Transcript

Transcript CWG DT-F 15 April.doc

Transcript CWG DT-F 15 April.pdf

Recording

The Adobe Connect recording is available here:   https://icann.adobeconnect.com/p1v1r3w0nhg/

The audio recording is available here:  http://audio.icann.org/gnso/gnso-dtf-15apr15-en.mp3

Notes

 DT F call 15 April 2015 

Comments received from Milton

Timing CWG: Recommendations  DT F need to be read by tomorrow 

Structure:

Recommendations that MUST be included 

Identify topics  to be addressed later. 

Two versions: short form of recomemndation and whatever size is necessary  

Suggestion use Milton's suggestions for short document 

Identifying NTIA "routine " Authorization goes away 

1 a. Section

Why say in short term?

Currently prohibted to make changes. 

Alternative: eiither change before software, or handled on interim basis, without any need to make change 

Chuck suggestion, make explicit. what is meant

Changes to be made 

1 b.

.Chuck suggestion, third sentences.

Add, nor whether what will be in place, 

Suggested change: 

The exact form of the latter transition is not currently known, nor what whether anything will replace the current 
Cooperative Agreement. However, there will likely may be a requirement to have a formal agreement between the IANA 
Function Operator and The Root Zone Maintainer.  

Does the group want to specify additional checks and balances or leave it for later stage.?

David: include as a recommendation caution not to go one entity

Chuck: suggest # 6 as suggestion 

Recommendation is: additional analysis is needed  

Recommendation 2

NTIA more extensive then anticipated first

NTIA is getting into fair amount 

Question: Is DT F going to make a recommendation or someone else 

David: Sensitivities involved, appropriate to make recommendation that CWG identify a party that explores issues. 

NTIA  approval needed fro different categories

Authorization of changes 

On day-to day basis

Publish reporting

Approval monthly reports

Document approval 

Question Alan: how often approval and is it real function? 

Question: what needs prior community approval, what can be done by IFO, with out prior approval by community 

Topic needs to be considered, what needs to be reviewed/ approved 

Unclear what rationale was for NTIA.  

Question What is  current rationale, to provide guidance.? 

Are there real issues or just concerns of USG.? 

Focus on structural aspects and not on presentation aspects.  

Question Alan: Are there topics that need to be discussed outside IAAN, but not fit for public comment? 

Issues specific to specific situation or cases could be an issue, but not in general items 

Capture notion that check for everything creates issues by itself, at the same some items may need to be  

Default should be on openness  and transparency as main principle, in some operational circumstances,confidentiality may be required -> need to be documented 

Interaction is vast majority of cases be open and transparent.  

Note that ATRT 1 and ATRT 2, transparency is needed and need to be in DNA 

Chuck: timing could make it an issue. >making it public when changes already made, non-issue 

Section 3. 

Question: is this in DT O? 

DT O recommended that as the solution get more well-defined to check whether budget has covered them

DT O has included general recommendation . However suggestion to re-amphasize point.  

Cheryl: In addition to DTO recommendation  DT F specially notes inclusion of   

Issues Warranting further discussion

Simply listing allows others to pursue these items ( CSC or part of Periodic review 

If considered critical infrastructure then items should be looked at in a consistent manner. Stronger suggestion that study must be done

to be considered by chair 

Section 5. What should be public and not be public

Recommendation  needs to be added on in line with discussion

David: Suggest to take into account phasing of change request process. Point for publication is when validation has been completed. 

What really matters is that changes has been accepted for processing

Agree it should be studied. 

In specific cases of delegation and redelgation of ccTLD that third parties. 

Introduce new section Principles that should be adhered to 

Include transparency as well ( Section 5) 

Section 6 

Making a statement of concentration of power is appropriate 

suggestion Chuck

Currently updating the Root Zone requires the active participation of three parties, the IANA Function Operator, the Root Zone Maintainer 
and the NTIA. Post transition there will only be the first two. DT-F recommends that the remaining two functions should 
never be awarded to a single entity. Note that the implications of this might vary depending on whether any or all of the 
robustness issues identified in Issue 1 have been addressed."

What is purpose of separation of IF and RZM, is to ensure two person control. Reduce risk of failures. malicious behaviour etc.

Risk mitigation tool

This looks at changes proposed to change implemented. 

Two parties control does not fix all problems, but catches some 

Suggestion: there should be no concentration of power but we need to explore how best to  implement it 

Section 7

Concern: if community wants to make slower

What is purpose of separation of IF and RZM, is to ensure two person control. Reduce risk of failures. malicious behaviour etc.

Risk mitigation tool

This looks at changes proposed to change implemented. 

Two parties control does not fix all problems, but catches some

Action Items

none noted

Chat Transcript

  Brenda Brewer: (4/15/2015 12:36) Hello and welcome to the DT-F Meeting #2 on 15 April.

  Cheryl Langdon-Orr: (12:52) Hi Brenda...   they can call me back any time from niw...   I did not get to pick up in time  easrlier

  Cheryl Langdon-Orr: (12:55) Hi there

  Bernard Turcotte - staff support: (12:56) helklo

  Bart Boswinkel: (12:56) Hi all

  Brenda Brewer: (12:57) and B

  Brenda Brewer: (12:57) :)

  jaap akkerhuis (SSAC): (13:00) Evening all

  David Conrad: (13:00) can anyone hear me?

  Cheryl Langdon-Orr: (13:01) no David  did not hear you

  Brenda Brewer: (13:01) I do not hear you David

  Cheryl Langdon-Orr: (13:01) Yup

  jaap akkerhuis (SSAC): (13:02) We hear you

  jaap akkerhuis (SSAC): (13:02) Someone is in an Airport?

  Chuck Gomes: (13:03) Just joined

  Cheryl Langdon-Orr: (13:03) Hi Chuck

  David Conrad: (13:04) yes 6508238783

  Bernard Turcotte - staff support: (13:04) please remember to mute if you are not speaking

  David Conrad: (13:05) can't hear anything

  Bernard Turcotte - staff support: (13:05) david you ok now

  David Conrad: (13:06) can hear but can't talk

  Brenda Brewer: (13:06) David, please see private chat

  Bernard Turcotte - staff support: (13:06) best of both worlds then

  David Conrad: (13:07) Brenda: please call +16508238783

  Brenda Brewer: (13:07) thank you David,  stand by.

  David Conrad: (13:07) can't type into private chat

  Kim Davies: (13:09) Hello, sorry for being late

  Brenda Brewer: (13:09) David, getting VM when we call you.

  Milton: (13:16) That is not going to happen

  Milton: (13:17) I can't stay, did you have any reactions to my proposed changes?

  Chuck Gomes: (13:17) The exact form of the latter transition is not currently known, nor what whether anything will replace the current Cooperative Agreement. However, there will likelymay be a requirement to have a formal agreement between the IANA Function Operator and The Root Zone Maintainer.

  Alan Greenberg: (13:18) Milton I will answer

  Cheryl Langdon-Orr: (13:18) We are going over them  line by line  along with other cotributions  Milton...

  Milton: (13:19) OK, thanks, sorry it overlaps w ARIN meeting so just wanted to see if I was lucky enough to be on when you were discussing them

  Cheryl Langdon-Orr: (13:19) Your comment on length  has already been noted

  Cheryl Langdon-Orr: (13:20) we are now at 1b/c

  Cheryl Langdon-Orr: (13:20) AGree with David

  Milton: (13:21) Did you retain the line about IANA "authorizing"  - seems odd when there will be no authorization function

  Kim Davies: (13:22) There was discussion a few minutes ago that that text be clarified that it is an interim approach to be able to conduct root zone changes without NTIA until necessary software updates are completed (if they are not in place on day 1)

  Milton: (13:22) That's for NTIA to sort out, not us

  Kim Davies: (13:23) (No)

  Milton: (13:25) ok so you mean literally 1 or 2 days

  Chuck Gomes: (13:26) I am going to have to drop off.

  Cheryl Langdon-Orr: (13:27) OK  Chuck

  Bernard Turcotte - staff support: (13:28) Sorry had to switch lines

  Chuck Gomes: (13:28) Here is my suggested text for 6: "6.        Currently updating the Root Zone requires the active participation of three parties, the IANA Function Operator, the Root Zone Maintainer and the NTIA. Post transition there will only be the first two. DT-F recommends that the remaining two functions should never It is conceivable that the IANA function and the Root Zone Maintainer agreements could be awarded to a single entity. Should the CWG explicitly recommend that the power/responsibility to modify/update the root zone not be concentrated in a single entity? Note that the implications of such an awardthis would might vary "

  Milton: (13:29) Chuck, there are incoherent fragments in there

  Kim Davies: (13:29) I'm still here

  Milton: (13:30) Some kind of copy-paste error in Chuck's suggestion. "the remaining 2 functions should never it is conceivable..."

  Chuck Gomes: (13:30) The chat doesn't show my redlined deletes so here is a clean version of what I recommend for 6: "6.         Currently updating the Root Zone requires the active participation of three parties, the IANA Function Operator, the Root Zone Maintainer and the NTIA. Post transition there will only be the first two. DT-F recommends that the remaining two functions should never be awarded to a single entity. Note that the implications of this might vary depending on whether any or all of the robustness issues identified in Issue 1 have been addressed."

  Cheryl Langdon-Orr: (13:31) You need to let us know  about the current Frequency of  Go/No Go decisions  being made... following  Davids intervention  I suspect @kim

  Cheryl Langdon-Orr: (13:31) ahh  that reads better  thanks @Chuck

  Milton: (13:32) In response to Chuck's suggestion, and reiterating the question I asked iin the email, can we identify what EXACTLY is gained by never combining these functions in a single entity?

  Milton: (13:33) if we are going to make this a recommendation (rather than a question that the CWG should answer) we need to be able to answer that question

  Milton: (13:37) Alan, why does that matter?

  Bernard Turcotte - staff support: (13:37) they have to go to NTIA to answer

  Milton: (13:37) ha ha

  Bernard Turcotte - staff support: (13:37) ;-)

  Milton: (13:37) :-)

  Milton: (13:38) Kim: Do you know what criteria the NTIA used to decide what you could publish or not?

  Milton: (13:38) Was there concern security, something else?

  Milton: (13:41) We would want to know what their rationale was - it might provide us guidance

  Milton: (13:42) So @David a lot of the need for review was caused by the USG's intrusion in the process in the first place

  Milton: (13:43) Right, Alan, but as noted David implies that most of the NTIA review was just the USG trying to cover its a**

  Chuck Gomes: (13:45) I am back for 15 minutes.

  Chuck Gomes: (13:49) @ Milton: Advantage of having to separate entities is checks & balances; two different entities doing checks.

  Chuck Gomes: (13:49) to = two separate entities

  Milton: (13:50) Chuck: so the process CANNOT be fully automated

  Milton: (13:51) ?? <-- that was meant to be a question

  Milton: (13:51) I think capture might be a better rationale

  Cheryl Langdon-Orr: (13:52) Agree absolutly  with this  default  to openness  tennant  it was also an issue raised in ATRT1

  Cheryl Langdon-Orr: (13:53) Agree totally here @David

  Cheryl Langdon-Orr: (13:53) yup

  Milton: (13:54) right - timing is important

  Milton: (13:55) is the transparency serving an error-checking function?

  Bernard Turcotte - staff support: (13:55) Dels and redels are special cases but there are a lot more than that

  Chuck Gomes: (13:58) :)

  Chuck Gomes: (13:59) Because I am co-chair and may have to chair the P&I WG call, I now have to depart for good.  I will check the document that Alan produces.

  Milton: (14:00) bye Chuck

  Cheryl Langdon-Orr: (14:03) yes

  Cheryl Langdon-Orr: (14:03) follow Davids rec

  Bernard Turcotte - staff support: (14:04) +1

  David Conrad: (14:09) ow

  David Conrad: (14:09) (i guess no one else was getting modem tones...)

  David Conrad: (14:10) really unimpressed with Adobe Connect

  Cheryl Langdon-Orr: (14:10) It has its moments @David   but so do all the other tools I've used...

  Milton: (14:14) We actually have two different interpretations of the separation of RZM and IANA principle: one talks about "concentration of power" the other about "checking errors"

  Cheryl Langdon-Orr: (14:14) speed AND Accuracy  need balance

  Milton: (14:15) what is the purpose of the separation?

  Milton: (14:16) No, I am asking

  Cheryl Langdon-Orr: (14:17) Yes  i is to better manage your  Critical Control Points  to reduce RISK

  Milton: (14:17) So you are saying that RZM can refuse to do what IANA proposes

  Milton: (14:17) But David, are we talking about avoiding mistakes, or abuses or power, or both?

  Cheryl Langdon-Orr: (14:20) it recures  concurance that the change is accuratly recorded and that due process ( however defined)  has properly occurred

  Milton: (14:20) OK, thanks

  Cheryl Langdon-Orr: (14:20) yes  Both Milkton...

  Cheryl Langdon-Orr: (14:20) sorry typos  in two calls at once  :-(

  Cheryl Langdon-Orr: (14:21) @Alan 2 party  is a risk MITIGATION  tool

  Milton: (14:22) can you hear?

  Milton: (14:22) I will use the cat instead

  Milton: (14:22) chat not cat

  Milton: (14:23) So my comment is this: given the purpose of the separation, does it make sense for ICANN to contract with RZM or should the contract come from someone else?

  David Conrad: (14:23) don't think it matters.  the key bit would be what was in the contract

  Milton: (14:24) a contract makes the RZM the agent of the IANA, not a check/balance on its power

  Milton: (14:24) agree about exploration

  Milton: (14:25) a lot of thought needs to be put into that

  Milton: (14:25) agree with that last formulatio David said

  Cheryl Langdon-Orr: (14:25) Yup  that works

  Bernard Turcotte - staff support: (14:26) we will not get it before its due

  Milton: (14:26) there should be no concentration of power but we need to explore how best to  implem,ent it

  Bernard Turcotte - staff support: (14:26) bye

  jaap akkerhuis (SSAC): (14:26) Bye bye

  Cheryl Langdon-Orr: (14:26) bye