Milton, I respectfully disagree. First, these are not "technical issues of minimal relevance." These are fundamental issues that go to clarifying the meaning and focus of this provision, which hinges on clarifying the meaning of "services" and "use". We can't define a technical process such as a "web service" without some technical references. But that doesn't make this a "technical issue." You are failing to focus on this on the technical layer; instead you keep trying to look at it on an economic layer. Given your background that's not surprising, but it doesn't take us in the right direction. As a result of your focusing on the wrong layer, you keep seeing "service providers" when you read "services," rather than seeing "web services" and similar technical processes occurring at the software-server level. You need to zoom in. I am confident this view is consistent with the comments made by Andrew Sullivan and Malcolm Hutty, among others, and we need to keep going in this direction to finalize this provision On Tue, Nov 17, 2015 at 11:13 AM, Mueller, Milton L <milton@gatech.edu> wrote:
Greg,
With all due respect, you are “hung up” on technical issues that are of minimal relevance.
Answers to some of your questions below should make this clear:
We have been referring to "services that use the Internet's unique identifiers." What kind of "services" actually "use" the Internet's unique identifiers?
MM: All of them. You can’t run a service on the internet of any sort without using IP addresses. Nearly all also use domain names.
GS: Again, you are looking at "service businesses," one among many types of Internet users. There's no reason this would be limited to service businesses and not include other businesses or other users of the Internet; it's just not a rational distinction. So the idea that we are talking about businesses that "run a service on the Internet" just doesn't hold water. Further, the business itself doesn't "use" an IP address; the business as such is barely aware that there is an IP address. The IP address is used by the servers and other things attached to the Internet in order to be located on the Internet. Those "things" could be owned by a business (whether product or service based), an individual, a government, etc.
Service *businesses* don't really "use" the Internet's unique identifiers.
MM: Yes, they do. I would guess that your problem here lies in the definition of “use” not “service.”
GS: It's both.
Sure, they may own domain names and they may type in domain names when they use email clients or web browsers -- but so do product-based businesses, end users, IGOs, sovereigns, etc., so a reference to services in particular and not to every other kind of user makes no sense.
MM: If I type in the domain for cnn.com I am “using” CNN’s service, and the distinction between a user and a provider is pretty clear. Because of your limited understanding of the tech I think you’ve completely lost sight of the purpose of this discussion. To repeat, it is to proscribe ICANN from regulating things that are outside its remit (content and services other than registries and registrars).
GS: Milton, Are you now a service? If we are talking about your example, when you type in cnn.com, your browser sends commands to the software on the server located using that domain name; that software than performs a "service" by assembling a result and sending it back to you, using the IP address associated with your computer (not with you), so that you see a "web page" on cnn.com. So, Milton Mueller may be looking at cnn.com, but the *service* in this equation is the software process, and the *use *of the Internet's unique identifiers is the addressing that allows the browser on your computer to reach cnn.com and vice versa. I've never claimed to have "unlimited understanding" of the underlying technology, but I think I have a more than adequate grasp of it all. This part of the provision is important to its understanding; it can't be "read away," as the NRA would like to do with the militia clause of the Second Amendment. It's all part of the overall meaning of the provision. I'm not going to impugn your "understanding of the tech," but I think that if you think about what actually takes place as Internet provides "information services" we'll be getting somewhere.
I refrain from responding to the rest of your message because it is more manifestations of confusion rather than anything that gets us anywhere.
I contend that it is perfectly clear, and it is really the crux of my communication. Your attempt to evade it by dismissing it as "manifestations of confusion" gives me additional evidence that it is logical and on the right track. Greg
--MM
Yet "services that use the Internet's unique identifiers" was chosen on purpose rather than some completely different expression. It's become increasingly clear to me that what this refers to is a service such as a web service, mail service, etc., that runs on a web server, mail server, etc., and the "use" of the Internet's unique identifiers that's referred to is the actual technical process that occurs when that software is "found" or communicates using the DNS. Other language bears this out -- a service *business* doesn't really "carry" content, but a web service or a mail service clearly does (think of "carry" as a variant of "carriage" or "carrier"). This understanding of "use" is also demonstrated by the Malcolm Hutty proposal on November 6, which referred to use of the Internet's unique identifiers by services "to enable or facilitate their reachability over the Internet." It's the software-service that needs to be "reached" using the DNS, not a person or a business that is being "reached."
So, I actually contend that we are on the right track here, and not "hung up." As my re-posted email indicates, Andrew Sullivan had a couple of constructions for the parenthetical that advanced discussions beyond my own attempt (and which are in my re-posted email), and which are more "technology neutral." Of course, there's a certain challenge to describing a technological process in a way that is "technology neutral" -- but since ICANN is an entity that carries out a technical mission, it's a challenge that has been met many times before and I'm confident it will be met again here. We're already closer than we were just a few days ago.
Greg
On Mon, Nov 16, 2015 at 6:12 PM, Mueller, Milton L <milton@gatech.edu> wrote:
Becky and Greg:
I am sorry I was not able to be on this call but it occurred in the middle of the IGF.
The problem with the conversation in the transcript below is that it is still hung up on some kind of technical definition of ‘service,” when all you need to do is come up with a way of differentiating registry and registrar service (which ICANN can legitimately regulate) from ALL other services (which it should not regulate. So again, I ask, what is wrong with this?
ICANN shall act strictly in accordance with, and only as reasonably appropriate to, achieve its Mission. Without limiting the foregoing:
- ICANN shall not impose regulations on:
- Information services which use the Internet’s unique identifiers but are not registries or registrars, or - The content that such services carry or provide
- ICANN shall have the ability to enter into and enforce agreements with contracted parties, insofar as those agreements are consistent with its Mission.
*From:* accountability-cross-community-bounces@icann.org [mailto: accountability-cross-community-bounces@icann.org] *On Behalf Of *Greg Shatan *Sent:* Monday, November 16, 2015 5:47 PM *To:* Andrew Sullivan *Cc:* accountability-cross-community@icann.org *Subject:* Re: [CCWG-ACCT] SIDE BY SIDE on Regulation/Contract language
*I sent an email Saturday in an attempt to move the ball on this provision -- particularly on the provision that Andrew highlights. Since it may have gotten lost in the email blizzard, I am pasting it in here as well. *
*Greg*
This topic was just discussed extensively on the call yesterday and there was considerable progress. The recordings, notes, chat and call transcripts are all available now at https://community.icann.org/pages/viewpage.action?pageId=56145283. The relevant discussion is on pp. 4-19 of the call transcript. It's important to read the whole section, but the concluding remarks from that discussion are helpful for us to build on:
Becky Burr: Well, no, I mean, I think that the point is - and Greg has just echoed me in this - what we are talking about is services, not service providers. So the question is what is the underlying service, not what is the nature of the business. So I guess I agree with - I agree with the way it is set up in this language here.
Thomas Rickert: Yes, and I guess that what you’ve just refined is a better description of what Greg I think meant by saying differing classes of services - class of businesses. So I guess that’s helpful. So I think we can confirm and please let me know if you do not agree with this, that that we are looking for a technology-neutral description of this.
But still if we’re using the technique that Kavouss has suggested, for example, we could have examples of technology to illustrate what we mean by the definition that should then go - or by the language that should then go into the bylaws. Alan, you’ve raised your hand.
Alan Greenberg: Yeah, thank you. It [dawns] on me that when we're looking at the two classes of service, someone may be offering a graphics design service over the web. Clearly, we are never saying we are going to be regulating the graphics design service. So at some level we could not - we don’t have to differentiate because we’re - the higher level service is always carved out. But since the word has two different very distinct meanings it’s probably better to be clear. Thank you.
Thomas Rickert: Thanks very much, Alan. And I think that these are excellent final words on this conversation. So I think this is as far as we can get during this call. And with that I’d like to thank you all for your contributions and for bearing with us for those who are not calling this their favorite item. And let’s now move to the next agenda item which is going to be chaired by Mathieu.
Andrew Sullivan had two suggested revisions to the first part of the provision
, both of which are consistent with this direction:
· **
*ICANN*
* shall not impose regulations on *
**
*services (i.e., any software process that accepts*
* *
*connections from the Internet) that use the Internet’s unique*
* *
*identifiers, or*
* the content that such services carry or provide*
or, alternatively
· **
*ICANN*
* shall not impose regulations on *
**
**
*services (i.e., any software process that accepts datagrams from the Internet, when those datagrams are not themselves necessarily the consequence of a datagram previously sent by the software process itself) that use the Internet’s unique identifiers, or*
* the content that such services carry or provide*
These parentheticals are both more "technically neutral" than the parenthetical I circulated ("(i.e., the software processes by which commands received via the Internet are processed and a response is generated and transmitted via the Internet, to be viewed in a web browser, email client, or the like")). I tend to prefer the first one, which has the virtue of brevity. That said, we should see if there are tweaks consistent with Thomas's summary of the call "we’re not talking about the service providers or the class of business that they're in but that we're talking about the technical processes, the technical services."
Another suggestion came from Milton Mueller:
· *ICANN*
* shall not impose regulations on *
**
*Information services which use the Internet’s unique identifiers but are not registries or registrars, or*
* *
*t*
*he content that such services carry or provide*
In response to James Bladel, Milton suggests "Instead of saying “registries or registrars” we could simply say “under the Registrar Accreditation Agreement (RAA) or the Registry Agreement (RA).”
·
*ICANN*
* shall not impose regulations on *
**
*Information services which use the Internet’s unique identifiers but are not under the Registrar Accreditation Agreement (RAA) or the Registry Agreement (RA), or*
* *
*t*
*he content that such services carry or provide*
Although "information services" could be preferable to "services," the first suggestion clearly takes us away from describing "technical processes" and instead describes "service providers" since it refers to "registries or registrars" as the kind of thing that would be included as "information services" (unless carved out). The second suggestion is not as clearly about "service providers" but it's also not clearly about "technical services" either (there's not enough information to read it clearly).
Perhaps there's a way to combine these various thoughts in a manner consistent with the overall direction, for instance in the following synthesis:
·
*ICANN*
* shall not impose regulations on:*
o **
*Information services (i.e., any software process that accepts*
* *
*connections from the Internet) that use the Internet’s unique identifiers, other than those covered by the Registrar Accreditation Agreement (RAA) or the Registry Agreement (RA), or*
* *
o **
*t*
*he content that such information services carry or provide*
or, alternatively (if the carve-out for the RAA/RA seems confusing, unnecessary or overbroad)
·
*ICANN*
* shall not impose regulations on:*
o **
*Information services (i.e., any software process that accepts*
* *
*connections from the Internet) that use the Internet’s unique identifiers, or*
* *
o **
*t*
*he content that such information services carry or provide*
In order to focus this email, I haven't touched on the second sentence, which we should finalize as well.
I look forward to your thoughts.
Greg
On Mon, Nov 16, 2015 at 5:27 PM, Andrew Sullivan <ajs@anvilwalrusden.com> wrote:
On Mon, Nov 16, 2015 at 10:15:31PM +0000, Burr, Becky wrote:
I am recirculating the slide that compares the 2nd Draft Report language with alternative language discussed on our last call. Although some who participated in the discussion found the definition of services to be a bit clunky, folks generally felt that this language could work as direction to drafters. We need to reach closure on this issue.
With my usual disclaimer in place, I really strongly advise against the "to be viewed" &c. language. I sent an alternative and I think there were some other suggestions as well. They're all much more neutral with respect to technology.
A -- Andrew Sullivan ajs@anvilwalrusden.com _______________________________________________ Accountability-Cross-Community mailing list Accountability-Cross-Community@icann.org https://mm.icann.org/mailman/listinfo/accountability-cross-community
_______________________________________________ Accountability-Cross-Community mailing list Accountability-Cross-Community@icann.org https://mm.icann.org/mailman/listinfo/accountability-cross-community