[00:00:41] orange leaves the room [00:03:17] Glenn Parsons joins the room [00:04:18] bortzmeyer joins the room [00:06:14] spencerdawkins leaves the room [00:07:30] spencerdawkins joins the room [00:07:57] Lars joins the room [00:08:08] James Galvin joins the room [00:08:54] Joe Hildebrand joins the room [00:09:00] resnick joins the room [00:09:23] sm joins the room [00:09:33] scott.mansfield joins the room [00:10:02] Here is the current list of topics I have so far: 1) Dan Wing: short announcement of GROBJ BOF (10 mins) 2) Dan Wing : Building IPv6 Applications which Access IPv4 Servers (draft-wing-v6ops-v6app-v4server-01.txt) - 15 mins 3) Peter Saint-Andre : Apps Review Team - 5 mins 4) Peter Saint-Andre : TLS server identity checks in application protocols (draft-saintandre-tls-server-id-check) - 20 mins [with discussions] 5) Magnus Westerlund : IANA ports and service names registry - 20 mins [with discussions] 6) John C Klensin

: FTP commands and extensions registry - 15 mins 7) Spencer Dawkins : Subscription/Notification for Lightweight Directory Access Protocol (LDAP) (draft-dawkins-ldapext-subnot-01.txt) - 20 mins 45 mins for the remaining of the session: 8) Short presentations about other Apps BOFs 9) Open mic [00:10:23] has anyone notified the secretariat of th eprojector issue yet? [00:10:26] shinta joins the room [00:11:03] coopdanger joins the room [00:11:19] fujiwara joins the room [00:11:23] rlbob joins the room [00:11:27] stpeter joins the room [00:11:32] Simon Perreault joins the room [00:12:10] Randall Gellens joins the room [00:12:11] Dan Wing to talk about Grobj [00:12:15] GROBJ [00:12:16] pronounced "grow"? [00:12:35] Vincent_Levigneron joins the room [00:12:41] (someone please fined a link to Dan's slides, and post here) [00:12:42] so i guess speakers should also check in via rfid? because dan is not lisa [00:12:43] "growb", I think. [00:12:54] there's a swiper on the char table. [00:12:55] Barry Leiba: thanks [00:12:57] chair [00:13:29] Dan: Solving redirect problems is hard, but it's even harder to do per-protocol [00:13:41] GROBJ BOF is Wed afternoon 1510 [00:13:52] GRO = Generic Referral Object [00:14:01] ray joins the room [00:14:23] GRO's allow referral between protocols, such as SIP and XMPP [00:14:49] BTW the log is at http://www.ietf.org/jabber/logs/apparea/2009-11-09.txt [00:15:05] Next topic: Building v6 Apps which access v4 servers [00:15:18] By "object", I assume we're talking about "interesting data structure"? [00:15:30] Atarashi Yoshifumi joins the room [00:15:38] BEHAVE scenario 1, scenario 5. [00:15:42] Jelte joins the room [00:15:44] orange joins the room [00:15:50] (I hate this silly overuse of "object" in this theoretical way.) [00:16:03] Lisa joins the room [00:16:10] alexey.melnikov joins the room [00:16:17] John C Klensin joins the room [00:16:20] Most of the time, the 6-4 problems don't exist, in particular, when you're using DNS. [00:16:28] Ohayou [00:16:45] おはよう。 [00:16:54] Gist of problem: address literals are bad. [00:17:04] marcos@de joins the room [00:17:42] App specific: dual-stack proxy [00:18:15] I will use the jabber room to announce next presenters [00:18:34] bhoeneis joins the room [00:18:46] General: make code changes to the IPv6 side to encode v4 in v6 [00:18:57] Ted Hardie: does TLS break this? Dan: No [00:20:11] ray leaves the room: Replaced by new connection [00:20:12] Dave Crocker: clarifies that problem is that v6-only host gets a v4 address [00:20:27] ray joins the room [00:20:29] [Next presentation: Spencer Dawkins : Subscription/Notification for Lightweight Directory Access Protocol (LDAP) (draft-dawkins-ldapext-subnot-01.txt) - 20 mins] [00:20:42] Dan proposes IPv6 apps get taught what to do with v4 literals [00:21:07] Douglas Otis: port stickiness? [00:21:21] Dan: no, that won't work any more [00:21:35] Joe Touch joins the room [00:22:28] magnus joins the room [00:22:56] (backing up: Dave Crocker also asked if Dan's stuff had been analyzed for effort vs. value, Dan said "not yet, but I welcome that") [00:23:47] ray leaves the room: Replaced by new connection [00:24:02] Spencer talking about LDAP pub/sub [00:24:17] gigix73 joins the room [00:24:23] 3GPP working on "Unified Subscriber Data" [00:24:30] gregdavies joins the room [00:24:32] Alternative is XML/SOAP [00:24:35] Agressive timelines [00:24:43] [The following presentation is: Peter Saint-Andre : Apps Review Team - 5 mins] [00:25:07] there has been feedback on the draft [00:25:13] Are there really no slides online? [00:25:25] why not use RFC 4533 or 3928? [00:25:30] jlcJohn: not sure [00:25:38] (spencer is talking too fast. help scribing?) [00:25:38] gregdavies leaves the room [00:26:05] Key difference: 3rd party subscriptions [00:26:30] gregdavies joins the room [00:26:31] Time travel needed to get done by Feb 2009 [00:27:09] just posted my slides at http://www.saint-andre.com/tmp/IETF76-apparea-appsreview.pdf and http://www.saint-andre.com/tmp/IETF76-apparea-serverid.pdf [00:27:14] (FWIW) [00:27:20] (2010 was intended) [00:27:39] Question: is this an appropriate LDAP extension? [00:27:48] hirotaka.sato joins the room [00:27:55] Alexey: We need to nail down requirements [00:28:01] gregdavies leaves the room [00:28:23] gregdavies joins the room [00:28:27] Question remaining: app-level acks of notifications [00:28:27] ohnolab.t-takashima joins the room [00:28:49] Alexey: LDAPext is the right place for discussions [00:29:05] draft-dawkins-ldapext-subnot-01 [00:29:14] draft-ietf-ldapext-psearch-03 [00:29:20] gregdavies leaves the room [00:30:06] http://tools.ietf.org/html/draft-dawkins-ldapext-subnot-01 [00:30:20] http://tools.ietf.org/html/draft-ietf-ldapext-psearch-03 [00:31:15] Stpeter talking about Apps-Review team [00:31:31] Apps area folk who review I-Ds [00:31:31] ray joins the room [00:31:41] Focus: Apps/RAI/Security [00:31:48] http://www.apps.ietf.org [00:32:08] Ping Peter if you're willing to help, or want review [00:32:37] Next topic: Server ID Checking [00:32:59] http://tools.ietf.org/html/draft-saintandre-tls-server-id-check-02 [00:33:36] Lots of protocols use X.509, each has to define it's own identity checking rules, so they're all slightly different [00:33:48] sal joins the room [00:33:48] I-D show similarities/differences [00:34:28] Scope: App servers as TLS servers [00:34:47] Joe Touch is next (talking abour port and service name registry) [00:34:54] Scope: just server identity [00:35:23] what kinds of identity? [00:35:53] doug.mtview joins the room [00:35:59] process: server presents multiple identities (i.e. subjectAltNames) [00:36:19] reference identity: ident that client expects to find [00:36:35] various rules for checking different idents [00:36:54] ko joins the room [00:37:00] domain names, IDNAs, fall back to CN [00:37:07] ko leaves the room [00:37:18] open issues: [00:37:22] - client checking? [00:37:30] - fingerprint matching? [00:37:35] non-X.509 certs [00:37:47] - beyond TLS (ipsec, dtls, etc) [00:37:57] - wildcard identities [00:38:22] rababy joins the room [00:38:26] ko joins the room [00:38:47] Pete Resnick: I don't get it. checking X.509 is seeing if what i typed matches the cert. [00:38:59] yao joins the room [00:39:00] magnus leaves the room [00:39:32] Pete cares more about what the CA asserts, than the cert generator [00:40:05] gregdavies joins the room [00:40:16] Pete: is this new rules for how CA generate certs? [00:40:18] coopdanger leaves the room [00:40:37] Pete: so we're making up rules for certificate signers? [00:40:48] StPeter: we're making up rules for clients [00:41:16] RLBob: it may be that what Pete Resnick thinks is obvious is not obvious, so we should write it down [00:41:35] RLBob: RFC 3709. [00:41:55] Logos are good [00:42:04] (aside: NOT accessible) [00:42:25] (can't get to mic since nobody else is scribing) [00:42:54] Dave Crocker: i'm going to ask a question even though i haven't read the draft [00:43:07] spencerdawkins leaves the room [00:43:18] DC: if this is a component, and nobody's going to use it, it's a waste of time [00:43:55] DC: is there a context specified in the I-D? [00:44:32] stpeter: so we don't have to define it every time, since it's hard to get right [00:44:53] Vincent_Levigneron leaves the room: Replaced by new connection [00:44:54] Vincent_Levigneron joins the room [00:45:04] DC: it's hard to get right. need to make sure we don't mess it up [00:45:30] Stpeter: after you read the draft, give feedback [00:46:08] RLBob: no significant changes to existing clients, since this more-or-less works today, it's just there are a lot of non-obvious edge cases [00:46:13] mellon joins the room [00:46:23] RLBob: this is a cleanup and reconciliation effort [00:46:50] Pete Resnick: why are clients not ok to choose any of these? [00:49:17] JoeH: it's per-protocol, not per-signer [00:49:29] Pete: If I understand your question, you're saying this is only half the issue (client checking) not the other half (CA signing)? [00:49:34] Steve Hanna: there are lots of ways, this I-D provides good guidance [00:49:41] Vincent_Levigneron leaves the room: Replaced by new connection [00:49:43] @mike: Steve Hanna - Juniper [00:49:43] Vincent_Levigneron joins the room [00:50:05] (in practice, CA's sign anything that looks kind of reasonable, or sign no certs with any "cool" extensions) [00:50:14] @Randy: Right, and I think the other half is the more important half. (Actually, necessary before you can even think about what the client is going to do.) [00:50:41] Lots of head-nodding in the room, since there are 3 different places to put domain names [00:51:28] (Pete: goal would be to get CA's to sign certs that have a well-understood, widely-implemented extension of at least one kind) [00:51:45] For the ports presentation: http://tools.ietf.org/html/draft-ietf-tsvwg-iana-ports-03 [00:52:03] Joe Touch leaves the room [00:52:37] ray leaves the room: Replaced by new connection [00:52:50] Carsten Bormann joins the room [00:53:28] Joe Touch: IANA updates [00:53:28] Ed J. joins the room [00:53:33] ray joins the room [00:53:48] Goal: update doc procedures [00:53:55] Goal: unify registries [00:54:06] updates exisiting RFCs [00:54:18] Not: guidance to app designers [00:54:25] Not: Specification of port/name spaces [00:55:16] Current registries, in different places [00:55:33] magnus joins the room [00:56:01] Q: I already own a name in /etc/services, do I have to register again in SRV? [00:56:02] yes. [00:56:10] calvin joins the room [00:56:25] Different managers: IANA, individuals, etc. [00:56:45] Goal: One unified table for all: transports, SRV [00:56:49] indexed by name [00:57:34] ray leaves the room: Replaced by new connection [00:57:44] allows IANA to specify format [00:57:51] Different syntax changes [00:58:04] US ASCII a-z, 0-9, hyphen, case insensitive [00:58:10] no hyphen start/end [00:58:13] not all numbers [00:58:15] 1-15 chars [00:58:32] whois++ -> whoisplusplus [00:58:47] new name replaces "specials" with '-' [00:59:11] (aside: "special" implies a value judgement, which is likely unintended) [00:59:48] Leslie Daigle: whoispp is canonical [01:00:03] Leslie: track down old protocols to find out what they would prefer [01:00:06] (rather than -) [01:00:20] JoeT: just adding these as aliases [01:00:24] gregdavies leaves the room: Replaced by new connection [01:00:31] gregdavies joins the room [01:00:46] Stuart Cheshire: 10k names, almost all human mnemonics [01:00:57] with SRV, they're being used programatically. [01:01:31] we'd like to know if anyone is using these strings in (e.g.) getservbyname [01:02:05] in reality, most apps use the well-known port directly [01:02:20] JoeT: some people make changes to their /etc/services: [01:02:37] Stuart: do you have an example of a browser that does that? [01:02:43] JoeT: address translator [01:02:50] Stuart: does that work? [01:03:00] masinter joins the room [01:03:07] JoeT: probably doesn't but should [01:03:25] Dave Crocker: is this just for SRV, or all underscore names? [01:03:39] Joe Hildebrand: "you gotta SRV somebody" :) [01:03:59] (SRV != Stevie Ray Vaughn in this crowd) [01:04:25] JoeT: we'll check on that and get back to you [01:04:28] Joe Hildebrand: I was thinking Robert Zimmerman, myself :) [01:04:54] (looking for agenda) [01:05:20] StuartC: we're defining the database of names for things other than SRV [01:05:22] (that was Olafur Gudmundsson at the mic) [01:05:32] there are 90ish id's that are 15 characters [01:05:36] people only on audio don't see the rfid-based name... [01:05:49] Jelte: sorry, trying to do the best i can. [01:06:01] please ask the RFID folks to implement an XMPP interface... [01:06:20] oh i'm not commenting on you, just repeating here that people should still *say* their name in the mic :) [01:06:31] Jelte: thx. not obvious IRL [01:06:53] Stuart talks about Bluetooth interop [01:07:02] (when not participating i often only listen to the audio logs later for instance) [01:07:25] Olafur: we're sacrificing flexibility? [01:07:34] StuartC: no: 15char is a lot. [01:07:40] gregdavies leaves the room [01:08:11] JoeT: 15 is big enough, 14 would require revisions to existing [01:08:51] previously: you got TCP and UDP, even if you didn't want both [01:09:08] Colman Ho joins the room [01:09:19] sometimes you'd get 30ports, sometimes not [01:09:49] IANA couldn't assign name w/o number [01:10:01] new: one port per service [01:10:13] new: one port for all versions. design your protocols. [01:10:21] new: includes secure!!! [01:10:29] (note: WSS needs start-tls) [01:10:53] new: one port for all device types [01:11:11] now: only one transport by default [01:11:24] IANA may recover/reuse/transfer [01:11:50] different names for UDP/TCP [01:12:12] sample: banana and banana-disc [01:12:46] First-come-first-serve without a port number, otherwise expert review [01:13:08] ~4k/year [01:13:20] constant over last 10y [01:13:42] new procedure: re-register/re-use/revoke/transfer [01:13:51] IANA is not going to yank ports [01:14:37] Owner is IETF for RFC-specified [01:15:08] 1021 and 1022 are experimental for non TCP/UDP from now on [01:15:28] Jelte leaves the room [01:15:32] DCCP edge cases [01:15:38] gregdavies joins the room [01:15:49] XML table [01:15:54] gregdavies leaves the room [01:15:57] did he say 4k a year? I think we would not run out in 70-80 years at the rate of 4k per year [01:16:12] gregdavies joins the room [01:16:13] (note: please ask app review to look at the XML) [01:16:39] Kenji Rikitake at the mic [01:17:03] Kenji Rikitake: clarify. many existing protocols actually *use* both TCP and UDP [01:17:14] John is next (talking about FTP extension registry) [01:18:08] JoeT: when you ask for TCP, you get TCP, and UDP is reserved [01:18:14] unless we run out of ports [01:18:29] person who registered TCP gets dibs on UDP [01:18:38] (associated UDP) [01:19:08] IANA is not planning on revoking/reserving old associated ports [01:19:17] but it might be interesting to make notes in the table [01:19:32] Olafur: questions wisdom of merging the registries [01:19:50] Olafur: has a draft to register just SRVs [01:20:38] Lisa: TSV? [01:20:47] Olafur: no, DNS guys don't know about app layer [01:21:15] Joe Hildebrand at the mic: I think apparea is the only area where this is used [01:21:24] JH: I think this is in-scope for apparea [01:21:37] Lisa: ought we take this to the list? [01:21:42] JH/JT: yes [01:21:51] "the list" = apps-discuss? [01:22:11] Lisa: please specify which list out loud [01:22:15] Lars Eggert at the mic [01:22:32] Lars Eggert: we've had some discussion, it got heated on the list [01:22:49] (I think apps-discuss is the list, because that's the list which has been cc'd to date) [01:23:35] JoeT: there being one place is good, so you don't have to worry about which is which [01:24:08] next presenter: John Klensin on "FTP commands and extensions registry" [01:24:33] shinta leaves the room [01:24:43] shinta joins the room [01:26:50] calvin leaves the room [01:27:07] Carsten Bormann at the mic on the 6lowapp BoF [01:27:14] gregdavies leaves the room [01:27:56] marcos@de leaves the room [01:28:07] CB: we have 6lowpan and roll WGs already, now it's time to talk about applications [01:28:13] marcos@de joins the room [01:28:14] Larry Masinter on IRI BoF [01:28:15] marcos@de leaves the room [01:28:37] LM: there is an RFC [01:29:12] LM: group of browser makers are also discussing this topic (HTML5) [01:29:42] HTML5 originally contained its own description, different processing rules, etc. -- describing what browsers currently do (which is not what's in the RFC) [01:30:51] LM: many different committees looking into this [01:31:07] Rich Woundy on the DECADE BoF [01:31:28] gregdavies joins the room [01:31:49] DECADE: storage in the network [01:31:57] masinter leaves the room: Replaced by new connection [01:32:15] problem: P2P is large % of network traffic [01:32:28] bhoeneis leaves the room: Replaced by new connection [01:32:43] existing p2p caches are complex/limited [01:33:29] stpeter wonders idly how DECADE is related to ALTO but might be operating on too little sleep [01:34:45] Complementary I think. [01:34:53] move data between network-based storage/caches [01:34:57] my storage device or yours? [01:35:45] hohno.46466 joins the room [01:35:45] (seems like closing the barn door too late, to me, since non-discoverability is now a feature for many P2P networks) [01:36:02] Cattle! Yah! [01:36:08] lmm joins the room [01:36:28] Ed Juskevicius: how do apps/people compete with each other for space? [01:36:41] A: good topic for bof [01:36:54] markus.isomaki joins the room [01:37:03] magnus leaves the room [01:37:25] Stuart Cheshire: is this just allowing ISPs to minimize the amount of traffic they're transfering? [01:37:33] transferring [01:37:39] well, cache happens, no? [01:37:45] A: cost for different links are different [01:37:55] (so it's a routing problem!) [01:38:22] JoeH on HyBi [01:38:37] HyBi = Hypertext Bidirectional [01:39:02] now morphing more into discussions of bidirectional communication between browser and server [01:39:35] Alexey: per-area office hours. [01:39:51] is a good idea? [01:39:54] shrugs in the room [01:40:25] calvin joins the room [01:40:35] Now: open mic night [01:40:54] communication is good, so no objection here to app area office hours [01:41:10] Barry Leiba: VWrap, please participate [01:41:13] Related to VWRAP: If you want to find me in Second LIfe, I'm Rue Linden! [01:41:19] virtual worlds [01:41:44] mellon leaves the room [01:41:58] DaveCrocker: need protocol help for VWRAP [01:42:29] doug.mtview leaves the room [01:42:42] Lisa: HTTP usage preso. anyone interested in a preview [01:42:57] Lisa leaves the room [01:43:53] room: sure [01:44:24] everyone says: just use HTTP [01:44:29] reality: HARD [01:45:28] marcos@de joins the room [01:46:26] intermediaries don't follow spec [01:46:59] link to the slides? [01:47:10] sm leaves the room [01:48:04] Many pitfalls [01:48:31] BCP 56 says HTTP bad as substrate [01:48:39] but now we're wiser [01:50:13] REST [01:50:47] stateful: bad [01:51:34] (note: much of this should go into draft-loreto-design-space-bidirectional-00 [01:51:38] ) [01:52:50] calvin leaves the room [01:53:04] compartmentalization = c18n? [01:54:43] Vincent_Levigneron leaves the room [01:54:49] counterdemonstration, counterexpostulation [01:55:06] Vincent Levigneron joins the room [01:55:28] REST is easy to deploy [01:55:54] orange leaves the room [01:56:43] REST drawbacks: [01:56:48] uniform interface not optimized [01:56:57] not for notifications [01:57:06] (HyBi plug) [01:57:25] no caching? REST adds overhead [01:57:30] resnick boos hybi [01:57:56] "The security properties of HTTP aren't awesome." - Lisa [01:58:08] candidate for understatement of the week [01:59:51] Barry Leiba leaves the room [02:00:15] Barry Leiba joins the room [02:00:16] RPC is different than REST [02:00:40] Alexey: Step out into the "hallway"? (hallway@jabber.ietf.org) [02:00:47] RPC doesn't benefit from caching [02:02:43] Lars leaves the room [02:02:45] Lars joins the room [02:03:18] (wants normative reference to RFC 2324) [02:03:33] ATOM: rfc 5023 [02:05:29] zdshelby joins the room [02:06:12] should we update BCP 56 then? [02:06:27] lmm: yes, she said that earlier [02:06:56] (sorry, i stopped re-typing the slides, just hitting the interesting highlights) [02:07:07] I rather understand that she criticized BCDP 56. Saying that we should go beyond it. [02:07:42] Ew. she said CORBA [02:08:01] Didn't she say that the slides were somewhere available? [02:09:37] marcos@de: yes, but i didn't get a link [02:10:44] sal leaves the room [02:12:44] JoeH: please get your bits into draft-loreto [02:12:55] Larry: please update BCP 56 [02:13:02] Lisa: working on a draft [02:13:54] Larry: Video over HTTP? [02:14:00] caching vs. overhead [02:14:20] Alexey: (snarkily?) DECADE [02:14:53] lisa: good points. httpbis in 7 chunks, so reuse might be possible [02:15:21] hohno.46466 leaves the room [02:15:30] Atarashi Yoshifumi leaves the room [02:15:47] Simon Perreault leaves the room [02:15:51] gigix73 leaves the room [02:15:57] bortzmeyer leaves the room [02:16:01] ko leaves the room [02:16:04] zdshelby leaves the room [02:16:17] hirotaka.sato leaves the room [02:16:59] Randall Gellens leaves the room [02:17:19] Carsten Bormann leaves the room [02:17:31] Joe Hildebrand leaves the room: Disconnected. [02:17:32] John C Klensin leaves the room [02:17:57] yone leaves the room [02:18:01] jlcJohn leaves the room [02:18:21] Colman Ho leaves the room [02:18:33] rlbob leaves the room [02:18:35] Ed J. leaves the room [02:18:39] resnick leaves the room [02:20:40] markus.isomaki leaves the room [02:20:47] marcos@de leaves the room [02:21:00] tonyhansen leaves the room [02:21:15] ohnolab.t-takashima leaves the room [02:21:19] Lars leaves the room [02:22:23] thanks to Joe for scribing [02:22:24] stpeter leaves the room [02:24:08] shinta leaves the room [02:24:46] James Galvin leaves the room [02:29:05] alexey.melnikov leaves the room: Computer went to sleep [02:30:04] rababy leaves the room: Computer went to sleep [02:33:27] sftcd leaves the room [02:34:29] gregdavies leaves the room [02:34:38] Barry Leiba leaves the room [02:35:20] Glenn Parsons leaves the room [02:36:28] ꈲ joins the room [02:37:11] gregdavies joins the room [02:37:12] scott.mansfield leaves the room [02:38:14] gregdavies leaves the room [02:38:17] Greg Davies joins the room [02:41:23] yao leaves the room [02:42:25] Greg Davies leaves the room [02:44:13] Vincent Levigneron leaves the room [02:47:28] lmm leaves the room [02:55:58] sal joins the room [03:00:40] sal leaves the room [03:47:49] Joe Hildebrand joins the room [03:48:53] redaka joins the room [03:49:08] redaka leaves the room [03:54:50] Vincent Levigneron joins the room [03:55:54] Vincent Levigneron leaves the room [03:57:35] Joe Hildebrand leaves the room [03:59:47] calvin joins the room [03:59:52] calvin leaves the room [04:01:04] kanda joins the room [04:01:12] kanda leaves the room [04:02:32] Lars joins the room [04:02:54] Lars leaves the room [04:07:27] alexey.melnikov joins the room [04:11:12] Glenn Parsons joins the room [04:11:17] Glenn Parsons leaves the room [04:16:12] bortzmeyer joins the room [04:30:25] fujiwara leaves the room [04:44:31] bortzmeyer leaves the room [05:47:55] alexey.melnikov leaves the room: Computer went to sleep [06:07:25] ꈲ leaves the room [06:23:38] ꈲ joins the room [07:04:56] ꈲ leaves the room [07:57:11] alexey.melnikov joins the room [08:16:23] alexey.melnikov leaves the room: Computer went to sleep [14:04:44] tonyhansen joins the room [14:33:23] tonyhansen leaves the room [22:52:32] alexey.melnikov joins the room