. (No, I am saying the affected parties need to be involved in figuring out the solution and that the decision does not reside solely with the registrars, resellers, etc.  Who do you think should make the decision?  And yes, I think the PDP process can be flawed in terms of balanced representation of interests, which may influence outcomes)
      8. (I have provided the clarification - scroll below to my other responses after this one).

      Rinalia

       

       

  4. Here are a few thoughts from the context of South Asia, a region that is as linguistically diverse as it can get: 1. Given the fact that a majority of Internet users in the region do not use ASCII/Latin/English as their first language/script, I believe that the option of using their own language/script to provide/access contact information would be a non-negotiable. I would therefore strongly support the Recommendation #1 in the Draft. 2. It is not clear if the proposed T/T is automated machine transliteration & translation (or whether it is human-intermediated). Automated T/T would be perhaps be ideal, but I suspect that the output may be less-than-perfect at this time. If we are taking the automated route, we may need to provide controlled mechanisms for the information providers to amend/edit the transliterated information (I'm not sure if such editing will impact search/sort operations that may involve string comparisons--if so, some kind of 'fuzzy' search may be required). 3. In order to reduce the risk of deformation/distortion in automated T/T, it may be interesting to explore if registrants can voluntarily provide ASCII representations at the time of entering contact information.
    1. Thanks, Satish.  I will incorporate your suggestion into the next version of the statement.

       

      Rinalia

  5. Apologies for the loss of formatting in the preceding post...it has been uploaded from a mobile device, as I've been having problems in adding comments from my regular machines.
  6. Thanks, Alan and Satish for the comments.

    Question for Alan, who do you think should make the decision regarding transformation of contact information?

     

    Rinalia

     

  7. Here is the distinction between general and specialized use:

    An example of general use is contact information that serves a public purpose and is made available to the general public without charging a fee (e.g., WHOIS).  Specialized use is a paid service where the data requester can choose the language of the data that it requires.  Groups that may require specialized use of contact information in particular languages may include Law Enforcement Agencies, the Intellectual Property Community, the Network Security Community, etc.   The cost of specialized use can be negotiated between the data provider and the data accessing entity and the level of accuracy required of transformed data would have bearing on the cost.

  8. Dear Alan,

     

    Regarding your proposal on the intermediate step (i.e., "An intermediate step to a fully internationalized Whois might be to have the ASCII-based Whois record point to a non-ASCII based record (maintained by the Registry and in parallel with the ASCII Whois)."

     

    An IDN expert provided the following comments:

    The option requires registry and registrar to implement additional mechanisms (each non-standard and ad hoc in a different way) to eventually move to the final IRD+transformation mechanism.  This is hard to implement, will create undue noise in the way WHOIS is done and will create an intermediate legacy technology and data beyond WHOIS once the final solution is implemented. 

    An alternate is to have just IRD as the interim, and then to add an additional transformation layer (when policy around it finalized).  This is cleaner and does not create and intermediate throw-away effort.  This does mean that in the interim period the IRD will only be available in local languages only, so whoever needs to see this data in another language will have to deal with it themselves. 

     

    Best regards,

     

    Rinalia

    1. I think that we are talking at cross purposes.

      If we had IRD (Internationalized Registration Data) now as an integral part of WHOIS, or we had a firm date when this would be available, the need for T/T would be far less. It is perhaps a nicety, to allow others not familiar with the script/language of registration to read the contact data.

      My suggestion was a method (yes, not an optimal one) to get access to the registration contact information before we have IRD, which today is likely to not be implemented until the directory services group EG concludes and THEN the PDP which is required. ie - years from now.

      Alan

  9. Revised Statement below incorporating input received. 

    Best regards,

    Rinalia

    ---------

    ALAC Statement on Translation and Transliteration of Contact Information

    The ALAC provides the following responses to the Translation and Transliteration of Contact Information Working Group Charter questions: 

    (Note:  The use of the word “transformation” of contact information pertains to the “translation and / or transliteration” of contact information.)

     

    (1) Whether it is desirable to translate contact information to a single common language or transliterate contact information to a single common script

    In principle, translating contact information to a single language privileges the users of that language over other users, just as the transliteration to a single script privileges users of that particular script over others.  The choice of one single language or script does not reflect or support a global Internet community, which is diverse in language and script use. 

    Ideally, registration contact information should be available in languages and scripts that reflect the global diversity of Internet users.  We recognize that the current WHOIS implementation is based on a limited 7-bit ASCII that cannot properly render many languages and scripts.  We are also aware that once the IETF completes its work on the Web Extensible Internet Registration Data Service (WEIRDS), a WHOIS database that can handle multiple scripts and character sets will be available. 

    Irrespective of whether contact information is transformed or not, we strongly believe that verifiable and verified contact information must be accessible to those with a right and need to access it via the WHOIS database.  The availability of contact information that is validated for accuracy and usability promotes consumer trust in the Domain Name System and is beneficial to the At-Large community.

    Until a fully internationalized WHOIS is available, ensuring that registrations created using non-ASCII-based scripts have meaningful and usable contact information in the current ASCII WHOIS can be achieved in various ways:

    One option is to transform the contact information into a single language or specific set of languages using scripts that are representable within the constraints of the current WHOIS.  Another option is to have the ASCII-based WHOIS record point to a non-ASCII based record, which would be maintained by the Registry and in parallel with the ASCII WHOIS.

    When considering the transformation of contact information (i.e., whether to translate or transliterate), transformation to a level where it is understandable to users of registration data may require a combination of both translation and transliteration.  For example, when transforming contact information, proper names or nouns (i.e., unique names of persons, places, events and things) should not be translated and should instead be transliterated.  Translation and / or transliteration may thus be appropriate for specific parts of WHOIS as identified in the Issue Report.

     

    (2) What exactly the benefits to the community are of translating and/or transliterating contact information, especially in light of the costs that may be connected to translation and/or transliteration?

    In the absence of transformation of contact information, non-ASCII represented scripts cannot be placed in the WHOIS record, which contravenes the Registrar Accreditation Agreement.

    Transformed contact information that is verified to be accurate and useable allows users who can understand the information to see who are the domain name registration holders and how to contact/reach them. 

    Contact information is used by a range of users for various purposes, which include regulatory/contract enforcement, domain name monitoring and research, domain name purchase/sale, individual Internet use, technical issue resolution, legal action, abuse mitigation and malicious Internet activities (see Expert Working Group on gTLD Directory Services Report). 

    In cases where harmful Internet activities are perpetrated, the availability of contact information that is verified to be accurate and useable facilitates those taking action to protect end users.

     

    (3) Should translation and/or transliteration of contact information be mandatory for all gTLDs?

    The transformation of contact information should be mandatory for gTLDs that allow registration of domains using non-representable scripts in the current ASCII WHOIS for contact information.

     

    (4) Should translation and/or transliteration of contact information be mandatory for all registrants or only those based in certain countries and/or using specific non-ASCII scripts?

    Transformation of contact information for the purposes of a limited ASCII WHOIS would be applicable for registrants that use non-ASCII scripts in providing their registration information.

    In principle registrants should not be expected or required to transform their own contact information.  It is unreasonable to assume that registrants will be able to enter contact information in scripts or languages other than their local script and language.  Nevertheless, it is possible that some registrants may have the ability to do so. 

    If transformation of contact information is automated, in order to reduce the risk of data deformation/distortion, the option of allowing registrants to provide ASCII representations voluntarily at the time of entering contact information can be explored.

     

    (5) What impact will translation/transliteration of contact information have on the WHOIS validation as set out under the 2013 Registrar Accreditation Agreement?

    No Comment 

     

    (6) When should any new policy relating to translation and transliteration of contact information come into effect?

    No Comment.

     

    (7) Who should decide who should bear the burden translating contact information to a single common language or transliterating contact information to a single common script? This question relates to the concern expressed by the Internationalized Registration Data Working Group (IRD-WG) in its report that there are costs associated with providing translation and transliteration of contact information. For example, if a policy development process (PDP) determined that the registrar must translate or transliterate contact information, this policy would place a cost burden on the registrar.

    The decision on who should bear the burden of transforming contact information should be informed by the views of and impact on all affected parties, which would include the provider of the information (registrants), those involved in collecting and maintaining the information (i.e., registrar, reseller, registry) and the range of users that include end users and regulatory authorities/law enforcement agencies. 

     

    (8) Who does your SG/C believe should bear the cost, bearing in mind, however, the limits in scope set in the Initial Report on this issue?

    In determining who should bear the cost, it would be important to consider whether the transformation of contact information is made for general use or specialized use. 

    If the transformation is required for general use, the cost should be borne by the process/entities that collect the information.  If the transformation is required for specialized use, the parties requiring the specialized service should bear the cost of contact information transformation. 

    An example of general use is contact information that serves a public purpose and is made available to the general public without charging a fee (e.g., WHOIS).  Specialized use is a paid service where the data requester can choose the language of required data.  Groups that may require specialized use of contact information in particular languages may include Law Enforcement Agencies, the Intellectual Property Community, the Network Security Community, etc.   The cost of specialized use can be negotiated between the data provider and the data accessing entity and the level of accuracy required of transformed data would have bearing on the cost.

     

    END