User:Andrew Linden/Office Hours/2008 07 22

Transcript of Andrew Linden's office hours:

[11:05]	The region you have entered is running a different simulator version. Click this message for details.
[11:05]	Kitto Flora:	Linden Lab is so name d cause it started on Linden Street
[11:05]	Arawn Spitteler:	Quite some fireworks, to REz Simon
[11:05]	RacerX Gullwing:	shields ha
[11:05]	Kitto Flora:	Linden Street may be named as it has lime trees on it?
[11:05]	Rex Cronon:	sorry
[11:06]	Kitto Flora:	Hello Andrew
[11:06]	Arawn Spitteler:	Linden Street was probably named, because people thousands of miles away wanted to imagine Lime Trees
[11:06]	Andrew Linden:	Hello everyone.
[11:06]	Arawn Spitteler:	Hi, Andrew
[11:06]	Andrew Linden:	Yup, Linden Lab started on Linden Street in SF
[11:06]	Rex Cronon:	hi andrew, simon
[11:06]	Andrew Linden:	which is really a very small one-way alley
[11:06]	Arawn Spitteler thinks there ought to be a build.
[11:06]	Kitto Flora:	Its not usual to put citrus trees on streets though. They are messy
[11:07]	Andrew Linden:	The original office was a converted warehouse, very close to where the old highway ran through the city before the '89 earthquake
[11:07]	Arawn Spitteler:	1889 or 1989?
[11:07]	Andrew Linden:	The neighborhood has changed quite a bit since we started there... gentrification is taking its toll
[11:07]	RacerX Gullwing:	well i'm here to find out if theres a fix in the works for my 1179 problem
[11:08]	Arawn Spitteler:	url?
[11:08]	Andrew Linden:	there is a new park, higher buildings, and the diagonal scar that was the raised highway location is harder to notice
[11:08]	Andrew Linden:	lessee... SVC-1179...
[11:08]	Andrew Linden:	lemme look it up
[11:08]	RacerX Gullwing:	and todays rollout didn't seem to effect my colision problem
[11:09]	Andrew Linden:	http://jira.secondlife.com/browse/SVC-1179
[11:09]	Andrew Linden:	I think I've fixed that, however the fix is not in the 1.23-Server update
[11:10]	RacerX Gullwing:	you think you fixed it when will you know?
[11:10]	Andrew Linden:	the 1.23 server has been in the pipe for a long time, and we're having trouble pushing it out -- we keep finding new problems when we do
[11:10]	Andrew Linden:	I believe it was going out today and is currently getting rolled back yet again
[11:10]	Andrew Linden:	the crash rate went way up, however it is not clear what was causing it
[11:11]	RacerX Gullwing:	oh well i tried after the sim was rolled both times no change
[11:11]	Andrew Linden:	it may have been a config issue rather than new crash modes
[11:11]	Seemok Capalini gave you [KMA] Kyuss Motors Amsterdam/ Seemok Capalini.
[11:11]	Kitto Flora:	Sim crash?
[11:11]	RacerX Gullwing:	no rollout
[11:11]	RacerX Gullwing:	twice today same sim
[11:12]	Andrew Linden:	Yes Kitto, simulator crashes.
[11:12]	Andrew Linden:	RacerX, the fix for SVC-1179 is not deployed yet anywhere for you to test.
[11:12]	Kitto Flora:	OK - probably explains a rash of 'returns - gone off world that I've seen
[11:12]	RacerX Gullwing:	ok
[11:12]	Andrew Linden:	When it is deployed somewhere I'll let you know.
[11:12]	Andrew Linden:	It will first be deployed to a test grid... probably ADITI
[11:12]	RacerX Gullwing:	well keep working on it someday i'd like my stuff to work again
[11:13]	Kitto Flora:	RacerX metioned collision problems, what are they?
[11:13]	Andrew Linden:	lesee... I wonder if I've got some announcements
[11:14]	Andrew Linden tries to remember...
[11:14]	RacerX Gullwing:	look at the snail there notice when he hi\ts the green thing it pops him up but the yellow one doesnt do anything till hes gone through it and is on the way down also he goes different distances each time
[11:14]	Andrew Linden:	I wasn't able to reproduce the "hopping avatar when hugging" bug, however I have a theoretical fix for it anyway.
[11:15]	Dahlia Trimble:	havent seen it lately
[11:15]	RacerX Gullwing:	oh one out of 6 or 7 it actauly works riight
[11:15]	Andrew Linden:	Ah, I suspect I've fixed that bug too RacerX, or at least made it much better
[11:15]	Andrew Linden:	lemme see if I can find the jira for that...
[11:15]	Andrew Linden:	RacerX, it is actually 1 out of 8
[11:16]	Andrew Linden:	that kind of stat makes me pretty confident we're talking about the same problem
[11:16]	RacerX Gullwing:	well we only let 6 snails on the track at a time and usally it works once
[11:16]	Arawn Spitteler:	Sounds like three bits
[11:17]	Andrew Linden:	no, there is "phasing" going on, and the collision events are currently batched up and may be processed up to 8 frames later than when they happened
[11:17]	RacerX Gullwing:	http://slcn.tv/media/racerx_samples_01jul08.mp4
[11:17]	RacerX Gullwing gave you RacerX better example film and object cannon.
[11:17]	RacerX Gullwing:	thats the example of before and after
[11:18]	RacerX Gullwing:	well i built a new course cause after 2 months it was just really gettiung old so we \no longer use cannons but hoping someday we can again
[11:18]	Arawn Spitteler:	Is this a viewer feature? I clicked on the URL, and the system wanted to dial up?
[11:19]	RacerX Gullwing:	http://slcn.tv/media/racerx_samples_01jul08.mp4
[11:19]	RacerX Gullwing:	hmm
[11:19]	Andrew Linden:	Ok, so when trying to fix SVC-2511 I noticed the collision event problem, and fixed it
[11:19]	Dahlia Trimble:	thats probably your internet settings Arawn
[11:19]	Andrew Linden:	meanwhile there have been other reports of collision events firing late
[11:20]	Andrew Linden:	https://jira.secondlife.com/browse/SVC-2511
[11:20]	Arawn Spitteler:	It should be just me, but I was wondering if it wasn't. My Firefox is already open, but I'm not opening hte URL
[11:20]	RacerX Gullwing:	yeah thats the latest problem
[11:20]	RacerX Gullwing:	before that one it was just inconsistent push
[11:21]	RacerX Gullwing:	now its they don't get pushed at all
[11:21]	Andrew Linden:	Ah, here is one of the recent ones I picked up... SVC-2655
[11:22]	Andrew Linden:	https://jira.secondlife.com/browse/SVC-2655
[11:22]	Dahlia Trimble:	I've noticed the collision lag in scripts since H1, thought it was just too many scripts running delaying events in general
[11:23]	Andrew Linden:	RacerX, I'll try to actually test the items you gave me, especially if there is no deploy of the fixed code by Thursday
[11:23]	RacerX Gullwing:	ok great thanks for caring
[11:24]	Andrew Linden:	Hrm... maybe I should fire up a region with some bleeing edge code on ADITI on Thursday and people can come test their bugs if they want.
[11:24]	Andrew Linden:	I'll make that the plan... after the Thursday office hour I'll hold a special session (most likely on some region in the ADITI grid) for the hour afterward
[11:25]	Andrew Linden:	so that would be... 18:00 - 19:00
[11:25]	Andrew Linden:	The final location will be announced on Thursday
[11:25]	Andrew Linden:	since I haven't yet set it up
[11:25]	Andrew Linden:	Note that the ADITI grid's inventory will be out of date
[11:25]	RacerX Gullwing:	so i'll need to come to that?
[11:25]	Andrew Linden:	probably by months
[11:26]	Andrew Linden:	RacerX, you only need attend if you like
[11:26]	RacerX Gullwing:	ok
[11:26]	Arawn Spitteler:	Transcripts should say where, once posted
[11:26]	Andrew Linden:	but it would be an opportunity to verify that the bugs I'm claiming I've fixed really are fixed.
[11:26]	Andrew Linden:	Consider it an early view on the bleeding edge
[11:26]	RacerX Gullwing:	i'm just here being the sqweeky wheel
[11:26]	RacerX Gullwing:	fix me
[11:27]	Andrew Linden:	I suspect most of the bugs fixed in that simulator won't have even passed QA yet, since the QA queue is full at
[11:27]	Andrew Linden:	at the moment
[11:27]	Andrew Linden:	and things are stalled, waiting for 1.23-Server to clear
[11:27]	Andrew Linden:	In any case, attendance is complete voluntary, and all are invited
[11:28]	Andrew Linden:	Ok, anybody have any other issues to raise?
[11:28]	Andrew Linden:	The table is open.
[11:28]	Arawn Spitteler has a class, at that time, but will try to make it in later.
[11:28]	Andrew Linden:	Or topics to discuss?
[11:28]	Kitto Flora:	Any hints of llSetPos() failing for PHYSICAL objects?
[11:28]	Arawn Spitteler:	I was wondering, about the rate of control input. Is tht predictable at all?
[11:29]	RacerX Gullwing:	oh one thing nice my castle wars game has gotten funner the cannonballls seem to bounce around now taking out multiple peices
[11:29]	Dahlia Trimble:	I have a question about collision geometry of twisted spheres
[11:29]	Andrew Linden:	Kitto, I haven't heard of llSetPos() failing for physical objects. Do you have a simple repro recipe?
[11:30]	Andrew Linden:	Arawn, what "control input"? Are you talking about "avatar control input"?
[11:30]	Kitto Flora:	None thats simple. I designed it out by setting object non-phys while it used the llSetPos() warping scheme. That made it work right. Just wondered if its a general problem.
[11:30]	Arawn Spitteler:	When the control event is called, in a script, I don't know how often
[11:30]	Andrew Linden:	Dahlia, just state your question about collision geometry and I'll try to get to it.
[11:31]	Andrew Linden:	Kitto, llSetPos() failures on dynamic objects would be news to me.
[11:32]	Dahlia Trimble:	If I have a sphere that is twisted in such a way that the surface normals are reversed, the surface isnt visible to the viewer but it acts like the collision mesh knows it's there. Does the collision mesh have 2 normals for each surface or something?
[11:32]	Andrew Linden:	Arawn, I'm not sure what the throttle is on control event rates in scripts
[11:32]	Andrew Linden:	however they woudl be affected by the load on the script engine
[11:33]	Andrew Linden:	and may have some throttle/rate-limit the the fast side
[11:33]	Arawn Spitteler:	So I should jira a new feature for something like control throttling, where consistency is desired?
[11:33]	Andrew Linden:	Also, the control input is stripped from the avatar contol input, and that is supposed to be sent about 10 updates/second
[11:34]	Simon Linden:	Dahlia - that sounds like a viewer problem ... it seems like the collisions are happening correctly, but it's not shown right in the viewer. Is that correct?
[11:34]	Andrew Linden:	so the script control event can't be faster than the client is updating it
[11:34]	Dahlia Trimble:	right
[11:35]	Dahlia Trimble:	heres an example
[11:35]	Simon Linden:	OK, well, I don't have much of a solution except it sounds like a good bug to file. Since it seems like a display issue, it will end up with the viewer team
[11:35]	Dahlia Trimble:	k
[11:36]	Dahlia Trimble:	it may break content, such as Arawn's example
[11:36]	Andrew Linden:	Dahlia, could you rephrase that question? You say the normals point inside and the triangles are then not renderd (back-face culling) but it collides how? Correct or wrong?
[11:36]	Dahlia Trimble:	correct
[11:36]	Dahlia Trimble:	the surface collides, but isnt visible
[11:36]	RacerX Gullwing:	bye
[11:37]	Andrew Linden:	Ah. Well, if the shape is fundamentally convex, then a convex shape is used and there are no real normals used in generating the geometry
[11:37]	Andrew Linden:	If the shape is concave, then we use a MeshShape, and feed the same triangles that are used for rendering (but sometimes at a lower LOD (level of detail))
[11:38]	Andrew Linden:	Now, if we feed total garbage to the MeshShape then bad things happen
[11:38]	Dahlia Trimble:	so there is a hull that surrounds the shape?
[11:38]	Andrew Linden:	I can't remember if we detect for garbage geometry and fix it...
[11:38]	Simon Linden:	I'd be curious to compare the physics shape with what's in the viewer
[11:38]	Andrew Linden:	we used to use the mass properties that Havok returned to detect when we had fed inverted triangles to the physics engine,
[11:39]	Andrew Linden:	however now we compute our own mass properties
[11:39]	Simon Linden:	There is some triangle reduction code for the physics shape
[11:39]	Andrew Linden:	Dahlia, could you create that shape again?
[11:39]	Andrew Linden:	I'll take a copy
[11:39]	Andrew Linden:	and we can look at it in the Havok Visual Debugger in the lag to see what the true collision shape is
[11:39]	Dahlia Trimble:	I have a better one at my skybox, I'll give you one of those
[11:39]	Andrew Linden:	in the 'lab'
[11:40]	Dahlia Trimble:	lol
[11:40]	Andrew Linden:	Ok I had two topics to offer up for discussion today
[11:41]	Andrew Linden:	The first was something Simon and I were just talking about before the hour started
[11:41]	Dahlia Trimble gave you strange collision shape.
[11:41]	Andrew Linden:	The 1.23-Server deploy has been having trouble getting out
[11:42]	Andrew Linden:	because we find a new blocker bug each time we try
[11:42]	Andrew Linden:	and the turnaround time for fixing the bug, QA'ing, and then letting sit on a test grid again for a few days is long
[11:43]	Andrew Linden:	Back when we had the Havok4 early adopter program we could sometimes get fixes out faster, at least within the EA regions
[11:43]	Andrew Linden:	And it seems that often our test grids just aren't big enoug to find all the bugs
[11:44]	Kitto Flora:	Test grids dont get the traffic/use to show up the bugs.
[11:44]	Andrew Linden:	so I was wondering if any of you thougth that it would be useful to have a voluntary "EA" program for the bleeding edge simualtor code
[11:44]	Andrew Linden:	and if so, what kinds of incentives might convince estate owners to participate?
[11:44]	Andrew Linden:	I'm just kicking this idea around, only Simon and I have talked about it as far as I know
[11:44]	Kitto Flora:	Hahaha - a free sim?
[11:44]	Rex Cronon:	reduce their tier:)
[11:45]	Dahlia Trimble:	I liked the H4 sims on the main grid like Crash Me was before the H4 intro
[11:45]	Seifert Surface:	do you want the sim owners who participate to bug report as well?
[11:45]	Seifert Surface:	if so, money may not be the right way to do it
[11:45]	Andrew Linden:	Yes Seifert, that would be one benefit -- bugs found sooner and reported
[11:45]	Andrew Linden:	then a fixed iteration deployed sooner
[11:46]	Seifert Surface:	getting good reports out of people isnt that easy to do...:)
[11:46]	Kitto Flora:	H4 was a big incentive - less sim crashes promised. But more recent updates - whats the incentive?
[11:46]	Arawn Spitteler:	We'd be happy to expand the sandbox of Spirit City to a free open grid. "Bleeding Edge Simulation!!!"
[11:46]	Andrew Linden:	Seifert, that is true, and an argument used in the past
[11:46]	Seifert Surface:	could just be entirely voluntary
[11:47]	Seifert Surface:	and some people, who use the jira anyway will probably get involved
[11:47]	Andrew Linden:	Right, H4 was kinda an exciting project to try, but most bleeding edge deployments are rather uninteresting
[11:47]	Seifert Surface:	yeah
[11:47]	Seifert Surface:	tricky though, if you could end up with a broken sim as a result of helping out
[11:48]	Andrew Linden:	Some of the bugs that held up 1.23 already were Estate managment bug... only estate owenrs and managers would have noticed
[11:48]	Kitto Flora:	And good ones of them are few/far between
[11:48]	Arawn Spitteler:	Offer Open Space Sandboxes, to the first 32 best qualified applicants.
[11:49]	Arawn Spitteler:	Open Space might harbour undisclosed standard sim failures, though
[11:49]	Seifert Surface:	tricky to do long term incentives like sims
[11:49]	Andrew Linden:	what if we had a penninsula of the main grid mainland that was always running bleeding edge regions, and parcel tier on them were halved?
[11:50]	Seifert Surface:	what if someone wants out of the program but has built stuff?
[11:50]	Seifert Surface:	i suspect youd get a ton of people that just want cheap tier
[11:50]	Andrew Linden:	Yeah, if we did it by estates then we'd have to allow the Estate managers an easy way to opt in and out
[11:51]	Seifert Surface:	youd get population but not necessarily bug reports
[11:51]	PulseBurst Flow:	Andrew..would another alternative be for LL just to hire some QA Testers and let them test the sims?
[11:51]	Andrew Linden:	You're probably right Seifert
[11:51]	Arawn Spitteler:	Allow the rental of open space sandboxes by the month, with discount on bleeding edge
[11:52]	Andrew Linden:	Yes PulseBurst, the solution may just be to beef up, overhaul, fix our QA process
[11:52]	Seifert Surface:	i hesitate to suggest it, but link continued participation in the freebie whatever to bug reports?
[11:52]	PulseBurst Flow:	rather then depend on volunteers..and deal with the issues of perks..and are enough volunteers reporting.
[11:52]	Andrew Linden:	I was curious to know if it would be possible to let some Residents try early.
[11:53]	Dahlia Trimble:	I think a way to teleport to the beta grid might be an incentive ;)
[11:53]	Arawn Spitteler:	An on-going competition for EA slots?
[11:53]	Andrew Linden:	Hrm... Dahlia, that is an interesting idea...
[11:53]	Seifert Surface:	could try straight volunteer, but perhaps with guaranteed technical concierge service
[11:53]	Andrew Linden:	Normally a beta grid is very empty, very little incentive to hang out there
[11:53]	Seifert Surface:	which youd want in case of problems anyway
[11:53]	Arawn Spitteler:	How soon will Aditi TP be available to Agni? It sounds consistent with OGP
[11:54]	Seifert Surface:	but some people might like to have access to anyway
[11:54]	Andrew Linden:	but if it were easy to get to maybe it would get more explorers
[11:54]	PulseBurst Flow:	yes, well..there are certainly some knowledgable and interested residents...no doubt.
[11:54]	Andrew Linden:	that is, make the beta grids near extenstions of the main grid
[11:54]	Seifert Surface:	soudns like the open grid stuff
[11:54]	Seifert Surface:	got to freeze money though if you go to beta?
[11:54]	Arawn Spitteler:	Beta Grid has differnt rules, for lasting value, though.
[11:54]	Andrew Linden:	Yup, but don't know the timeline on that getting done.
[11:54]	PulseBurst Flow:	" little incentive to hang out there" would be different for people who's job it is to "hang out there"
[11:55]	PulseBurst Flow:	I think M said...that LL is doing well now.
[11:55]	Andrew Linden:	Ok, well thanks for the discussion. I'll re-read the transcripts after I post them
[11:55]	Andrew Linden:	now what was that other topic I was going to talk about?...
[11:56]	Andrew Linden:	Oh yeah, I was brainstorming about "phantom avatars" as a feature
[11:56]	Andrew Linden:	lemme find the jira item...
[11:56]	Dahlia Trimble has another meeting, thanks all and bye :)
[11:56]	Rex Cronon:	bye dahlia
[11:56]	Andrew Linden:	MISC-1407
[11:56]	Seifert Surface:	bye
[11:56]	PulseBurst Flow:	bye
[11:56]	Rex Cronon:	like the don't intereact with the h4 engine?
[11:56]	Andrew Linden:	https://jira.secondlife.com/browse/MISC-1407
[11:57]	Andrew Linden:	There is currently a bug (SVC-2570) that allows your avatar to become the same collision group as VOLUME_DETECT objects
[11:57]	Andrew Linden:	which is essentially collisionless
[11:57]	Andrew Linden:	I've already fixed the bug internally
[11:57]	Andrew Linden:	but some people would like to be collisionless to avoid the "shoot-em-up" gameplay (griefing)
[11:58]	Seifert Surface:	as i understand it, the argument against this in the past has been about wanting people not walking through your walls
[11:58]	Arawn Spitteler:	I dont see any problem, with letting avatars be phantom when sitting on phantom vehicles, but understand that's been squelched for control of cheating.
[11:58]	PulseBurst Flow:	yes, I don't understand..wouldn't you just fall through things?
[11:58]	Andrew Linden:	I was wondering what you all thought about the feature idea... allow people to optionally set their avatar collisionless (or only collides with some stuff)
[11:58]	Andrew Linden:	and have that be a per-region setting that can be disabled
[11:59]	Seifert Surface:	might not be popular with some people
[11:59]	Rex Cronon:	i wonder if u can restrict the ability of those that are phantom from rezzing things?
[11:59]	Kitto Flora:	Does that make the avatar 'phantom' ?
[11:59]	Seifert Surface:	although people who want to get through walls can do so already
[11:59]	Andrew Linden:	Yes, the bug currently allows your avatar to become effectively "phantom"
[11:59]	Arawn Spitteler can get through Access Lines, if he can rez nearby.
[12:00]	Kitto Flora:	I think that change would make little difference to anything that I work on/make.
[12:00]	Arawn Spitteler:	I'd like Static Avies to be an option.
[12:00]	Rex Cronon:	can those that become phantom loose the ablity to rezz things from their attachments?
[12:00]	Seifert Surface:	if you hide it as some sort of primitive params call or some such, it will be as obscure as the current methods for cheating around walls
[12:00]	Seifert Surface:	rex: why is that useful?
[12:01]	Seifert Surface:	arawn: static avie == sitting on a prim?
[12:01]	Rex Cronon:	if i become phantom at rausch, than i can shoot everybody without getting hurt
[12:01]	Arawn Spitteler:	Some sims, might not require smothed motion
[12:01]	Andrew Linden:	Right Rex, so shoot-em-up play would only be fun in a region that had that feature disabled
[12:01]	Seifert Surface:	rex: then rausch should turn off phantom av ability
[12:01]	Rex Cronon:	no
[12:02]	Rex Cronon:	maybe i want to watch
[12:02]	Arawn Spitteler:	Whether features are used for Cheating at games, is a challenge for hte Marshallate.
[12:02]	Andrew Linden:	It MUST be possible to disable on a per region basis since Rausch (and regions like it) requires it
[12:02]	Seifert Surface:	then sit on a cube?
[12:02]	Rex Cronon:	but i shouldn't be able to interfere
[12:02]	Seifert Surface:	ah, maybe this should be split
[12:02]	Seifert Surface:	links to the idea of "viewer avs" that was floating around a while back
[12:02]	Arawn Spitteler:	Cheating happnes, and ever sport negotiates its accomodations
[12:02]	Andrew Linden:	No, there are other uses for rezzing objects than to shoot them
[12:03]	Seifert Surface:	so you could fit more than 40 avs into a concert sim
[12:03]	Andrew Linden:	the simulator could not be made smart enough to tell what rezzes were bullets and what were not
[12:03]	Seifert Surface:	that turns off physics for the av, and also turns off a whole bunch of other things
[12:03]	Rex Cronon:	ok, than everything that is rezzed by those that are phantom should be phantom too:)
[12:03]	Andrew Linden:	Suppose your avatar could be made to not collide with dynamic objects only
[12:03]	Andrew Linden:	woudl that be enough?
[12:04]	Seifert Surface:	rex: what if i want to build in a sandbox and build non phantom, and not be shot at?
[12:04]	Andrew Linden:	Then the "collisionless" avs could walk on floors and bump into walls
[12:04]	Rex Cronon:	u do that in a damage enabled sim?
[12:04]	Seifert Surface:	andrew: there must be cages that work non physically surely?
[12:04]	Rex Cronon:	i am only talking about damage enabled sims
[12:04]	Andrew Linden:	yes Seifer, static cages are common
[12:05]	Seifert Surface:	then only working with dynamic prims wouldnt help
[12:05]	Andrew Linden:	but... suppose the avatar could always tunnel through static content... when moving up (but not down or sideways)?
[12:06]	Seifert Surface:	make very tall cages
[12:06]	Arawn Spitteler:	It would be nice, if we could sit on a prim, and direct the prim to move as an avatar should.
[12:06]	Andrew Linden:	hehe, I'm glad I've got you guys helping me out
[12:06]	Seifert Surface:	;)
[12:06]	Seifert Surface:	arawn: isnt that equivalent to phantom avs?
[12:07]	Arawn Spitteler:	Anti Griefing Devices already exist, but move badly
[12:07]	Rex Cronon:	not all arawn:)
[12:07]	Rex Cronon:	u should try smoothmove
[12:07]	Andrew Linden:	Ok, well thanks for the discussion. I've got to run.
[12:08]	PulseBurst Flow:	thanks andrew
[12:08]	Andrew Linden:	As I mentioned earlier, on Thursday after the office hour I'll be holding a 1 hour session on a test grid
[12:08]	Rex Cronon:	i have to go to
[12:08]	Rex Cronon:	bye everybody
[12:08]	Andrew Linden:	where the latest unreleased code simulator can be tested
[12:08]	Andrew Linden:	for those who want to see if their bugs are fixed (but not yet deployed)
[12:08]	Andrew Linden:	Location to be announced at the next office hour
[12:08]	Seifert Surface:	including that air resistance thing?
[12:09]	PulseBurst Flow:	bye all
[12:09]	Kitto Flora:	OK
[12:09]	Rex Cronon:	have fun
[12:09]	Kitto Flora:	Depends on whats on my schdule...
[12:09]	Kitto Flora:	Bye all
[12:09]	Kitto Flora runs for lunch