Re: the iab net neutrality
On Mar 29, 2006, at 10:56 AM, Henning Schulzrinne wrote: We could ask the IEEE, since the relationship between the WiFi folks and IEEE 802.11 seems to be somewhat similar. One of the problems I see is that many of the industry associations (SIP Forum, IPv6 forum, to name two I'm somewhat familiar with) tend to focus on service providers, not consumers. But an organization such as the SIP Forum could provide a VoIP-optimized label for NAT boxes and maybe even ISPs. I'm a board member of the SIP Forum, so I'd like to respond to Henning. (I'm speaking as an individual here who happens to be on the SIP Forum board so these are personal views neither discussed with nor agreed to by the rest of the board. Ditto for the IAB.) The SIP Forum is a creature of our members, which today are almost exclusively service providers and equipment vendors. We try to respond to the pain points they bring us and add value by bridging the gap between protocol standardization through the IETF and needs in the market. So far, we've been pretty successful at running interoperability testing through the SIPIT program, and coming up with deployment and feature bundling specifications in areas like hooking up SIP-based enterprise VoIP systems to service providers who are offering PSTN origination and termination services. The question of how to help the consumer market segment is one we are stumped on, for a number of reasons. First, there is no obvious advocate for the needs of consumers among our membership. Second, few to none of the vendors who sell consumer gear (e.g. Linksys, Netgear, Sony, Apple) are members. Third, much of that market segment is driven by offshore manufacturers who have little incentive to lead in engineering. Their expertise is in channel and brand management, and in minimizing all costs, including engineering (not to mention forum memberships...). That said, a number of us believe that we are having a modest effect. For example, in the enterprise interconnect specification, we worked very hard to make sure straightforward interconnect worked without mandating extra firewall, NAT or B2BUA functionality. The idea of having the SIP Forum sponsor work to at least partially drain the NAT/firewall traversal swamp is a good one. So - seeing as this is on the IETF public list, let me offer a plea: if you work with or build SIP products for consumers, JOIN THE SIP FORUM and help us put together a program in our Technical Working Group to address these issues. We are driven by our members. Membership is free for individuals and of modest cost for companies. Cheers, Dave Oran Thus, I think we need a separate organization (or work with a separate organization) that does branding and certification. It's hard to buy a non-WiFi device in stores today; the equivalent consumer assurance needs to be true for core consumer and small- business network devices, and possibly services. I don't know how this would work, but if it could be made to work, that might be very helpful. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
We could ask the IEEE, since the relationship between the WiFi folks and IEEE 802.11 seems to be somewhat similar. One of the problems I see is that many of the industry associations (SIP Forum, IPv6 forum, to name two I'm somewhat familiar with) tend to focus on service providers, not consumers. But an organization such as the SIP Forum could provide a VoIP-optimized label for NAT boxes and maybe even ISPs. Thus, I think we need a separate organization (or work with a separate organization) that does branding and certification. It's hard to buy a non-WiFi device in stores today; the equivalent consumer assurance needs to be true for core consumer and small-business network devices, and possibly services. I don't know how this would work, but if it could be made to work, that might be very helpful. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
Dear All, My apologies for not being clearer - my intention was not to criticize WG or IAB actions in the past, but to point out that we are now in an escalating game of whack-a-mole with our applications as the moles that NATs and FWs are finding new ways to frustrate. The security guys have taught us that holding the mallet and waiting for the next mole to pop up is not fun; I think the NAT/FW guys are teaching us that being the mole is just as bad (gee, this used to work, until someone decided that yet another normal operation was a security threat and configured their FW to block it, or someone just installed a NAT that is broken in some new and exciting way). Thanks, Spencer p.s. I apologize for the use of culture-specific analogies. For an explanation of Whack-a-mole, see http://en.wikipedia.org/wiki/Whack_a_mole or http://whacamole.com/ (the official website - now, that's scary). From Wikipedia: Colloquial usage: The term Whac-a-Mole, or Whack-a-mole, has been used in the computer and networking industry to describe the phenomenon of fending off recurring spammers, vandals or miscreants. The connotation is that of a repetitious and futile task: each time the attacker is whacked or kicked off of a service, he only pops up again from another direction. Also used in the military to refer to opposing troops who keep re-appearing: Whack the mole here and it dies, but another pops up in a different spot. From: Melinda Shore [EMAIL PROTECTED] On 3/25/06 7:47 PM, Spencer Dawkins [EMAIL PROTECTED] wrote: So my point was, I'd really like to take a chance on some IAB statements about things that need to be stated about our architecture. They might be ignored. Would the result be any worse? This is a somewhat bothersome case, because the IAB *did* issue an RFC explaining what many of the problems were with Unilateral Network Self-Address Fixing (i.e. STUN). They included a list of conditions they felt that an UNSAF protocol had to meet in order to be published, including a description of a transition mechanism away from itself and towards something more robust. I don't know what more the IAB could have done in order to kill what I think is a clearly pathological approach to NAT traversal (and I chaired the working group that standardized it, so I accept a great deal of responsibility for this mess), but if putting out a document that says These are the reasons that this isn't a good protocol isn't enough, well, I'm not sure. But it seems to me that trying to fix it this late in the process (my other .sig is software longa, hardware brevis) has less to do with architecture and more to do with oncology. At any rate, I do think that in this case the IAB did do their job and it was the rest of us louts who messed up. And I'll tell you where I think it happened: when we accepted the idea that something might be transitional and would eventually go away. Melinda ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
On 3/27/06 6:45 AM, Spencer Dawkins [EMAIL PROTECTED] wrote: My apologies for not being clearer - my intention was not to criticize WG or IAB actions in the past, but to point out that we are now in an escalating game of whack-a-mole with our applications as the moles that NATs and FWs are finding new ways to frustrate. I think we're actually making similar points - if you find yourself playing whack-a-mole (and we are) there's a good chance that you're taking the wrong approach. In this particular case there's a reliance on using side-effects for NAT traversal, which suggests that we need sufficiently similar side-effects from all NATs for the approach to work predictably. Rather than concluding that maybe betting on uniformity in side-effects inside closed boxes isn't a great design decision there's been a tendency to respond to all these problems that have cropped up as if applying heaps of baling wire and duct tape will eventually get everything working properly. Melinda ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
Traditionally, it was sufficient for the IETF to publish an RFC specifying requirements or behavior; the purchasing process, through RFIs and RFPs, then cited the long list of RFCs, essentially creating the protocol police force that the IETF doesn't have. That list-of-RFC-numbers approach is clearly not workable for consumer gear. The consumer wireless providers recognized that a while ago. Thus, 802.11b became WiFi and an easily recognizable logo, with interoperability testing. Same for BlueTooth. Trying to devise ever more elaborate NAT traversal mechanisms that include sending keep-alives every few seconds and various let's try this and then that algorithms clearly don't scale if we want to get to consumer-grade reliability, not just interesting toys that try to stay one step ahead of the next stupid NAT trick. Thus, I think we need a separate organization (or work with a separate organization) that does branding and certification. It's hard to buy a non-WiFi device in stores today; the equivalent consumer assurance needs to be true for core consumer and small-business network devices, and possibly services. Henning Spencer Dawkins wrote: Dear All, My apologies for not being clearer - my intention was not to criticize WG or IAB actions in the past, but to point out that we are now in an escalating game of whack-a-mole with our applications as the moles that NATs and FWs are finding new ways to frustrate. The security guys have taught us that holding the mallet and waiting for the next mole to pop up is not fun; I think the NAT/FW guys are teaching us that being the mole is just as bad (gee, this used to work, until someone decided that yet another normal operation was a security threat and configured their FW to block it, or someone just installed a NAT that is broken in some new and exciting way). Thanks, Spencer p.s. I apologize for the use of culture-specific analogies. For an explanation of Whack-a-mole, see http://en.wikipedia.org/wiki/Whack_a_mole or http://whacamole.com/ (the official website - now, that's scary). From Wikipedia: Colloquial usage: The term Whac-a-Mole, or Whack-a-mole, has been used in the computer and networking industry to describe the phenomenon of fending off recurring spammers, vandals or miscreants. The connotation is that of a repetitious and futile task: each time the attacker is whacked or kicked off of a service, he only pops up again from another direction. Also used in the military to refer to opposing troops who keep re-appearing: Whack the mole here and it dies, but another pops up in a different spot. From: Melinda Shore [EMAIL PROTECTED] On 3/25/06 7:47 PM, Spencer Dawkins [EMAIL PROTECTED] wrote: So my point was, I'd really like to take a chance on some IAB statements about things that need to be stated about our architecture. They might be ignored. Would the result be any worse? This is a somewhat bothersome case, because the IAB *did* issue an RFC explaining what many of the problems were with Unilateral Network Self-Address Fixing (i.e. STUN). They included a list of conditions they felt that an UNSAF protocol had to meet in order to be published, including a description of a transition mechanism away from itself and towards something more robust. I don't know what more the IAB could have done in order to kill what I think is a clearly pathological approach to NAT traversal (and I chaired the working group that standardized it, so I accept a great deal of responsibility for this mess), but if putting out a document that says These are the reasons that this isn't a good protocol isn't enough, well, I'm not sure. But it seems to me that trying to fix it this late in the process (my other .sig is software longa, hardware brevis) has less to do with architecture and more to do with oncology. At any rate, I do think that in this case the IAB did do their job and it was the rest of us louts who messed up. And I'll tell you where I think it happened: when we accepted the idea that something might be transitional and would eventually go away. Melinda ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
I think Henning and I are saying the same thing (he's just saying it better). From: Henning Schulzrinne [EMAIL PROTECTED] Trying to devise ever more elaborate NAT traversal mechanisms that include sending keep-alives every few seconds and various let's try this and then that algorithms clearly don't scale if we want to get to consumer-grade reliability, not just interesting toys that try to stay one step ahead of the next stupid NAT trick. I know that I agree with this. Thus, I think we need a separate organization (or work with a separate organization) that does branding and certification. It's hard to buy a non-WiFi device in stores today; the equivalent consumer assurance needs to be true for core consumer and small-business network devices, and possibly services. I don't know how this would work, but if it could be made to work, that might be very helpful. Thanks, Spencer ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
John, everyone, I think it's fair to say that the IAB has heard the concern at this point -- about the net neutrality issue, and the desire to see some concrete IAB action. I've also seen a fair bit of discussion about what an appropriate stance *is*, and whether or how to express it as a useful and usable (IAB) action. I'm not sure that timid is the right word -- in at least some instances, considered would be better. Which means there could (should?) be some time delay between heard the concern and that concrete IAB action. Leslie. John C Klensin wrote: Tony, I agree completely and believe the IAB has, of late, been altogether too timid in this area. I think you know all of what I'm about to say, but your note is, IMO, easily misread, so an additional observation about 4084 and its potential relatives: In this sphere, a document that says XYZ is evil is essential worthless. The companies considering XYZ will almost always say hmm, there is a tradeoff here. One possibility is that I can increase profitability. The other is that I can pay attention to that group of geeks who are living in some other reality. Guess which wins, almost every time? There are two strategies that make more sense and have more chance of success. One is precisely what 4084 attempted to do: lay out categories and boundaries that, if adopted, make better information available to potential users/customers and provide a foundation for regulation about what must be accurately disclosed (as compared to what is required). That said, I've been quite disappointed with the results of 4084: from the comments and input I got before we did the work, I was optimistic that we would see at least some ISPs, and maybe even some regulators, pick the concepts and terminology up. To put it mildly, it hasn't happened. The other approach, with thanks to Dave Clark for pointing it out to me a few years ago, is to carefully write a neutral and balanced document whose theme is of course the Internet architecture permits you do this, but, if you do, it will have the following good and bad consequences which you should understand in making your decisions. Either approach requires serious work and people on the IAB who are interested, willing, and have the skills to do it. I can't speak for the current IAB at all but, if the sort of output Tony and I are talking about is wanted, then people need to tell the Nomcom(s) that the ability and willingness to generate it should be an important candidate selection criterion. john --On Thursday, March 23, 2006 20:20 -0600 Tony Hain [EMAIL PROTECTED] wrote: I didn't make it to the mic fast enough at the end, but Brian's comment about the proposal to outlaw diffserv actually gets to the heart of why the IAB needs to take specific stands and make public comments. Telling the telco's they are evil is not the point. General statements of principle or observations of past behavior like 'walled gardens are not conducive to open application innovation and frequently result in additional layering complexity to traverse the walls', or 'allowing people to elect going to the head of the line is what the QoS toolset is about'. I am not sure what the right language is but there is probably something the IAB could say about misusing the tools to effectively set up an extortion/protection racket being a possible side effect that regulators might want to consider, but that cutting off the tools outright would actually hamper some potential new service and application development. The point is that if the IAB stands back without making any statement there will be no guidance about the impacts of various business/deployment models. Something along the lines of 4084 that takes no particular position of right or wrong, but identifies the consequences of potential actions might help to stabilize the public debate. After all even open application development might be considered wrong by some, but when coupled with the observation that it happens anyway with more complexity and cost might get all the fundamental issues on the table. Tony ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
On 3/25/06 7:47 PM, Spencer Dawkins [EMAIL PROTECTED] wrote: So my point was, I'd really like to take a chance on some IAB statements about things that need to be stated about our architecture. They might be ignored. Would the result be any worse? This is a somewhat bothersome case, because the IAB *did* issue an RFC explaining what many of the problems were with Unilateral Network Self-Address Fixing (i.e. STUN). They included a list of conditions they felt that an UNSAF protocol had to meet in order to be published, including a description of a transition mechanism away from itself and towards something more robust. I don't know what more the IAB could have done in order to kill what I think is a clearly pathological approach to NAT traversal (and I chaired the working group that standardized it, so I accept a great deal of responsibility for this mess), but if putting out a document that says These are the reasons that this isn't a good protocol isn't enough, well, I'm not sure. But it seems to me that trying to fix it this late in the process (my other .sig is software longa, hardware brevis) has less to do with architecture and more to do with oncology. At any rate, I do think that in this case the IAB did do their job and it was the rest of us louts who messed up. And I'll tell you where I think it happened: when we accepted the idea that something might be transitional and would eventually go away. Melinda ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
On 24 mar 2006, at 18.07, Hallam-Baker, Phillip wrote: If I am going to send a copy of a $200 million action movie to a viewer I am going to expect to be paid for that. The viewer is going to expect a high quality viewing experience. The problem is that the bandwidth they subscribe to for Web browsing purposes may not be great enough to support that viewing experience. If I am charging $8 for a movie I might well be willing to pay $0.50 to the carrier as a distribution fee in exchange for access to high bandwith pipe for an interval. See the document Geoff sent a link to. The way it works is that IF the end user want to be able to see high quality movies, he need to buy a good quality Internet Connection. One with good quality on the connections all the path that the receiver pay. The company sending the movie have to pay to get quality for the full path that the sender pay. What does not work is to route money over the Internet. It has never worked. The problem is that end users today pay for low quality Internet access, and then ask why they do not get high quality. Where we have a problem is when the access provider that have a bad quality packet exchange relation with some other ISP is also providing a video service. They the bad quality is used as a lock-in tool to keep the customer. I.e. there are many access providers/ISP's that have no interest what so ever to sell good quality interconnect to other ISP's. There is not enough economical force behind building it. Only path forward is, I think, that end users start to demand better service, and the ones that do are prepared on paying more. Like if you just want broadband, buy blue service, but if you want better quality, buy red service tied together with to be able to use our movie distribution service, you need a broadband access of at least red service. Then the market can show how many consumers want that better service, they pay more, and that money can be used for creating higher-quality interconnect. Patrik ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
Geoff, things were indeed different then, as long distance bandwidth costs were a serious concern. That has changed. I think the fact that content providers who are paid for that content don't (in effect) pay for the congestion that they cause hasn't changed. But mainly I was interested to see PHB making arguments quite close to the ones I made ten years ago. Brian Geoff Huston wrote: To quote from the Carpenter draft:... One approach to resolving the current crisis in Internet performance is to institute an efficient system of inter-carrier settlements. Progress is often hard when you are heading in off in the weeds. Try http://www.potaroo.net/ispcol/2005-01/interconn.html as an alternative view of the ISP settlement world. regards, Geoff At 12:12 PM 25/03/2006, Brian E Carpenter wrote: I know I'm going to regret saying this, but we haven't made much progress in ten years. http://tools.ietf.org/id/draft-carpenter-metrics-00.txt I got a lot of interest in that draft, none of which came from ISPs... Brian Hallam-Baker, Phillip wrote: I think that people need to consider that maybe there might be advantages to non-flat rate, non-consumer pays charging models. I don't expect the attempted shakedown of Google to work and there are certainly tactics that they could use to preclude any desire on the part of the carriers to do any such thing. A much more interesting case would be delivery of video on demand. This is surely what the proponents of the sender pays scheme are really thinking about. If I am going to send a copy of a $200 million action movie to a viewer I am going to expect to be paid for that. The viewer is going to expect a high quality viewing experience. The problem is that the bandwidth they subscribe to for Web browsing purposes may not be great enough to support that viewing experience. If I am charging $8 for a movie I might well be willing to pay $0.50 to the carrier as a distribution fee in exchange for access to high bandwith pipe for an interval. The point here is that higher bandwidth costs more to provide. If the bandwidth is provided to every subscriber all the time the costs are much greater than providing the ultra-high speed to a small pool of subscribers who need it for a limited time and purpose. If the high bandwidth is added to the general pool then it will be diluted by contention and the folk running file sharing ct. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
--On Friday, 24 March, 2006 16:28 -0600 Scott W Brim sbrim@cisco.com wrote: On Fri, Mar 24, 2006 05:00:07AM -0500, John C Klensin allegedly wrote: There are two strategies that make more sense and have more chance of success. One is precisely what 4084 attempted to do: lay out categories and boundaries that, if adopted, make ... Either approach requires serious work and people on the IAB who are interested, willing, and have the skills to do it. I can't speak for the current IAB at all but, if the sort of output Tony and I are talking about is wanted, then people need to tell the Nomcom(s) that the ability and willingness to generate it should be an important candidate selection criterion. These are great, John, but as you say, both approaches require serious work -- both before and after publication. In fact spreading an idea can take much more work, over a longer time, than agreeing on it, writing it up, and implementing it in the first place. Of course. The 4084 effort was just a first step. As has recently been pointed out to me, it may have been a first step that lost focus by digging down into the interests and issues of some Internet-consumer communities more than others, thereby neither maintaining a consistent high-level view nor clearly focusing in on an area or two. That said, if I had understood that focus problem at the time (and I understood it enough to be nervous, but not enough to get articulate about it), I would not have done anything differently because it seemed clear that there wasn't sufficient community interest to cope with a half-dozen documents, rather than one. But it, or even producing a revision or a few updates that focus better on specific communities and clusters of needs, are fairly easy: just as with 4084, someone can sit down and write, round up a handful of people to comment, and then write some more. The harder part requires people to stand up and call attention to the statements. That is where the analogy to RFC 1984 applies -- IAB and IESG statements carry far more weight than a random BCP and can be an important tool in focusing interest on a subject where policy or commercial interests become problematic for the Internet. A healthy Internet requires effort on three fronts: innovation to start with, deployment (not just of new ideas, but of what we have already to lesser developed areas), and finally trying to get our principles, conceptual framework, and attitudes accepted elsewhere. The first is the usual focus of IETF WGs. These days the third is increasingly important. In all cases it's not enough to launch something -- it needs to be nursed and championed for a long time after its birth. As Scott correctly points out, documents such as 4084, or even 1948, isn't all there is to do either. But, if the decision of the IETF community is that it is more important to spend energy exclusively on low-level technical issues or on administrative and procedural navel-gazing, then we should keep our expectations about leverage on this type of issues very low. The IAB's primary orientation should be toward breadth, not depth. Individual members can focus in particular areas but the IAB as a whole needs to cover a great deal of material on all three of these fronts. Doing a good job on all three legs of the stool takes hundreds of people. We non-IABers can generate the sort of thing you're talking about as well as the IAB, and we should. We should use the IAB as a focal point, lookouts, facilitators, instigators, conveners, as well as as individuals for their expertise and dedication. I think these capabilities are at least as important as being able to write up results of deliberation. We should take as least as much responsibility for doing the grunt work, including coming up with innovative ideas, writing documents like those you describe, and making sure results happen in the real world, as we expect IAB members to. I think we agree. I drew the effort that produced 4084 together after hearing that it was needed from too many people who wouldn't (or didn't feel that they could) do so themselves. So, turning Scott's discussion above around -- what are the rest of you doing? See you in Montreal. I hope to see enough action on this, including some drafts and some expression of interest from our leadership, long enough before Montreal that focused discussion and some conclusions there become possible. Perhaps that is a silly hope but if not now, when? john ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
On Sat, Mar 25, 2006 at 05:56:09PM +0100, Patrik F?ltstr?m wrote: Only path forward is, I think, that end users start to demand better service, and the ones that do are prepared on paying more. Like if you just want broadband, buy blue service, but if you want better quality, buy red service tied together with to be able to use our movie distribution service, you need a broadband access of at least red service. Of course the other model is that the content provider temporarily upgrades your blue service to red service for their streams. If content providers cannot do this and most people have blue service, then the market for content that requires red service may be less then the critical mass required to make providing the content viable. The problem is how do we differentiate between cases where content providers pay to get a higher then default QOS for their streams vs. the case where the provider pays to prevent the ISP from intentionally interfering with their streams. I believe the former is reasonable while the later is extortion. Then there is the threat that ISP's will permit their default level of service to degrade over time because all those who care must now pay for higher QOS. -Jeff -- = Jeffrey I. Schiller MIT Network Manager Information Services and Technology Massachusetts Institute of Technology 77 Massachusetts Avenue Room W92-190 Cambridge, MA 02139-4307 617.253.0161 - Voice [EMAIL PROTECTED] signature.asc Description: Digital signature ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
--On Saturday, 25 March, 2006 12:54 -0500 Jeffrey I. Schiller [EMAIL PROTECTED] wrote: The problem is how do we differentiate between cases where content providers pay to get a higher then default QOS for their streams vs. the case where the provider pays to prevent the ISP from intentionally interfering with their streams. I believe the former is reasonable while the later is extortion. Then there is the threat that ISP's will permit their default level of service to degrade over time because all those who care must now pay for higher QOS. The latter is, of course, one of the interesting behaviors that have been observed in parts of the US (at least) from TV cable providers. And, behold, several of them have become major ISPs for the residential / telecommuter/ SOHO market, so we are dealing with a group that has already learned that particular trick. :-( john ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
Brian, Actually the document I referenced is also around 9 years old - so even then we were having a Fine Debate about settlement systems in this industry. The introduction of Content into this debate has also been interesting with the earliest intersection of the two groups (ISPs and content factories) resulting in the claims of you have to pay me coming from the content industry and being directed to the IP access providers, while the precise opposite is the case today. (Some reflections arising from the first set of encounters are at http://www.potaroo.net/ispcol/2001-06/2001-06-content.html if anyone is vaguely interested in such things!) Content was, and remains, a distinct overlay economy and making claims that content providers should pay ISPs for the shortcomings in the ISP's own network engineering are around as specious as earlier claims that that ISPs should pay content providers for content that their customers may well have been completely uninterested in! (Bundling service and infrastructure, in whatever form, also strikes me as yet another reprise of that 'convergence' nonsense that has been inflicted on this industry for some decades now, primarily by folk looking desperately for monopolistic relief from the harsh realities of a highly competitive deregulated communications industry.) regards, Geoff At 02:02 AM 26/03/2006, Brian E Carpenter wrote: Geoff, things were indeed different then, as long distance bandwidth costs were a serious concern. That has changed. I think the fact that content providers who are paid for that content don't (in effect) pay for the congestion that they cause hasn't changed. But mainly I was interested to see PHB making arguments quite close to the ones I made ten years ago. Brian ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
I don't mean to hijack this conversation, only add a data point... I have a great deal of respect for the people who have done the heavy lifting in BEHAVE, but it seems like every time we meet, someone discovers a new and previously un-observed NAT behavior that Is Not Helpful. This week was the best yet... In a recent posting (http://list.sipfoundry.org/archive/ietf-behave/msg01189.html), Dan Wing said: That isn't quite the scenario. The scenario where the UDP/TCP interworking is useful is this: Alice---[NAT/firewall]---[TURN][NAT]---Bob ---TCP|---UDP Where Alice's NAT/firewall device blocks UDP. Many company firewalls have that behavior, which cause applications such as Skype and Yahoo Voice to send their traffic over TCP. Bob, on the other hand, has a 'normal' NAT, however Bob's endpoint has no support for framing RTP over TCP. This is pretty common -- many endpoints have no ability to send their RTP traffic over TCP. So, this functionality provides a way for Alice and Bob to communicate where they couldn't communicate if they were both forced to use UDP (because Alice's firewall blocks UDP). OTOH we definetly need to nail down framing in the TCP-TURN-UDP scenario. I think we can use the TCP framing mechanism that Jonthan proposed at the behave meeting; the server would emit every TCP 'chunk' as a UDP packet. I'm still thinking about the telling our children about the ancient past when it was possible/legal for an ordinary person to put a server on the Internet remark from Thursday's plenary. Dan's note, quoted above, is an excellent summary of where we are now, and it is not too much of a leap to JohnK's favorite rhetorical question, where are we going, and how did we get in this handbasket? So my point was, I'd really like to take a chance on some IAB statements about things that need to be stated about our architecture. They might be ignored. Would the result be any worse? Thanks, Spence P.s. And how many working group meetings did YOU sit in this past week, where firewall and NAT traversal were affecting our protocol designs? I think I counted six different working groups (including BEHAVE). Dan Wing could be thinking about USEFUL problems, if we weren't distracting him with stuff like this. Ditto the rest of the ICE/TURN/STUN crew. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
Tony, I agree completely and believe the IAB has, of late, been altogether too timid in this area. I think you know all of what I'm about to say, but your note is, IMO, easily misread, so an additional observation about 4084 and its potential relatives: In this sphere, a document that says XYZ is evil is essential worthless. The companies considering XYZ will almost always say hmm, there is a tradeoff here. One possibility is that I can increase profitability. The other is that I can pay attention to that group of geeks who are living in some other reality. Guess which wins, almost every time? There are two strategies that make more sense and have more chance of success. One is precisely what 4084 attempted to do: lay out categories and boundaries that, if adopted, make better information available to potential users/customers and provide a foundation for regulation about what must be accurately disclosed (as compared to what is required).That said, I've been quite disappointed with the results of 4084: from the comments and input I got before we did the work, I was optimistic that we would see at least some ISPs, and maybe even some regulators, pick the concepts and terminology up. To put it mildly, it hasn't happened. The other approach, with thanks to Dave Clark for pointing it out to me a few years ago, is to carefully write a neutral and balanced document whose theme is of course the Internet architecture permits you do this, but, if you do, it will have the following good and bad consequences which you should understand in making your decisions. Either approach requires serious work and people on the IAB who are interested, willing, and have the skills to do it. I can't speak for the current IAB at all but, if the sort of output Tony and I are talking about is wanted, then people need to tell the Nomcom(s) that the ability and willingness to generate it should be an important candidate selection criterion. john --On Thursday, March 23, 2006 20:20 -0600 Tony Hain [EMAIL PROTECTED] wrote: I didn't make it to the mic fast enough at the end, but Brian's comment about the proposal to outlaw diffserv actually gets to the heart of why the IAB needs to take specific stands and make public comments. Telling the telco's they are evil is not the point. General statements of principle or observations of past behavior like 'walled gardens are not conducive to open application innovation and frequently result in additional layering complexity to traverse the walls', or 'allowing people to elect going to the head of the line is what the QoS toolset is about'. I am not sure what the right language is but there is probably something the IAB could say about misusing the tools to effectively set up an extortion/protection racket being a possible side effect that regulators might want to consider, but that cutting off the tools outright would actually hamper some potential new service and application development. The point is that if the IAB stands back without making any statement there will be no guidance about the impacts of various business/deployment models. Something along the lines of 4084 that takes no particular position of right or wrong, but identifies the consequences of potential actions might help to stabilize the public debate. After all even open application development might be considered wrong by some, but when coupled with the observation that it happens anyway with more complexity and cost might get all the fundamental issues on the table. Tony ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
RE: the iab net neutrality
May be if you think the other way around, you reinvent the Minitel model? Not sure as the final text is not voted and is _very_ confused, but this _may_ be what the French DADVSI law _may_ lead to. jfc At 18:07 24/03/2006, Hallam-Baker, Phillip wrote: Content-class: urn:content-classes:message Content-Type: multipart/signed; protocol=application/x-pkcs7-signature; micalg=SHA1; boundary==_NextPart_000_0014_01C64F3B.8ABDFF10; x-avg-checked=avg-ok-77F14237 I think that people need to consider that maybe there might be advantages to non-flat rate, non-consumer pays charging models. I don't expect the attempted shakedown of Google to work and there are certainly tactics that they could use to preclude any desire on the part of the carriers to do any such thing. A much more interesting case would be delivery of video on demand. This is surely what the proponents of the sender pays scheme are really thinking about. If I am going to send a copy of a $200 million action movie to a viewer I am going to expect to be paid for that. The viewer is going to expect a high quality viewing experience. The problem is that the bandwidth they subscribe to for Web browsing purposes may not be great enough to support that viewing experience. If I am charging $8 for a movie I might well be willing to pay $0.50 to the carrier as a distribution fee in exchange for access to high bandwith pipe for an interval. The point here is that higher bandwidth costs more to provide. If the bandwidth is provided to every subscriber all the time the costs are much greater than providing the ultra-high speed to a small pool of subscribers who need it for a limited time and purpose. If the high bandwidth is added to the general pool then it will be diluted by contention and the folk running file sharing ct. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
maybe I can summerize John's note by asking if this IAB has the will to write a RFC 1984 about net neutrality Scott ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
I know I'm going to regret saying this, but we haven't made much progress in ten years. http://tools.ietf.org/id/draft-carpenter-metrics-00.txt I got a lot of interest in that draft, none of which came from ISPs... Brian Hallam-Baker, Phillip wrote: I think that people need to consider that maybe there might be advantages to non-flat rate, non-consumer pays charging models. I don't expect the attempted shakedown of Google to work and there are certainly tactics that they could use to preclude any desire on the part of the carriers to do any such thing. A much more interesting case would be delivery of video on demand. This is surely what the proponents of the sender pays scheme are really thinking about. If I am going to send a copy of a $200 million action movie to a viewer I am going to expect to be paid for that. The viewer is going to expect a high quality viewing experience. The problem is that the bandwidth they subscribe to for Web browsing purposes may not be great enough to support that viewing experience. If I am charging $8 for a movie I might well be willing to pay $0.50 to the carrier as a distribution fee in exchange for access to high bandwith pipe for an interval. The point here is that higher bandwidth costs more to provide. If the bandwidth is provided to every subscriber all the time the costs are much greater than providing the ultra-high speed to a small pool of subscribers who need it for a limited time and purpose. If the high bandwidth is added to the general pool then it will be diluted by contention and the folk running file sharing ct. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
On Fri, Mar 24, 2006 05:00:07AM -0500, John C Klensin allegedly wrote: There are two strategies that make more sense and have more chance of success. One is precisely what 4084 attempted to do: lay out categories and boundaries that, if adopted, make better information available to potential users/customers and provide a foundation for regulation about what must be accurately disclosed (as compared to what is required).That said, I've been quite disappointed with the results of 4084: from the comments and input I got before we did the work, I was optimistic that we would see at least some ISPs, and maybe even some regulators, pick the concepts and terminology up. To put it mildly, it hasn't happened. The other approach, with thanks to Dave Clark for pointing it out to me a few years ago, is to carefully write a neutral and balanced document whose theme is of course the Internet architecture permits you do this, but, if you do, it will have the following good and bad consequences which you should understand in making your decisions. Either approach requires serious work and people on the IAB who are interested, willing, and have the skills to do it. I can't speak for the current IAB at all but, if the sort of output Tony and I are talking about is wanted, then people need to tell the Nomcom(s) that the ability and willingness to generate it should be an important candidate selection criterion. These are great, John, but as you say, both approaches require serious work -- both before and after publication. In fact spreading an idea can take much more work, over a longer time, than agreeing on it, writing it up, and implementing it in the first place. A healthy Internet requires effort on three fronts: innovation to start with, deployment (not just of new ideas, but of what we have already to lesser developed areas), and finally trying to get our principles, conceptual framework, and attitudes accepted elsewhere. The first is the usual focus of IETF WGs. These days the third is increasingly important. In all cases it's not enough to launch something -- it needs to be nursed and championed for a long time after its birth. The IAB's primary orientation should be toward breadth, not depth. Individual members can focus in particular areas but the IAB as a whole needs to cover a great deal of material on all three of these fronts. Doing a good job on all three legs of the stool takes hundreds of people. We non-IABers can generate the sort of thing you're talking about as well as the IAB, and we should. We should use the IAB as a focal point, lookouts, facilitators, instigators, conveners, as well as as individuals for their expertise and dedication. I think these capabilities are at least as important as being able to write up results of deliberation. We should take as least as much responsibility for doing the grunt work, including coming up with innovative ideas, writing documents like those you describe, and making sure results happen in the real world, as we expect IAB members to. See you in Montreal. swb ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
To quote from the Carpenter draft:... One approach to resolving the current crisis in Internet performance is to institute an efficient system of inter-carrier settlements. Progress is often hard when you are heading in off in the weeds. Try http://www.potaroo.net/ispcol/2005-01/interconn.html as an alternative view of the ISP settlement world. regards, Geoff At 12:12 PM 25/03/2006, Brian E Carpenter wrote: I know I'm going to regret saying this, but we haven't made much progress in ten years. http://tools.ietf.org/id/draft-carpenter-metrics-00.txt I got a lot of interest in that draft, none of which came from ISPs... Brian Hallam-Baker, Phillip wrote: I think that people need to consider that maybe there might be advantages to non-flat rate, non-consumer pays charging models. I don't expect the attempted shakedown of Google to work and there are certainly tactics that they could use to preclude any desire on the part of the carriers to do any such thing. A much more interesting case would be delivery of video on demand. This is surely what the proponents of the sender pays scheme are really thinking about. If I am going to send a copy of a $200 million action movie to a viewer I am going to expect to be paid for that. The viewer is going to expect a high quality viewing experience. The problem is that the bandwidth they subscribe to for Web browsing purposes may not be great enough to support that viewing experience. If I am charging $8 for a movie I might well be willing to pay $0.50 to the carrier as a distribution fee in exchange for access to high bandwith pipe for an interval. The point here is that higher bandwidth costs more to provide. If the bandwidth is provided to every subscriber all the time the costs are much greater than providing the ultra-high speed to a small pool of subscribers who need it for a limited time and purpose. If the high bandwidth is added to the general pool then it will be diluted by contention and the folk running file sharing ct. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
the iab net neutrality
I didn't make it to the mic fast enough at the end, but Brian's comment about the proposal to outlaw diffserv actually gets to the heart of why the IAB needs to take specific stands and make public comments. Telling the telco's they are evil is not the point. General statements of principle or observations of past behavior like 'walled gardens are not conducive to open application innovation and frequently result in additional layering complexity to traverse the walls', or 'allowing people to elect going to the head of the line is what the QoS toolset is about'. I am not sure what the right language is but there is probably something the IAB could say about misusing the tools to effectively set up an extortion/protection racket being a possible side effect that regulators might want to consider, but that cutting off the tools outright would actually hamper some potential new service and application development. The point is that if the IAB stands back without making any statement there will be no guidance about the impacts of various business/deployment models. Something along the lines of 4084 that takes no particular position of right or wrong, but identifies the consequences of potential actions might help to stabilize the public debate. After all even open application development might be considered wrong by some, but when coupled with the observation that it happens anyway with more complexity and cost might get all the fundamental issues on the table. Tony ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: the iab net neutrality
This directly relates to the Skype discussion during the plenary. Skype will, if necessary, tunnel media on port 80 and port 443. To some extent, the debate also highlights a lack of usable protocol tools: One reason, albeit likely not the only one, that there is talk about per-source wholesale charging for improved QoS is that we don't currently have a viable inter-provider retail mechanism that allows individuals and small companies, for example, to request and pay for a fixed-bandwidth pipe between random points on the Internet on short notice. The inability to offer such services also biases things like IPTV towards being provided by those owning the wires and DSLAMs, rather than third parties, even without explicit discrimination. Henning Tony Hain wrote: I didn't make it to the mic fast enough at the end, but Brian's comment about the proposal to outlaw diffserv actually gets to the heart of why the IAB needs to take specific stands and make public comments. Telling the telco's they are evil is not the point. General statements of principle or observations of past behavior like 'walled gardens are not conducive to open application innovation and frequently result in additional layering complexity to traverse the walls', or 'allowing people to elect going to the head of the line is what the QoS toolset is about'. I am not sure what the right language is but there is probably something the IAB could say about misusing the tools to effectively set up an extortion/protection racket being a possible side effect that regulators might want to consider, but that cutting off the tools outright would actually hamper some potential new service and application development. The point is that if the IAB stands back without making any statement there will be no guidance about the impacts of various business/deployment models. Something along the lines of 4084 that takes no particular position of right or wrong, but identifies the consequences of potential actions might help to stabilize the public debate. After all even open application development might be considered wrong by some, but when coupled with the observation that it happens anyway with more complexity and cost might get all the fundamental issues on the table. Tony ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
