User:Zero Linden/Office Hours/2009 mar 03

From Second Life Wiki
Jump to: navigation, search
  • [12:59] Infinity Linden: i'm gonna be embarassed if i just sat on someone
  • [12:59] Torrid Luna: /mew^^
  • [12:59] SharedRealm Engineer: hi
  • [13:00] Torrid Luna: hahaha^^
  • http://slurl.com/secondlife/Grasmere/163/111/27 (Starts now)
  • [13:00] Torrid Luna:  ;-))
  • [13:00] Rex Cronon: don't drop the cat:)
  • [13:00] Torrid Luna: I do my best^^
  • [13:00] Morgaine Dinova: Haha
  • [13:01] Infinity Linden: put your hands over your head and step away from the cat!
  • [13:01] Torrid Luna: Okies.^^
  • [13:01] Jonit Ivory: good day lovely people :)
  • [13:01] Morgaine Dinova: Hiya Jonit
  • [13:01] Rex Cronon: hi
  • [13:01] Saijanai Kuhn: https://wiki.secondlife.com/wiki/AW_Groupies/Chat_Logs/AWGroupies-2009-03-03 transcript from this morning
  • [13:02] Jonit Ivory: for whom unrezzed I may sit on, I apologise
  • [13:02] Infinity Linden: for those about to sit, we salute you?
  • [13:02] Jonit Ivory: as you wish
  • [13:02] Jonit Ivory: I found a seat
  • [13:02] Jonit Ivory: it's a start
  • [13:04] Rex Cronon: testing...
  • [13:04] Asterion Coen: hello folks
  • [13:04] Asterion Coen: oh, a bear :)
  • http://slurl.com/secondlife/Grasmere/163/111/27 (Started 5 minutes ago)
  • [13:05] Asterion Coen: so, now we got a kittie, a bear and a dog ;)
  • [13:05] Asterion Coen: hmm i will build some honey, i promize :)
  • [13:05] Jonit Ivory: it's a 32 bit dog
  • [13:05] Rex Cronon: thanks jonit. i don't need a tp. i am right here:)
  • [13:06] Jonit Ivory: bu**ger wrong person
  • [13:06] Rex Cronon: soon we will have an entire zoo:)
  • [13:06] Morgaine Dinova: Hi Zero
  • [13:06] Asterion Coen: heje
  • [13:06] Zero Linden: hello all
  • [13:06] Zha Ewry: Good afternoon
  • [13:06] SharedRealm Engineer: hi
  • [13:06] Jonit Ivory: ciao Zero
  • [13:06] Mordack Byron: tips hat
  • [13:06] Saijanai Kuhn: https://wiki.secondlife.com/wiki/AW_Groupies/Chat_Logs/AWGroupies-2009-03-03 chat log from this morning Zero
  • [13:06] Asterion Coen: hello zero :)
  • [13:06] Harleen Gretzky: Hi Zero
  • [13:07] Widget Whiteberry: mutes the cat
  • [13:07] Rex Cronon: hello zero
  • [13:07] Zha Ewry: Wyn, you are sitting in Infinitty
  • [13:07] Latha Serevi: Zero (or anyone), any objection if I mirror this meeting into IRC, just for a bit of gratuitous interop?
  • [13:07] Wyn Galbraith: That would be nice
  • [13:07] Wyn Galbraith: I don't know where I'm sitting
  • [13:07] Wyn Galbraith: waiting to rez
  • [13:07] Jonit Ivory: +100 to Wyn
  • [13:07] Morgaine Dinova: Thank you kindly, Ast :-)
  • [13:08] Asterion Coen:  ;)
  • [13:08] Saijanai Kuhn: what's that IRC channel again, Latha?
  • [13:08] Infinity Linden: lol
  • [13:08] Zero Linden: Welcome all
  • [13:08] Zero Linden: Mirroring to IRC is just fine
  • [13:08] Zero Linden: this office hours, like all of mine, are on-the-record - speek freely, speak publically
  • [13:09] Zero Linden: I first will appologize for being remiss in either setting up my office hours page to reflect the new format
  • [13:09] Asterion Coen: /so i will try to dont tell jokes
  • [13:09] Latha Serevi: brb
  • [13:09] Zero Linden: and for failing to put out a call for agenda items in advance
  • [13:09] Zero Linden: I will get it right next month (thanks to my trusty GTD app....)
  • [13:09] Couch sit: 2(male model setup) RED: Asterion Coen, say '/1 Hide' to hide me, or '/1 Show' to make me show. Or just right-click and sit on me to use me.
  • [13:09] Zha Ewry: as long as you don't require the in IETF <fark> annoying </fark> format
  • [13:09] Zero Linden: (which I have JUST created a to do item in...)
  • [13:10] Saijanai Kuhn: Latha what's the irc mirror channel?
  • [13:10] Zero Linden: What - listening here isn't enough - you want it scrolled on more than one screen?
  • [13:10] Zha Ewry: spent slightly too long recently, bashing something simple into a form the IETF would eat
  • [13:10] Saijanai Kuhn: I like talking to myself...
  • [13:11] Infinity Linden: as long as i can setup a camera of me typing in this meeting and stream it as the parcel media
  • [13:11] Asterion Coen: what irc network about ?
  • [13:11] Latha Serevi: In caswe folks want to attend but hate SL, they can go to irc://irc.quickfox.net/mmox
  • [13:11] Zero Linden: okay
  • [13:11] Zero Linden: it is now 1:11
  • [13:11] Saijanai Kuhn: mirrors this local chat to mmox@irc.quick
  • [13:11] Saijanai Kuhn: etc
  • [13:11] Infinity Linden: ohoh. you mean jon watte?
  • [13:11] Jonit Ivory: that's just narcissism Infinity
  • [13:12] Asterion Coen: 6667 ?
  • [13:12] Latha Serevi: Jon probably doesn't know about our mirroring attempts; we haven't said anything on the mailing list yet.
  • [13:12] Infinity Linden: it's telling that a standard IRC port is the number of the beast times ten plus seven
  • [13:13] Mordack Byron: forget about IRC, the lag here in-world is nicer cause more challenging
  • [13:13] Zero Linden: well - in any event, this isn't an MMOX meeting, it is an AWG meeting
  • [13:13] Wyn Galbraith: LOL
  • [13:13] Zero Linden: though the overlap is notable
  • [13:13] Zha Ewry: ponders "Slightly more than ten times beastly then merely beastly"
  • [13:13] Wyn Galbraith: I think it cancels out Zha
  • [13:14] person: hello: folks !
  • [13:14] Teravus Ousley: thinks it's the number of Rex
  • [13:14] Zero Linden: Now - Rather than the previous rambly discussions we've had - as enjoyable and fruitful as they have been -- no, serously - I really loved them!
  • [13:14] Zero Linden: we are going to be a bit more formal here
  • [13:14] Morgaine Dinova: Yout can't cancel out Zha
  • [13:14] Zero Linden: So - I'd like agenda items, and I'd like them to refer to written docs for discussion if at all possible
  • [13:14] Zero Linden: Anyone?
  • [13:15] Jonit Ivory: Zero acan you statr with your curent focus?
  • [13:15] Jonit Ivory: what do you want to discuss hee?
  • [13:15] Jonit Ivory: here
  • [13:15] Zero Linden: well - there are several OGP drafts in existance
  • [13:15] Teravus Ousley: someone suggested back at the AWG meeting to bring up the drafts... for discussion.
  • [13:16] Zero Linden: and as many of you know, we got them working a bit this Summer and early Fall
  • [13:16] Zero Linden: Now, there are many areas not covered by those drafts that I think LL would like to address with new work this year
  • [13:17] Zero Linden: and there are details of those drafts that we'd like to nail down, smooth out, etc... as well
  • [13:17] Zero Linden: At least one of those, LLSD has been submitted as an Internet-Draft
  • [13:17] Teravus Ousley: the XML serialization of LLSD changes suggested by Morgaine?: [1]
  • [13:17] Teravus Ousley: err.. no wait.. that was Catherine Pfeffer
  • [13:18] Latha Serevi: Agenda item: clearly this month's flurry of activity related to the MMOX BOF in 3 weeks, is where the action is. So perhaps "how OGP fits into a broader MMOX" as a topic?
  • [13:18] Morgaine Dinova: Not me
  • [13:18] Teravus Ousley: She 'attached' them to the e-mail.. so.. I copy and pasted to pastebin.com
  • [13:18] Infinity Linden: well.. to the degree it would make something that was not compatible with LL, OpenSim or PyOGP...
  • [13:18] Infinity Linden: i don't know that we're planning on including them in our draft
  • [13:19] Latha Serevi: (personally, I have no interest in LLSD, and would prefer a broader and less nuts-and-bolts-of-this-particular-low-level-internet-draft.)
  • [13:19] Latha Serevi: *discussion
  • [13:19] Infinity Linden: she is, of course, welcome to define another technique for transporting application layer data
  • [13:19] Zero Linden: Latha - then please, by all means write one!
  • [13:19] Teravus Ousley: I think there is a very clear cry about where LLSD lacks..
  • [13:19] Infinity Linden: but she would have to come up with a couple of implementations that implement the draft to get it past draft stage
  • [13:19] Teravus Ousley: .. and, several people have suggested changes :D
  • [13:20] Zero Linden: I won't be using this forum for the kind of open discussion that is "how X fits into a broader MMOX"
  • [13:20] Infinity Linden: @teravus... i don't see the drafts
  • [13:20] Infinity Linden: and i don't see the code
  • [13:20] Zero Linden: that is an MMOX discussion and should occur in the mailing list and in the BoF
  • [13:20] Latha Serevi: That was a non sequitur, Zero. I was saying, let's talk today NOT about LLSD. More on the level of Jon Watte's LESS and Zha's [2]
  • [13:20] Infinity Linden: and i don't see people deploying the code
  • [13:20] Infinity Linden: mea culpa
  • [13:21] Infinity Linden: i'll bait teravus in a private channel after this meeting
  • [13:21] Teravus Ousley: Infinity. In essence, making that argument, you are limiting any input to 'people who make worlds that interop'
  • [13:21] Morgaine Dinova: Teravus++
  • [13:21] Latha Serevi: Zero, OK, my suggestion for a topic is rejected. But is there any role for enhancing the MMOX process today at all, or should we stay away from it?
  • [13:21] Morgaine Dinova: All IETF Contributions are equally germane to the process, if well founded technically.
  • [13:22] Zero Linden: we should stay away from it, I think
  • [13:22] Zero Linden: I think we can talk about specific suggestions to change LLSD - in so far as we can point at the proposal
  • [13:23] Mordack Byron: i dont understand much about the scientific side of the whole IETF chatting, more interesting to me would be, whether people can look at some code _anywhere_
  • [13:23] Morgaine Dinova: Zero, today is about AWG. But as a primary architect here, you need to have a presence on MMOX as well. We're missing your viewpoint.
  • [13:23] Infinity Linden: i could setup a different meeting on thursday to chat about it
  • [13:24] Zero Linden: My viewpoint will be conveyed soon..... (looks at the draft tool....)
  • [13:24] Mordack Byron: cause people can argue over theoretical concepts for a lifetime
  • [13:24] Zero Linden: Yes, no more theoretical concepts here - please - concrete is the name of the game
  • [13:24] Zero Linden: well - if you can believe that coming out of my mouth!
  • [13:24] Mordack Byron: i would like that very much
  • [13:25] Morgaine Dinova: Zero: just writing drafts is not enough. You can't expect people to just endorse your drafts as fait acompli. IETF groups are about discussion, and we need to hold those disucssions with you as well.
  • [13:25] Zero Linden: So - I haven't read draft-ietf-mmox-problem-00.txt yet
  • [13:25] Zero Linden: and don't konw if that is appropriate here
  • [13:25] Zero Linden: HERE
  • [13:25] Zero Linden: the AWG
  • [13:25] Asterion Coen:  :)
  • [13:25] Zero Linden: we are focused on making our interoperable virtual worlds more interoperable
  • [13:26] Zero Linden: Here, I'm not going to be shy about the scope being all about worlds with roughly the same conceptual basis as Second Life
  • [13:26] Latha Serevi: Zero, since it's directly on "scope of MMOX" and you rejected such general discussion, I suggest that mmox-problem-00 will not meet your approval as a topic today.
  • [13:26] Rex Cronon: which VWs are interoperable?
  • [13:26] Zero Linden: I don't think it will, Latha
  • [13:26] Zero Linden: and I didn't see Zha suggesting it
  • [13:26] Latha Serevi: No worries.
  • [13:27] Zha Ewry: Well, I was wickedly crashed
  • [13:27] Latha Serevi: Nothing but LLSD today? (sigh)
  • [13:27] Zero Linden: Seems to me that OpenSim based virtual worlds and Second Life (at least a beta grid) have shown them selves to be interoperable
  • [13:27] Zero Linden: probably RealExtend's version of OpenSim (is that a separate version) falls into that camp
  • [13:27] Zero Linden: I suspect that even HyperGrid falls into the general set
  • [13:27] Morgaine Dinova: The fact that they implement the same model might have something to do with that.
  • [13:28] Zero Linden: Indeed - so - my aim in these meetings is to get those worlds, and worlds that want to join them, to be even more interoperable --
  • [13:28] Zero Linden: and to define what we need to, to make that happen
  • [13:28] Zha Ewry: Which is a subset of everything we could call interop, but a very practical subset
  • [13:29] Zero Linden: Yes, and unabashedly what serves this particular set of worlds -----
  • [13:29] Zha Ewry: nods
  • [13:29] Zero Linden: Basically, we all accomplished something great last Summer - and the next hurdle will require more work and more specing
  • [13:29] Zha Ewry: Tho, one of the challanges, when taking something lke this to the more public square is other people develop opinions too
  • [13:30] Latha Serevi: How about this more provincial version of my topic: "Staying within the OGP side of interop, what list of internet-drafts do we have cooking right now, and how do they serve this particular set of worlds?"
  • [13:30] Saijanai Kuhn: we have serveral parallel things going on right now for interop. A new one is the metanomics-style media streaming between OpenSim/ SL and Croquet which should happen for the BOF meeting
  • [13:30] Zero Linden: excellent
  • [13:31] Zero Linden: and as if on cue: "Your Internet-Draft has been successfully posted in IETF servers and an announcement has been sent out."
  • [13:31] Morgaine Dinova: Zero: you mentioned the many sections of OGP that will be forthcoming. While the detailed info of those may not yet be known, do you have a specific list of the items that will be covered? It was one of the problems identified early on in MMOX, that we don't know the scape of OGP.
  • [13:31] Zero Linden: I'd like to annoucne draft-lentczner-ogp-base-00.txt
  • [13:31] Latha Serevi: woot!
  • [13:31] Jonit Ivory: woot indeed
  • [13:31] Zha Ewry: Do you have a url to that?
  • [13:32] Zero Linden: [3]
  • [13:32] Zero Linden: SO
  • [13:32] Zha Ewry: /m feels like a straight ave there
  • [13:32] Zero Linden: now we've got two agenda topics
  • [13:32] Zero Linden: and I'll take the perogative to order them
  • [13:32] Jonit Ivory: this is rather a defining moment, Zero you should be happy :)
  • [13:32] Zero Linden: 1) OGP drafts in the works and needed next
  • [13:33] Zero Linden: 2) LLSD XML serialization suggestions
  • [13:33] Zero Linden: I'm actually rather relieved....
  • [13:33] Jonit Ivory: hehe
  • [13:33] Zero Linden: and yes, it feels quite nice to have that out there now
  • [13:33] Zero Linden: So - moving forward
  • [13:33] Zero Linden: Drafts we have written about OGP already (not all IETF published):
  • [13:33] Zha Ewry: Ca I offer a 3) based on 2) "Best practices for using LLSD?"
  • [13:33] Zero Linden: LLSD
  • [13:33] Zero Linden: OGP Base
  • [13:34] Zero Linden: OGP Authorization
  • [13:34] Zero Linden: OGP Teleport
  • [13:34] Zero Linden: Zha - is there at least a wiki page on that yet?
  • [13:34] Zero Linden: So - we have these drafts so far
  • [13:34] Zha Ewry: I'll happily make one for 2 weeks hence
  • [13:34] Zero Linden: Zha - that would be most excellent
  • [13:35] Zha Ewry: But, seriously, from the MMOX discussion, it would be a good idea
  • [13:35] Zero Linden: and then it would make a great topic for here
  • [13:35] Zha Ewry: k
  • [13:36] Morgaine Dinova: And apart from those existing drafts, do you know which drafts you will produce? Or at least which topics they will cover?
  • [13:36] Zero Linden: So - https://wiki.secondlife.com/wiki/Open_Grid_Protocol
  • [13:36] Zero Linden: here's the list of drafts on OGP to date
  • [13:37] Zero Linden: Things I'd like to see next
  • [13:37] Zero Linden: OGP Messaging (IM, chat?)
  • [13:37] Zero Linden: OGP Inventory
  • [13:37] Saijanai Kuhn: group IM?
  • [13:37] Zero Linden: yes, IM and Group IM
  • [13:37] Zero Linden: OGP Trust
  • [13:37] Zero Linden: (glances at Zha)
  • [13:38] Zero Linden: OGP Assets (?)
  • [13:38] Zero Linden: OGP Land
  • [13:38] Zero Linden: maybe?
  • [13:38] Zero Linden: anyone else have some thoughts here?
  • [13:39] Jonit Ivory: profiles and friendlists?
  • [13:39] Saijanai Kuhn: appearance?
  • [13:39] Zero Linden: I'm still sort of assuming that we'll hobble along on the old legacy UDP object transport and update system for the whiel
  • [13:40] Morgaine Dinova: So the set of things that OGP will cover is still being worked out.
  • [13:40] Zero Linden: and that getting experience with these other things will give us the fortitude to do those harder ones down the line
  • [13:40] Latha Serevi: that's an intresting ilst. Question, where do 3-d object models (prims, meshes) fit into the OGP spec'ing effort?
  • [13:40] Saijanai Kuhn: oh, and service discovery to allow for custom AD/AD, AD/client and even client/client interactions
  • [13:41] Zero Linden: well- that's what I meant by object transport and update -- those 3-d models and all the stuff that gets that working
  • [13:41] Mystical Demina: doesn't there to be a specification on the endpoint to where you teleport to so things like SLURL, OSURL, etc. can work together and open up portal sites that are multiverse compatabile. Seems ever before full teleport to just be ablt to point the client to a new location and login there is something that will get used
  • [13:41] Zero Linden: is very very touchy in our experience - and so seems like we should get some solid protocol under out belts before attacking that stuff
  • [13:41] Saijanai Kuhn: Zero have you read Avi Bar-Zdev's discussion on SL prims and possible extensions?
  • [13:42] Saijanai Kuhn: Bar-Zeev*
  • [13:42] Zero Linden: no, pointer?
  • [13:42] Saijanai Kuhn: [4]
  • [13:42] Saijanai Kuhn: he wrote the oiriginal code
  • [13:42] Zero Linden: Mystical - good point
  • [13:43] Zero Linden: OGP Viewer Control
  • [13:43] Zero Linden: (?)
  • [13:43] Mystical Demina: can always change the name once it is on the list of things
  • [13:43] Latha Serevi: I'm trying to understand how OGP handles "parts not yet defined yet" such as 3-d. Do we give such pieces of data containers and MIME types that OGP knows how to pass along, for instance, and just not look inside the container?
  • [13:44] Zha Ewry: When we get to the client/sim relationship full up, I'm expecting we'll want to fully factor it as much as possible
  • [13:44] Saijanai Kuhn: There's teh realXtend server-side bots code. WOuld that be suitable for OGP?
  • [13:44] Zero Linden: in the past we've added "legacy" fields that just provide IP,port to the old messaging system
  • [13:44] Zha Ewry: (Seperate out upstream.downstream, and generalize the basics)
  • http://slurl.com/secondlife/Portage/14/154/98 (Starts in 15 minutes)
  • Preview Grid (Aditi) (Starts in 15 minutes)
  • [13:45] Zero Linden: Sai - possible - I'd like to see a write up to know how that fits in.... we have, like many many protocols, stayed way from
  • [13:45] Zha Ewry: (right now, the whole set of tasks in UDP and the stuff that's moved to caps is sort of painfully co-dependent adn entangled)
  • [13:45] Zero Linden: mandating how things "behind the socket" work or can be extended
  • [13:45] Widget Whiteberry: nods to T.Ousley
  • [13:45] Zero Linden: But here's a big one---- OGP HyperGrid?
  • [13:45] Teravus Ousley: nods back
  • [13:46] Zero Linden: I'd love to see some way to fit HyperGrid into the fold of OGP
  • [13:46] Zha Ewry: Only, if you can get hypergid off of .NET remote objects
  • [13:46] Zero Linden: I think they share substantial space
  • [13:46] Zero Linden: well - I think we're talking about the concept and construction of HyperGrid ---
  • [13:46] Saijanai Kuhn: seems traightforward at one level: Agent Domain for external transport, hypergrid for same trust domain transport
  • [13:46] Teravus Ousley: You can use a subset of hypergrid :D
  • [13:46] Zero Linden: if it can be worked into an OGP framework - that would make both stronger
  • [13:46] Zha Ewry: Absolutely
  • [13:47] SharedRealm Engineer: zha, I think hypergrid has dropped .NET remoting.
  • [13:47] Zero Linden: as they could share a substantial amount of common protocol elements
  • [13:47] Zha Ewry: Not last I looked, but it was a few weeks ago, and amen, if so
  • [13:47] Zero Linden: build on each other's work
  • [13:47] Teravus Ousley: I know it's moving off of it currently in increments.
  • [13:47] Morgaine Dinova: Zha: do you think it will it be possible to untangle those co-dependencies into a form suitable for an IETF spec?
  • [13:47] Infinity Linden: wonders if LL would ever open the maingrid to HG though
  • [13:47] Zero Linden: and it would show that what we've built is rather flexible
  • [13:47] Zero Linden: I wouldn't need to, Infinity
  • [13:48] Zero Linden: The point of interoperability is soley that every player does the entirety of any given interoperable spec
  • [13:48] Zha Ewry: I think so, Morgaine
  • [13:48] Mystical Demina: i konw with WCF possible to have any enpoint, transport and message serialization so i am sure if OpenSim need to use something different it would be done
  • [13:48] Morgaine Dinova: Good
  • [13:48] Infinity Linden: to be a HG protocol endpoint?
  • [13:48] Zero Linden: just because Jabber admits mutliple kinds of connection and authentication doesn't mean that every installation (or even implementation) has to
  • [13:48] Infinity Linden: right
  • [13:49] Zha Ewry: And you seperate down into as many small parts as possible, so you can chose which ones you need, ratehr than omnibus ones
  • [13:49] Infinity Linden: i'm just saying.. there's room for HG if there's support for it in the community at large
  • [13:49] Infinity Linden: but
  • [13:49] Infinity Linden: we can't mandate it's ues
  • [13:49] Infinity Linden: use
  • [13:49] Zero Linden: so - if we can all build on a common base, re use common components, then even if different grids choose different parts to support, we are still all winning
  • [13:50] Infinity Linden: but we want to have a clear enough definition of the benefits and drawbacks of HG and OGP/Teleport so peeps know which to use in which situations
  • http://slurl.com/secondlife/Portage/14/154/98 (Starts in 10 minutes)
  • Preview Grid (Aditi) (Starts in 10 minutes)
  • [13:50] Zero Linden: of course
  • [13:51] SharedRealm Engineer: No more .NET remoting in OpenSim: https://lists.berlios.de/pipermail/opensim-users/2009-February/001512.html
  • [13:51] Zero Linden: Perhaps someone would like to collect the list of possible "next drafts" and stick them in the AWG wiki page?
  • [13:51] Zha Ewry: ahh cool
  • [13:51] Asterion Coen: /(i have to run! thx for meeting, have fun folks, kitties, dog, furs, other undefined !
  • [13:51] Zero Linden: okay
  • [13:52] Asterion Coen: take care :)
  • [13:52] Zero Linden: I think we've got that one down
  • [13:52] Rex Cronon: tc
  • [13:52] Zero Linden: and I think have a sense of possible next directions
  • [13:52] Teravus Ousley: listens
  • [13:52] Zero Linden: I will say that this coming month is going to be consuming with the IETF and some end of quarter things for Infinity and I -- so I don't think we'll have anything major newly written for next month.....
  • [13:53] Zero Linden: but we aim to thereafter
  • [13:53] Zero Linden: now
  • [13:53] Zero Linden: Topic 2
  • [13:53] Zero Linden: LLSD XML serializations
  • [13:53] Zero Linden: someone posted a link
  • [13:53] Zero Linden: can they repost it?
  • [13:53] Teravus Ousley: [5] <--- Suggested changes from Catherine Pfeffer
  • [13:53] Infinity Linden: screams quietly to herself
  • [13:54] Infinity Linden: so.. is otherdata a map?
  • [13:54] Zero Linden: Catherine isn't here, which is a detraction
  • [13:54] Infinity Linden: and <binary> is now
  • http://slurl.com/secondlife/Portage/14/154/98 (Starts in 5 minutes)
  • Preview Grid (Aditi) (Starts in 5 minutes)
  • [13:55] Zero Linden: I think, basicallhy, this says that each and every message in the protocol should have it's own custom XML schema
  • [13:55] Zero Linden: is that how others read this?
  • [13:55] Infinity Linden: NO
  • [13:55] Infinity Linden: ACK
  • [13:55] Saijanai Kuhn: Catherine was online earlier hmmm
  • [13:55] Infinity Linden: that kinda leaves a bad taste in my mouth
  • [13:55] Zha Ewry: chuckles softly "What, no fully staticaly typed world with versions and all that fun?"
  • [13:56] Zero Linden: does anyone want to pull for this? It seems very antithetical to the needs we found when building a virtual world
  • [13:56] Latha Serevi: (I'll flee now; will keep IRC open and return here later to delete the IRC relay, or it can be bought for L$0 and deleted by whoever wants. Cheers. )
  • [13:56] Infinity Linden: and why not just say... "hey! here's a bunch of DTDs for application messages"
  • [13:56] Teravus Ousley: Infinity has stated to me in IM that LLSD is pretty much set in stone (with maybe a 64 bit integer) add on. If this is the case, then I'm going to suggest and possibly propose something else be the 'deliverable'.
  • [13:57] Zero Linden: well -
  • [13:57] Zha Ewry: well, one of the sort of design points in LLSD seems to be to avoid the whole "We have a formal schema for every message"
  • [13:57] Infinity Linden: i think the context of the discussion was... no, we're not changing LLSD to include every possible application layer message
  • [13:57] Zero Linden: I'm going to leave the "set in stone" aside -- since nothing is until it has an standards stamp on it...
  • [13:57] Zero Linden: BUT
  • [13:57] Teravus Ousley: I understand that it costs money to change implmentations... and we're all working to create the best solution possible. I don't think it's very 'open' to pull the 'you must have code that is in two implementations' card for every proposal she disagrees with. :D
  • [13:57] Morgaine Dinova: Agree with Teravus. That was a big discussion at the start of MMOX, the need for flexibility in our building bricks.
  • [13:58] Zero Linden: Someone proposed this kind of change in direction for messaging
  • [13:58] Infinity Linden: well.. it's not my card
  • [13:58] Zero Linden: are the assembled here just universally "uhm, no way"?
  • [13:58] Zha Ewry: Its a very traditional XML construsction approach to messaging
  • [13:58] Teravus Ousley: no, though, I think the fact that LLSD is not open to discussion.. may result in a revolt.
  • [13:58] Zero Linden: so- Teravus and Morgaine
  • [13:59] Teravus Ousley: at the very least, it will rehash the topic.
  • [13:59] Zero Linden: HERE we are discussing this particular proposition
  • [13:59] Infinity Linden: and. i've stated repeatedly and publicly that if you want to create DTDs for each application message, i have no problem with that. but don't put it in LLSD
  • [13:59] Zero Linden: to LLSD
  • [13:59] Saijanai Kuhn: it seemed to me that tehre was a lot of ground covered in discussing how to extend/revise/optimize xxSD
  • [13:59] Zha Ewry: wonders what people would actually do with those DTDs (or schemas)
  • [13:59] Saijanai Kuhn: that could be very frutiful
  • [13:59] Zero Linden: I'm not hearing a single persone stand up and say "Yup, we should have XML like this"
  • [13:59] Morgaine Dinova: John Hurliman made a very thorough technical destruction job on LLSD. The fact that he's not here may not be relevant to AWG, but it's definitely relevant to MMOX.
  • http://slurl.com/secondlife/Portage/14/154/98 (Starts now)
  • Preview Grid (Aditi) (Starts now)
  • [14:00] Zero Linden: So -
  • [14:00] Zero Linden: practically, in this forum, with respect to the interoperability work we have done, and will do in the next year
  • [14:00] Zero Linden: no one here is really proposing the kind of change indicated by that web page
  • [14:01] Morgaine Dinova: The issue is being painted in the wrong light anyway, as somehow adversarial. It's not. It's simply a matter of finding the IDL and ADT and serialization system that is best for MMOX. That may not be the same as what is best for SL.
  • [14:01] Zero Linden: It kind of closes that topic -- but I point out that the people who proposed it aren't here
  • [14:01] Rex Cronon: unless the author of that was here:)
  • [14:02] Zero Linden: no - nothing adversarial
  • [14:02] Morgaine Dinova: Zero: they're not here because it was proposed in MMOX. This is AWG, not MMOX --- their presence here would be unusual.
  • [14:02] Zero Linden: right - but someone here proposed it as a topic for AWg
  • [14:02] Zero Linden: and so -
  • [14:02] Zero Linden: it's close
  • [14:03] Zero Linden: closed
  • [14:03] Teravus Ousley: (that person who proposed it was me)
  • [14:03] Infinity Linden: (which might also imply that it's not an appropriate discussion for this group)
  • [14:03] Teravus Ousley: .. (as a topic)
  • [14:03] Zha Ewry: [6]
  • [14:03] Zha Ewry: if you want John's comments
  • [14:03] Morgaine Dinova: AWG was always commited to LLSD/OGP, it's never been an issue here. :-)
  • [14:03] Zero Linden: all's well - I think we've established that we here are quite happy with LLSD not being being that kind of XML serialization
  • [14:04] Zero Linden: okay -
  • [14:04] Zero Linden: well -
  • [14:04] Saijanai Kuhn: though, I'm not sure what "that kind of XML serialization" means since the people who were talking about "that kind" aren't here to clarify
  • [14:04] Zero Linden: we've come to the end of our agenda
  • http://slurl.com/secondlife/Portage/14/154/98 (Started 5 minutes ago)
  • Preview Grid (Aditi) (Started 5 minutes ago)
  • [14:06] Saijanai Kuhn: new business?
  • [14:06] Zero Linden: the going and coming of Zha!
  • [14:06] Teravus Ousley: heh
  • [14:06] Morgaine Dinova: Zero: would it be possible for you to give some (extended) thought to what the items to be covered by OGP will be? Unless you want it hammered out in MMOX ... but that might not lead to what you want (speaking with my AWG hat on). A well though out list would be good.
  • [14:06] Zha Ewry: says very rude things about the loander laptop being flacky, while the laptop is being fixed for being flaky
  • [14:07] Zero Linden: Well - with respect to AWG and OGP - the first agenda item gave a list of next areas to tackle
  • [14:07] Saijanai Kuhn: FRed the wonder Laptop needs sacking?
  • [14:07] Zero Linden: at least as I saw it, with some input from others
  • [14:07] Zero Linden: and I didn't hear too much dissention
  • [14:07] Zha Ewry: Did the link to John's comments I chatted make it onto the sim before I crashed?
  • [14:07] Zero Linden: as for within MMOX - I will say that now that the OGP base document is published
  • [14:07] Morgaine Dinova: Yes Zha
  • [14:07] Saijanai Kuhn: [14:03
  • [14:07] Morgaine Dinova: But no comment was made
  • [14:08] Zero Linden: I can talk to the MMOX list about where I see OGP fitting in, what it does now, and where I think it should head
  • [14:08] Zero Linden: look for a post from me on that to MMOX later this week
  • [14:08] Zha Ewry: I think that some of the "Why we did this" discussion would help
  • [14:08] Morgaine Dinova: That would be excellent.
  • [14:08] Zha Ewry: There's a lot of "WHy on earth did Linden do X"
  • [14:08] Zha Ewry: when, I strongly expect that some exigesis would help
  • [14:08] Zero Linden: I have some replies to J.'s analysis of LLSD/LLIDL
  • [14:09] Zero Linden: and will try to assemble those
  • [14:09] Saijanai Kuhn: quite a few new comments on mmox list in the last hour BTW
  • [14:09] Infinity Linden: fwiw... i posted the beginning of the "why we do things the LLSD way" to the MMOX list
  • [14:09] Zha Ewry: nods
  • [14:09] Zha Ewry: I know, and that was good
  • [14:10] Morgaine Dinova: Yes. I think that the reason there is so much disquiet in MMOX over OGP is because no techical case has been made for it, looks like an imposed solution. We've been discussing it for 16 months, we have a pretty good idea of the "why" (in principle). But for MMOX as a whole, it's unknown why.
  • [14:10] Infinity Linden: i probably need to move that to a web page
  • [14:10] Zha Ewry: There are some related things, that probably would help a bit.. and.. some ammount of "Why caps, and why yet another layer of description language"
  • [14:10] Infinity Linden: and i would like to link to an extended discussion of JHurliman's concerns
  • [14:10] Infinity Linden: and catherines
  • [14:10] Teravus Ousley: (and mine?)
  • [14:10] Infinity Linden: right.
  • [14:11] Zha Ewry: I think that, on the whole, that would be wickedly useful
  • [14:11] Zha Ewry: and. I think some "best practices"
  • [14:11] !Blazed.Cigarette : puffs
  • [14:11] Infinity Linden: didn't you want to see catherine's stuff extended
  • [14:11] Zero Linden: I don't think things in the IETF are imposed
  • [14:11] Infinity Linden:  ?
  • [14:11] Zha Ewry: As in.. "Wow, don't do what was done on the parcel flags" for example.
  • [14:11] Zero Linden: It is a common protocol, used by several implementations, that seemed ready for IETF refinement
  • [14:11] Infinity Linden: good
  • [14:11] Infinity Linden: teravus... were you just advocating for a 64/128 bit int?
  • [14:12] Teravus Ousley: I was advocating for more specific data types and best practices.
  • [14:12] Zha Ewry: There have been several feature requests, such as that, and some cleanu p on how ints/booleans/and maps are done
  • [14:12] Infinity Linden: "more specific" meaning ??
  • [14:12] !Blazed.Cigarette : puffs
  • [14:12] Zero Linden: okay - but
  • [14:12] Teravus Ousley: UInt != binary.
  • [14:13] Infinity Linden: int does not equal binary already
  • [14:13] Teravus Ousley: .. unsigned integer.
  • [14:13] Morgaine Dinova: Refinement implies change. It would be good to know that change is possible --- LLSD seems immoveable, we can barely get a single change on integers considered.
  • [14:13] Infinity Linden: ohoh
  • [14:13] Infinity Linden:  !!
  • [14:13] Infinity Linden: yes
  • [14:13] Zero Linden: I'd like to move that we've run out of agenda
  • [14:13] Infinity Linden: i think zha was saying
  • [14:13] Morgaine Dinova: So add LLSD extensions to the agenda.
  • [14:13] Infinity Linden: "infinity... you almost came out and said 'we made mistakes in the past.' come out and say that and people will be happier"
  • [14:13] Zero Linden: I'd be happy to entertain a HUGH discussion on integers or any other topics of this sort in LLSD next month
  • [14:13] Zero Linden: BUT
  • [14:13] Zero Linden: PLEASE write up specific suggestions somewhere
  • [14:13] !Blazed.Cigarette : puffs
  • [14:13] Zero Linden: and let us know them
  • [14:14] Zero Linden: in advance
  • [14:14] Zha Ewry: Anyeon who has best practice suggestions
  • [14:14] Teravus Ousley: well, I think Infinity will want us to formally propse it and get OpenSimulator and pyOGP to use it before bringing it to them.
  • [14:14] Zha Ewry: Pass them to me and I'll happily coalate them
  • [14:14] Saijanai Kuhn: I've got a Category:MMOX set up on teh wiki, so don't forget to put that at the bottom of any wiki page about mmox type stuff
  • [14:14] Teravus Ousley: .. which I think is being very 'closed minded' bout it.
  • [14:14] Zero Linden: Teravus - I just invited you to write them up and dsicuss them here next month
  • [14:15] Morgaine Dinova: Zero: do you mean that *in principle* we will be able to get multiple integer types into LLSD, as a start, if MMOX reaches agreement that they would be a good thing?
  • [14:15] Teravus Ousley: .. not much of an open discussion.. such is written in that the chair's job is
  • [14:15] !Blazed.Cigarette : puffs
  • [14:15] Infinity Linden: but if you want them to be an IETF standard... you're well served to think how they can be implemented by multiple implementations
  • [14:15] Zero Linden: Look there are two issues at stake
  • [14:15] Zero Linden: er, on the table here
  • [14:15] Zha Ewry: I'm hoping that a well reason discussion will be fruitful
  • [14:16] Zero Linden: 1) How IETF standards get written
  • [14:16] Infinity Linden: terevus... if you have a complaint about how i'm moderating the mailing list to block discussion regarding LLSD, you are welcome to take it up with Lisa
  • [14:16] Zero Linden: for that - please read the IETF docs, history, etc.....
  • [14:16] Zero Linden: 2) What we, the AWG, want to do about LLSD in our interoperability work
  • [14:16] !Blazed.Cigarette : puffs
  • [14:16] Zero Linden: for that, I'm happy to discuss things that need to get changed
  • [14:17] Zha Ewry: With at least some thought about the cost of various changes, and idealy an upwards compatible path
  • [14:17] Teravus Ousley: thank you. I'm mostly saying that because it's written in the working group RFC. Though, I guess we each have our own interpertation of it :D
  • [14:17] Zero Linden: I will point out that the working group doesn't exist yet, and the charter isn't written in stone yet
  • [14:17] Zha Ewry: I'd also gently suggest people try to keep somewhat seperate the
  • [14:17] !Blazed.Cigarette : puffs
  • [14:18] Zha Ewry: "Oh, god, why did you USE LLSD like that"
  • [14:18] Morgaine Dinova: 3) What we, in MMOX (MMOX hat on now), want to do to make LLSD suitable for the MMOX protocol.
  • [14:18] Zha Ewry: from "LLSD could be better"
  • [14:18] Zha Ewry: both have thier place, but.. they are different topics
  • [14:19] !Blazed.Cigarette : puffs
  • [14:19] Zero Linden: wonders where he can buy an MMOX hat...
  • [14:19] Tracy Welles:  :)
  • [14:19] Infinity Linden: Zero!
  • [14:20] Infinity Linden: though i publicly disagree with morgaine on a number of issues, i _would_ like to point out that her contributions to hte list are valued
  • [14:20] Zero Linden: oh - no ---- misunderstood
  • [14:20] Zero Linden: I was thinking of hats in the prim-second-life sense
  • [14:20] Morgaine Dinova: Hehe. MMOX hats are freely available. And it doesn't hurt to swap hats occasionally and give a non-LL view. IETF is about individuals after all, not company reps.
  • [14:20] !Blazed.Cigarette : puffs
  • [14:20] Zero Linden: really
  • [14:20] Zha Ewry: wonders if she can get Chigadee London to do a high fashion one, mens and ladies ;-)
  • [14:21] Zero Linden: exactly!
  • [14:21] Saijanai Kuhn: was thinking something similar to Rex's hat
  • [14:21] Zha Ewry: would not be seen dead in same ;-)
  • [14:21] Saijanai Kuhn: exactly
  • [14:21] Jonit Ivory: it has it's point
  • [14:21] Rex Cronon: lol
  • [14:21] Zha Ewry: One challange for any spec like LLSD
  • [14:21] Zha Ewry: is that just like languages, you can write good LLSD or bad LLSD
  • [14:22] Zha Ewry: I'm hoping some best practices would help there
  • [14:22] Zha Ewry: "Don't do things which require the person consuming the packet to know the semantics when parsing"
  • [14:23] Zero Linden: well- that was a design requirement
  • [14:23] Zha Ewry: (At least ina few spots you end up need ing to know if an Int is a booleano r not, to guess endianness at the moment)
  • [14:23] Morgaine Dinova: Oh, that last one is a great one Zha.
  • [14:23] Zero Linden: parsing has to be generic and ignorant of the specific message involved
  • [14:23] Zha Ewry: I'm pretty sure that's not true int he current use
  • [14:23] Zha Ewry: *in the
  • [14:23] Zha Ewry: Tho "int the" is not uncogent
  • [14:24] Zha Ewry: There is serious oddness when ints could be int, unsigned or boolean
  • [14:24] Morgaine Dinova: John "ambiguity" post, right at the start.
  • [14:25] Zero Linden: if we have ever serialized an unsigned int as an LLSD int (signed) - then I bow down and wail "mea culpa"
  • [14:25] Zha Ewry: I think Teravus pointed out a few of those ;-(
  • [14:26] Zero Linden: and really, region ids that are 64 bit unsigned horrors of a x,y location smashed together are also a sin for whch we will do eternal pennance
  • [14:26] Zha Ewry: But, its not supposed to work that way, I know
  • [14:27] Zero Linden: okay, on that note
  • [14:27] Zero Linden: I think it is time to call this a wrap
  • [14:27] Zero Linden: thanks all for coming
  • [14:27] Infinity Linden: kudos to teravus for (repeatedly) bringing it up.. and keeping it real wrt that example
  • [14:27] Rex Cronon: it could be possible to have 2 versions on th e wire. one horrible and one nice:)
  • [14:27] Zha Ewry: ttoally true, Rex
  • [14:27] Saijanai Kuhn: t;he division is naughty and nice
  • [14:28] Morgaine Dinova: Thanks Zero
  • [14:28] Zero Linden: later all
  • [14:28] Rex Cronon: bye zero
  • [14:28] Jonit Ivory: cheers Zero
  • [14:28] Saijanai Kuhn: laters Zero
  • [14:28] Zha Ewry: Thanks, people. And.. anyone who hasn't looked at the MMOX scope document is strongly encouraged to do so
  • [14:28] Jonit Ivory: ghe leaves so abruptly
  • [14:28] Jonit Ivory: he^^
  • [14:28] Saijanai Kuhn: in Second LIfe, "ghe" might be appropriate
  • [14:28] Rex Cronon: like somebody yanks the plug:)
  • [14:28] Jonit Ivory: hehe
  • [14:28] Harleen Gretzky: bye Zero
  • [14:28] Infinity Linden: lol
  • [14:29] Torrid Luna: thanks all, and good bye^^
  • [14:29] Rex Cronon: tc
  • [14:29] Jonit Ivory: ciao ciao
  • [14:29] Saijanai Kuhn: end transcript
  • [14:30] Saijanai Kuhn: nods off (again)