- Wednesday, November 16, 2016

homenet@jabber.ietf.org

Wednesday, November 16, 2016< ^ >

Barbara Stark has set the subject to: Homenet @ IETF96

Room Configuration

Room Occupants

GMT+0
[04:14:33] Ralph Droms joins the room
[04:16:53] Ralph Droms joins the room
[04:17:17] Ralph Droms leaves the room
[04:17:47] Ralph Droms leaves the room
[04:22:22] Meetecho joins the room
[04:25:12] John Border joins the room
[04:25:12] Mark Andrews joins the room
[04:25:12] Ralph Droms joins the room
[04:25:12] Juliusz Chroboczek joins the room
[04:28:20] Terry Manderson joins the room
[04:28:33] Brian Carpenter joins the room
[04:30:04] Wentao Shang joins the room
[04:30:42] Yingzhen Qu joins the room
[04:31:07] Tim Chown joins the room
[04:31:16] Ray Atarashi joins the room
[04:31:52] Brian Carpenter leaves the room
[04:33:27] Tim Chown leaves the room
[04:33:40] Mikael Abrahamsson joins the room
[04:33:50] <Mikael Abrahamsson> I'll be jabber scribe.
[04:34:13] <Juliusz Chroboczek> (waves)
[04:34:51] <Ralph Droms> How do I get the presentation materials feed back up in my browser?
[04:35:17] Suzanne joins the room
[04:35:31] Andrew Sullivan joins the room
[04:35:44] <Mikael Abrahamsson> ralph, not sure what you mean... are you in meetecho?
[04:35:50] <Ralph Droms> yes.
[04:35:54] <Ralph Droms> yes.
[04:36:05] <Wentao Shang> try the buttons on the top right of the interface?
[04:36:08] click speaker
[04:36:34] <Mikael Abrahamsson> ralph, you want to "raise hand" to get into the "I want to speak" queue?
[04:37:00] presentation mode
[04:37:05] <Ralph Droms> No, I want to see the presentation materials in addition to the video feed of the chairs.
[04:37:11] <Meetecho> what you you mean by presentation materials? the pdf?
[04:37:13] <Ralph Droms> It was visible earlier.
[04:37:27] <Ralph Droms> What's being projected on the screen in the meeting.
[04:37:32] <Meetecho> aaah the slides
[04:37:36] <Meetecho> they should be up
[04:37:37] lectern icon
[04:37:47] <Meetecho> there's a podium icon
[04:37:54] <Ralph Droms> Yeah, got it.
[04:37:57] <Meetecho> try clicking that if you changed layout for some reason
[04:38:04] <Meetecho> (y)
[04:38:09] <Ralph Droms> Thanks.
[04:38:35] <Mikael Abrahamsson> Dave Thaler at mic
[04:39:11] <Juliusz Chroboczek> How many people are in the room?
[04:39:42] <Mikael Abrahamsson> about 50-60 people
[04:39:47] Steve Olshansky joins the room
[04:39:52] <Juliusz Chroboczek> ty
[04:39:57] <Mikael Abrahamsson> 40-60 perhaps, I didn't count exactly
[04:40:37] wmt joins the room
[04:40:47] <wmt> Whew, made it in.
[04:41:53] Victor Kuarsingh joins the room
[04:42:18] marco@nl joins the room
[04:42:35] <Mikael Abrahamsson> Dave Thaler at mic
[04:43:19] <Mikael Abrahamsson> Ted Lemon at mic
[04:43:30] Ray Pelletier joins the room
[04:43:43] Jason Weil joins the room
[04:43:44] Jun Hu joins the room
[04:44:13] danyork joins the room
[04:44:18] <Andrew Sullivan> I am in favour of all extended bikeshed metaphors here
[04:44:32] Mikael Abrahamsson leaves the room
[04:45:57] <Juliusz Chroboczek> Meetecho: being able to see the mike line during questions is unusual and highly appreciated.
[04:46:17] <Meetecho> Juliusz Chroboczek: we try to do that when we can!
[04:46:59] <Meetecho> ps: are chairs aware of Ralph Droms in the virtual queue?
[04:49:17] Ray Atarashi leaves the room
[04:49:48] <Ralph Droms> Can someone nudge the chairs for me?
[04:49:54] Mikael Abrahamsson2 joins the room
[04:50:05] Audric Schiltknecht joins the room
[04:50:14] <Mikael Abrahamsson2> ok, I was disconnected
[04:50:24] <Mikael Abrahamsson2> so right now Peter Koch and Ted Lemon is arguing at the mic
[04:50:31] <Mikael Abrahamsson2> Ted Lemon
[04:50:33] <Andrew Sullivan> I think that speculation on people's future mental states is maybe not a good use of WG time
[04:50:50] mcr joins the room
[04:50:54] <mcr> http://graphemica.com/%F0%9F%8F%A0
[04:50:57] <Juliusz Chroboczek> (giggle)
[04:51:39] <mcr> 🏠.arpa
[04:51:42] <Terry Manderson> fwiw related to .arpa
[04:51:43] <Terry Manderson> The IAB shall only recommend the creation of "arpa" sub-domains
 corresponding to protocol entities where:
 - the delegation, and the hierarchical name structure, is described
 by an IETF Standards Track document [4], and
 - the use of the "arpa" domain is explicitly recommended in the
 "IANA Considerations" section of that document.
[04:51:52] <Mikael Abrahamsson2> Andrew Sullivan
[04:52:37] <Terry Manderson> I'm not pushing for either option at this point as AD btw.
[04:52:37] Simon Pietro Romano joins the room
[04:53:08] <Andrew Sullivan> @Terry, yes, and I am comfortable suggesting that this WG has the ability to meet those criteria
[04:53:45] Simon Romano joins the room
[04:54:41] <Andrew Sullivan> I don't think that home _did_ go through the IETF process, because it violated the 6761 processes which is why some of us missed it
[04:55:00] <Andrew Sullivan> But I agree we should change to something, and again I do not care what the string is
[04:55:14] <Andrew Sullivan> but I pointed out what the trade-offs are for each strategy
[04:55:41] <Terry Manderson> I think well noted Andrew.
[04:55:55] <Suzanne> @ajs likewise….I'm increasingly convinced that the point about *who* is being relied upon to protect a name from collision is important.
[04:55:58] <Mikael Abrahamsson2> Jason Livingood at mic
[04:56:59] <Andrew Sullivan> I don't think "homenet" is likely to be any friendlier for a user. We could speculate all day long about what will be intuitive to users
[04:57:11] <Andrew Sullivan> I think _anything_ that exposes _any of these names_ to users will be mystifying
[04:57:14] <Suzanne> especially if we stop implicitly stipulating the user is an American.
[04:57:20] <Andrew Sullivan> indeed
[04:57:30] <Mikael Abrahamsson2> Tim Chown at mic
[04:57:36] <Juliusz Chroboczek> Was that Mark?
[04:57:44] <Andrew Sullivan> That was Mark speaking yes
[04:57:47] <Mikael Abrahamsson2> yes, that was Mark
[04:58:07] <Mikael Abrahamsson2> Pierre at mic
[04:58:11] <Andrew Sullivan> and you can't treat emojis as "ordinary domain names" in some cases
[04:58:29] <Andrew Sullivan> (depends basically on what "ordinary" means there)
[04:58:33] <Ralph Droms> I disagree that RFC 7788 defines the behavior of the homenet name resolution mechanism.
[04:59:01] <Mikael Abrahamsson2> Ted Lemon at mic
[04:59:39] <Juliusz Chroboczek> Ralph: agreed. However, it does define a number of TLVs that contain names, and implementors need to know what to stick into those TLVs.
[04:59:53] <Andrew Sullivan> The technical reason a TLD is the wrong name to choose is that the namespace control is under dispute
[05:00:11] <Juliusz Chroboczek> So the more we delay on choosing the string, the higher the chances that implementations using .home escape in the wild.
[05:00:11] <Ralph Droms> Mark said that the resolution mechanism is defined in RFC 7788 and, therefore, we can move on to discussing the name.
[05:00:26] <Andrew Sullivan> (The response to that might be that the namespace isn't, because of 6761. My point is that this is an argument one could make, which could yield an appeal.)
[05:00:36] <Mikael Abrahamsson2> Avri Doria at mic
[05:00:36] <marco@nl> 🏠is technically feasable ;-)
[05:00:44] <Andrew Sullivan> Agree with Juliusz
[05:00:58] <Ralph Droms> Juliusz: is the idea that the name can be reserved and the resolution mechanism can be defined later?
[05:01:03] <Suzanne> I object on the basis that a "TLD" involves dependency on an external body, outside of the IETF and its accountability mechanisms, to protect the name from collision.
[05:01:05] <Andrew Sullivan> Actually, 🏠 is not technically feasible
[05:01:20] <Andrew Sullivan> because it's unclear what its treatment ought to be
[05:01:54] <Mikael Abrahamsson2> Dave Thaler at mic
[05:01:56] <Juliusz Chroboczek> The idea is that we need to reserve a name ASAP.
[05:02:03] <Andrew Sullivan> because IDNA2003 might possbly or maybe not handle it as IDNA, and IDNA2008 says it's not allowed, and DNS says it is but you don't know for sure that it's UTF-8 since there's no encoding bits
[05:02:20] <Juliusz Chroboczek> A name resolution mechanism is pretty much orthogonal.
[05:03:11] <Andrew Sullivan> for "TLD" hear "domain name label in the top-most position"
[05:03:12] <Mikael Abrahamsson2> Bob Hinden at mic
[05:03:22] <Andrew Sullivan> in an effort to get out of the "DNS root" argument
[05:03:30] <Mikael Abrahamsson2> Stuart Cheshire at mic
[05:03:32] <marco@nl> So, xn--um8h is not an option?
[05:03:50] <Suzanne> @ajs I like references to "single domain name label" even better :)
[05:03:57] <marco@nl> (reading RFC5895 now)
[05:03:58] Ray Atarashi joins the room
[05:04:10] <Andrew Sullivan> @marco, no
[05:04:18] <marco@nl> ack
[05:04:34] <Andrew Sullivan> because IDNA2008 requires labels to be made up of "letters" or "numbers"
[05:04:40] <Andrew Sullivan> and emoji are neither
[05:05:08] <Mikael Abrahamsson2> Terry Manderson at mic
[05:05:15] <Andrew Sullivan> we anyway don't want to dig into the rathole that is the interop among IDNA2003, IDNA2008, and UTS-46
[05:05:52] <Mikael Abrahamsson2> Mark T speaking now
[05:05:53] danyork leaves the room: Disconnected: closed
[05:06:07] <Mikael Abrahamsson2> Terry speaking again
[05:06:14] <Meetecho> note to chairs: Ralph Droms in virtual queue
[05:06:18] <Andrew Sullivan> Terry can't actually promisse that, of course, because of IETF consensus
[05:06:21] <Andrew Sullivan> maybe one of them would die
[05:06:27] <Meetecho> (just writing in case they're not aware, let me know if that's not needed)
[05:06:28] <Andrew Sullivan> which is what Stuart was saying
[05:07:14] Dave Thaler joins the room
[05:07:19] Colin Petrie joins the room
[05:07:38] Peter Koch joins the room
[05:07:50] <Terry Manderson> If the WG consensus AND the ietf consensus is for one document then so be it. but i have to get this moving.. and this is a way.
[05:08:15] <Terry Manderson> I don't see the second documnet dying as then homenet would be broken. and stay broken.
[05:09:18] <Andrew Sullivan> I agree with all that. Just being super clear that AD can't promise what the IETF will do, no matter who asks you :)
[05:09:29] <Juliusz Chroboczek> I just lost video by clicking at the wrong place. How do I get it back?
[05:09:29] <Terry Manderson> lol
[05:10:07] <Mikael Abrahamsson2> juliusz, there are different icons that represent different displaymodes, in the top right. Try "presentation mode"
[05:10:13] <mcr> My son (11yr) just asked me what my meeting is about. I said: It's about how we build networks in homes, and how we name computers in the home, and we tried to name home computers things like "tyson.home", but people objected to the use of ".home", saying that we didn't go through the process properly.
[05:10:20] <Juliusz Chroboczek> Ah, found it.
[05:10:29] <mcr> (my won recently named our printer tyson after a Percy Jackson character)
[05:11:33] <Mikael Abrahamsson2> So personally I think we should continue with the current proposal unless we know with high degree of probability that it will fail. If we think it's at least likely it will succeed, let's go with this.
[05:12:10] Moritz joins the room
[05:13:10] <Terry Manderson> @mcr its more than that. not going through the process meant that technical considerations of collision in that .home is already used by some vendors in an ad-hoc way was completely missed… but that is now history.
[05:14:06] <mcr> Terry, my only point was, could I simplify things enough to explain to my kid, in one sentence :-)
[05:14:10] danyork joins the room
[05:14:34] <Juliusz Chroboczek> (The nice thing about it being 6am here is that I find it really difficult to care.)
[05:15:07] <Andrew Sullivan> I can simplify most things so that people who don't know anything about the topic will understand me as long as I'm not constrained to being accurate
[05:15:58] wmt leaves the room
[05:16:12] <Andrew Sullivan> I won't go to the mic, but to get it captured, there's a tiny flaw in the 6761 template
[05:16:29] <Andrew Sullivan> 6. DNS server operators should not configure DNS servers to act as
 authoritative for any name ending in '.homenet'.
[05:16:47] wmt joins the room
[05:16:47] <Andrew Sullivan> This should probably be 6. DNS server operators should not configure DNS servers to act as
 authoritative for any name ending in '.homenet' unless they are providing services to a homenet
[05:16:51] <Juliusz Chroboczek> Ralph, I think that you're right from a technical point of view, but either we define a default, or ask the HNCP implementors to rip out that part of the code until the naming mechanisms are fully and accurately defined. And I don't think that's gonna happen.
[05:16:52] <Andrew Sullivan> or something like that
[05:17:11] <wmt> Thank you for the clarification on that, Terry.
[05:18:29] <Mikael Abrahamsson2> Stuart Cheshire at mic
[05:18:39] <Ralph Droms> Andrew - I find "DNS server operators" to be confusing. *Which* operators; "operators" always implies to me something like an ISP.
[05:18:45] except for DS HOMENET which needs a NOERROR NODATA response
[05:19:06] <Dave Thaler> I didn't hear an explicit statement of consensus from the chairs. Is there WG consensus or not? (I want it minuted :)
[05:19:27] <Andrew Sullivan> These points plus what Stuart is saying suggests that some additional work on the document is needed
[05:19:35] <Andrew Sullivan> and I agree Dave, there was no explicit statement
[05:19:37] <Mikael Abrahamsson2> Barbara Stark at mic
[05:19:46] <Andrew Sullivan> which means that there's no determination of consensus yet
[05:19:55] <Ralph Droms> I suggested in e-mail just before the mtg "DNS servers should not be configured to be authoritative for .homenet unless part of a home network naming system"
[05:20:05] <Dave Thaler> I can ask again at the end of the mic whether the chairs call consensus and thus the issue is closed, or not.
[05:20:25] <Suzanne> I'm not sure what scope Stuart just suggested for HNCP names, so I'm not sure what he's suggesting is consistent with what's in 7788
[05:20:29] <Dave Thaler> I also need to relay Joe's point about the redact doc, there's not a separate presentation on that doc right?
[05:20:29] <Ralph Droms> Sorry, Dave, I've reached my multitasking limit. Consensus on *what*?
[05:20:30] HOMENET needs a insecure delegating in the root zone so that validators don't need to know that homenet is special.
[05:21:01] <Dave Thaler> TLD vs not-a-TLD (and thus consensus on .homenet which seems to be what the likely consensus would be from what I've heard)
[05:21:03] <Andrew Sullivan> I don't believe that the IETF can specify an insecure delegation in the root, Mark
[05:21:15] <Andrew Sullivan> so that is in fact a technical argument in favour of homenet.arpa
[05:21:23] <Mikael Abrahamsson2> Barbara Stark at mic
[05:21:32] <Peter Koch> @Mark Andrews: how is that different from the onion and local (non)TLDs?
[05:21:54] <Mikael Abrahamsson2> Michael Richardson at mic
[05:22:13] Ray Pelletier leaves the room
[05:22:40] .local in mdns
[05:22:48] .onion is tor
[05:22:51] <Mikael Abrahamsson2> Stuart Cheshire at mic
[05:22:59] .homenet is DNS
[05:23:00] <Andrew Sullivan> I would suggest that word-smithing efforts around this doc could be done better on the list
[05:23:27] DNS has DNSSEC
[05:23:40] <Andrew Sullivan> I think Mark has a point here
[05:24:02] <Andrew Sullivan> Do you want it relayed, Mark, or will do you it yourself via pac-man?
[05:24:11] <Andrew Sullivan> (Also, I note Ralph is in the remote queue)
[05:24:35] <Andrew Sullivan> Now 2 remote queue members
[05:24:42] <Mikael Abrahamsson2> Wes Hardaker at mic
[05:25:38] <Mikael Abrahamsson2> Stuart Cheshire at mic
[05:26:24] <Mikael Abrahamsson2> Mark T speaking now
[05:26:41] <Mikael Abrahamsson2> Ralph Droms speaking remotely
[05:26:59] <Andrew Sullivan> In the interests of time, I sat down
[05:27:17] <Andrew Sullivan> but I got up because of the claim that things won't leak because it says "MUST NOT forward"
[05:27:26] <Andrew Sullivan> things that don't implement this spec will leak those names
[05:27:34] <Andrew Sullivan> because they're just ordinary domain names
[05:27:43] <Andrew Sullivan> the names _will_ leak, is Wes's point
[05:27:51] Ray Pelletier joins the room
[05:28:45] <Andrew Sullivan> Please note that what Mark is saying means this _can't_ be a right-most label in a domain name, though
[05:28:51] <Mikael Abrahamsson2> Tim Chown at mic
[05:28:58] <Andrew Sullivan> because the draft can't instruct ICANN to put an insecure delegation in the root
[05:29:12] <Mikael Abrahamsson2> Dave Thaler at mic
[05:29:14] <Ralph Droms> Yes. And a device that is used inside a non-leaking homenet is quite likely to continue using names from the .homenet zone after it moves out of the homenet ...
[05:29:15] <Juliusz Chroboczek> what does "delegation" mean in this context? Actually putting NS glue? Or a purely administrative thing?
[05:29:24] <Mikael Abrahamsson2> Stuart Cheshire at mic
[05:29:37] <Andrew Sullivan> You have to put a DS record that results in NODATA NOERROR
[05:29:53] <Juliusz Chroboczek> right, clear.
[05:29:56] Mark Andrews leaves the room
[05:30:00] Mark Andrews joins the room
[05:30:13] Ray Pelletier leaves the room
[05:30:28] <Andrew Sullivan> But if we mean it that this has to work with DNSSEC, then I think Mark's point is devastating for any "TLD" argument
[05:30:43] delegation
[05:31:10] you delegate back to the root servers
[05:31:30] <Peter Koch> @Andrew: the 6761 template actually already asks for a delegation of the TLD to IANA servers
[05:31:32] foo.homenet -> NXDOMAIN
[05:31:53] <Mikael Abrahamsson2> Dave Thaler at mic
[05:32:03] DS HOMENET -> NOERROR NODATA
[05:32:14] <Andrew Sullivan> @Peter: ?
[05:32:16] <Andrew Sullivan> In https://tools.ietf.org/html/draft-ietf-homenet-dot-00
[05:32:17] <Andrew Sullivan> ?
[05:32:22] Jason Weil leaves the room
[05:32:23] NS HOMENET -> NS RRSET -> root servers
[05:32:27] Jason Weil joins the room
[05:32:34] Jason Weil leaves the room
[05:32:43] <Andrew Sullivan> where?
[05:32:48] <Peter Koch> yes; section 3, question 7
[05:33:16] <Andrew Sullivan> that's not a delegation in the root zone
[05:33:30] <Andrew Sullivan> oh, wait, line before
[05:33:31] <Andrew Sullivan> hmm
[05:33:33] <Andrew Sullivan> even worse
[05:33:38] hummmmm
[05:33:42] <Andrew Sullivan> I think we don't have the ability to direct IANA to do that
[05:33:43] <John Border> hummmmm
[05:33:48] <Ralph Droms> hummmmm
[05:34:24] <Terry Manderson> Thankyou for clarifying with the consensus call.
[05:34:34] <wmt> You're welcome.
[05:34:49] <Andrew Sullivan> I am not convinced we are in a position to create that delegation from the root zone, and I think it really _is_ beyond the MoU between the IETF and ICANN
[05:34:57] <Andrew Sullivan> and I think we're going to have a problem therefore.
[05:35:19] <wmt> Well, I hope that's not the case, but at least we will find out.
[05:35:40] <Juliusz Chroboczek> Meetecho, could we have the camera on the speaker?
[05:35:41] <Suzanne> Anyone know where to find the process for asking IANA to make an unsecured delegation?
[05:35:59] <Meetecho> (y)
[05:36:00] <Juliusz Chroboczek> ty
[05:36:19] default local zone required every zone to have a insecure delegation
[05:36:36] <Terry Manderson> Suz: The only process that I am aware of to have IANA to make any delegations is related to .arpa
[05:36:56] <Andrew Sullivan> 4.3. Two particular assigned spaces present policy issues in addition
 to the technical considerations specified by the IETF: the assignment
 of domain names, and the assignment of IP address blocks. These
 policy issues are outside the scope of this MOU.
 Note that (a) assignments of domain names for technical uses (such as
 domain names for inverse DNS lookup), (b) assignments of specialised
 address blocks (such as multicast or anycast blocks), and (c)
 experimental assignments are not considered to be policy issues, and
 shall remain subject to the provisions of this Section 4. (For
 purposes of this MOU, the term "assignments" includes allocations.)
 In the event ICANN adopts a policy that prevents it from complying
 with the provisions of this Section 4 with respect to the assignments
 described in (a) - (c) above, ICANN will notify the IETF, which may
 then exercise its ability to cancel this MOU under Section 2 above.
[05:37:07] <Andrew Sullivan> This is from RFC 2860
[05:37:12] IANA known how to this and can even co-ordinate this with ARIN
[05:37:14] <Andrew Sullivan> we're just not in a position to direct ICANN to do those.
[05:37:22] <Andrew Sullivan> I am prepared to believe IANA knows how to do this
[05:37:33] <Andrew Sullivan> I just don't believe we have the authority to direct it
[05:37:55] Victor Kuarsingh leaves the room: Disconnected: closed
[05:38:06] special names don't have to return NXDOMAIN
[05:38:08] <Andrew Sullivan> If we want those delegations, we have to follow ICANN's procedures for requesting a delegation (or non-delegation)
[05:38:09] <Mikael Abrahamsson2> well, let's kick this document upstairs and let's hope it can be worked out?
[05:38:21] <Andrew Sullivan> There is no upstairs here
[05:38:34] <Andrew Sullivan> There is only someone else's IANA registry
[05:38:37] <Suzanne> @ajs was afraid of that….a history of doing it as a favor doesn't mean they have to, and a single-label name may very well pose different issues to a third-level or lower under in-addr.arpa
[05:38:40] <Andrew Sullivan> and we have to go through their procedures for it
[05:38:48] <Mikael Abrahamsson2> well, AD and up in IETF is "upstairs" for me as participant in this WG.
[05:38:57] <Andrew Sullivan> stuff under arpa is the IAB's problem
[05:39:02] <Andrew Sullivan> and you can get the insecure delegation if you want
[05:39:03] <wmt> I hope we can figure that out, because it is clearly bad to have TLDs leaking all over the place when we could otherwise prevent it. As Pierre said in his talk, if we leave it up to vendors to pick their own value, goodness knows what is going to happen.
[05:39:04] <Andrew Sullivan> that's up to us
[05:39:16] <Andrew Sullivan> Completely agree with that
[05:44:12] <Juliusz Chroboczek> mic: there's no chance to implement Netconf in a web browser.
[05:44:45] <Mikael Abrahamsson2> I don't think it was meant to be done in browser.
[05:45:28] <Mikael Abrahamsson2> do you still want it relayed to mic?
[05:45:37] <Juliusz Chroboczek> Mikael, reading through Ted' s document, it seems clear to me that he intends the interface to be used by a web interface.
[05:45:40] <Juliusz Chroboczek> Nah, forget it.
[05:46:33] <Andrew Sullivan> +1 to DNSSEC in here
[05:46:52] <Mikael Abrahamsson2> juliusz, I think the comment on netconf was in case the administration was done using an app or a m2m interface. But yes, the examples might be more browser specific.
[05:47:28] the hard part is installing the trust anchor in every validator using the homenet
[05:47:38] <Mikael Abrahamsson2> Stuart Cheshire at mic
[05:47:58] danyork leaves the room: Disconnected: closed
[05:48:34] <Peter Koch> like the DNSSEC idea, but don't see a reason not to use "real" names
[05:49:09] <Andrew Sullivan> @Peter: everyone isn't going to spend $n additional per year in order to run their home network
[05:49:13] <wmt> For example, I can only update the config on my Freebox by logging into a webpage (not on the home router itself), making changes, and when I submit something talks to the Freebox from the "cloud" and configures it. Question is whether an app fully in the home would have to be different than that.
[05:49:56] wmt leaves the room
[05:50:08] <Mikael Abrahamsson2> Bob Hinden at mic
[05:50:11] wmt joins the room
[05:50:43] Colin Petrie leaves the room
[05:51:33] <Mikael Abrahamsson2> Michael Richardson at mic
[05:52:08] <Peter Koch> the additional name and DNSSEC transaction by the ISP won't be "no cost" (as in: there is work to be done)
[05:52:42] Colin Petrie joins the room
[05:53:59] danyork joins the room
[05:55:47] <Mikael Abrahamsson2> Harald Alvestrand at mic
[05:55:58] <Mikael Abrahamsson2> David Schninazi at mic
[05:56:29] <Ralph Droms> We have a test case in front of us. Is the name resolution mechanism defined in Ted's doc independent of the name suffix for homent-local names? Is the text in draft-ietf-homenet-dot compatible with the mechanism described in Ted's document?
[05:57:01] Balazs Varga joins the room
[05:57:19] step 3 is only needed if you want to publish the zone contents
[05:57:29] <Mikael Abrahamsson2> James Woodyatt at mic
[05:57:31] <Andrew Sullivan> What I am hearing here is an argument that all those botnet devices that are easy are not also failures :)
[05:58:35] <Juliusz Chroboczek> It's javascript served from the router's web interface.
[06:00:14] danyork leaves the room: Disconnected: closed
[06:00:15] Colin Petrie leaves the room
[06:00:27] Audric Schiltknecht leaves the room: offline
[06:00:27] <Andrew Sullivan> Is this Townsley's Law, a little like Godwin's?
[06:00:34] <wmt> :-)
[06:00:35] Colin Petrie joins the room
[06:01:03] Ray Atarashi leaves the room
[06:01:05] Terry Manderson leaves the room
[06:01:21] <Dave Thaler> FYI, interesting fact: ".arpa" does not (since 2000) stand for the name of a US defense agency. it stands for "Address and Routing Parameter Area". This is from RFC 3172 / BCP 52, and this expansion has been at the top of the wikipedia page for many years which wiki page is the #1 hit when I did a web search on "arpa". Doesn't change consensus, but good to know.
[06:01:24] Yingzhen Qu leaves the room
[06:01:29] marco@nl leaves the room: Replaced by new connection
[06:01:43] Peter Koch leaves the room
[06:01:49] Andrew Sullivan leaves the room
[06:01:50] wmt leaves the room
[06:01:53] Simon Pietro Romano leaves the room
[06:01:54] marco@nl joins the room
[06:02:03] Steve Olshansky leaves the room
[06:02:07] Jun Hu leaves the room
[06:02:07] Juliusz Chroboczek leaves the room
[06:02:07] John Border leaves the room
[06:02:07] Balazs Varga leaves the room
[06:02:07] Ralph Droms leaves the room
[06:02:07] Mark Andrews leaves the room
[06:02:07] Wentao Shang leaves the room
[06:02:08] Mikael Abrahamsson2 leaves the room
[06:02:10] Moritz leaves the room
[06:02:17] Suzanne leaves the room
[06:02:30] Simon Romano leaves the room
[06:02:30] Ralph Droms joins the room
[06:03:34] Meetecho leaves the room
[06:04:55] Ralph Droms leaves the room
[06:06:05] marco@nl leaves the room
[06:08:52] danyork joins the room
[06:09:57] Colin Petrie leaves the room
[06:17:17] Dave Thaler leaves the room
[06:19:07] mcr leaves the room
[06:21:53] Dave Thaler joins the room
[06:22:10] marco@nl joins the room
[06:22:18] Dave Thaler leaves the room
[06:23:00] marco@nl leaves the room: Replaced by new connection
[06:23:16] marco@nl joins the room
[06:24:07] marco@nl leaves the room: Replaced by new connection
[06:24:23] Moritz joins the room
[06:24:25] marco@nl joins the room
[06:24:29] mcr joins the room
[06:24:58] Moritz leaves the room
[06:26:13] Terry Manderson joins the room
[06:26:13] marco@nl leaves the room
[06:32:35] Colin Petrie joins the room
[06:46:01] Suzanne joins the room
[06:48:26] Terry Manderson leaves the room
[06:55:11] mcr leaves the room: Disconnected: No route to host
[07:04:18] Suzanne leaves the room
[07:14:49] danyork leaves the room
[07:19:35] Colin Petrie leaves the room
[07:44:05] Suzanne joins the room
[07:57:16] mcr joins the room
[08:01:24] wmt joins the room
[08:02:44] <wmt> But Google points me to this: https://en.wikipedia.org/wiki/ARPA
[08:02:55] <wmt> Which lists • ARPA, the Advanced Research Projects Agency, renamed the Defense Advanced Research Projects Agency (DARPA <https://en.wikipedia.org/wiki/DARPA>) in 1972
[08:02:59] <wmt> on the first bullet
[08:03:17] <wmt> But good to know about Address and Routing Parameter - I didn't know about that! :-)
[08:04:41] wmt leaves the room
[08:11:34] Peter Koch joins the room
[08:16:06] Peter Koch leaves the room
[08:23:52] mcr leaves the room
[08:48:18] Suzanne leaves the room
[11:40:27] Colin Petrie joins the room
[12:13:57] Colin Petrie leaves the room
[12:16:01] Colin Petrie joins the room
[12:30:15] Colin Petrie leaves the room
[12:31:30] Colin Petrie joins the room
[15:39:48] Colin Petrie leaves the room
[17:01:11] Jason Weil joins the room
[17:03:33] Jason Weil leaves the room