User:Andrew Linden/Office Hours/2009 02 19

Transcript of Andrew Linden's office hours:

[17:04]	pickle Flagon:	hello
[17:04]	Sindy Tsure:	hi andrew
[17:04]	Andrew Linden:	Hello everyone
[17:04]	Moon Metty:	hi Andrew :)
[17:05]	pickle Flagon:	oh sorry
[17:05]	Simon Linden:	no problem
[17:05]	xstorm Radek:	hi Andrew
[17:05]	Chaser: Hello, Avatar!
[17:05]	Andrew Linden:	Ok, I've got a few announcements to make...
[17:05]	Yukinoroh Kamachi:	good morning
[17:05]	Sindy Tsure waves hi
[17:06]	Andrew Linden:	(1) I've convinced the #release team to consider including all the bug fixes that have piled up in the maint-server-6 branch for including into server-1.26
[17:06]	xstorm Radek:	hi Yukinoroh Kamachi
[17:06]	Andrew Linden:	(that branch was previously destined for server-1.27 which is months away)
[17:06]	Chaser serves:	Arawn Spitteler
[17:07]	Andrew Linden:	There are a bunch of public jira bugs that I flagged as "fix pending for server-1.27" but there is a chance they might actually get into server-1.26
[17:07]	Sindy Tsure claps!
[17:07]	Andrew Linden:	the final decision has not yet been made, but they bug fixes are being retested right now
[17:07]	Arawn Spitteler hasn't heard of this as a griefing mode, yet: If you click thins, it won't register on your hits.
[17:07]	Andrew Linden:	Anyway, if they are accepted then I'll have to go through and update some public jira's
[17:08]	Moon Metty:	cool
[17:08]	xstorm Radek:	what about ghosting problems and Stuck people will that get fix to ?
[17:08]	Andrew Linden:	and I'll mention the more interesting bugs in next week's office hours (if maint-server-6 is accepted)
[17:08]	Yukinoroh Kamachi:	what about buoyancy ?
[17:09]	Andrew Linden:	(2) I just locked maint-server-7 which has a lot of bug fixes (but none have been verified/tested yet)
[17:09]	Sindy Tsure:	hi again, neuro
[17:09]	Andrew Linden:	that branch is destined for server-1.27 for sure
[17:09]	Neuro Linden:	Evening Andrew, hi sindy, evening all :)
[17:09]	Andrew Linden:	I don't have a list of bugs fixed in maint-server-7 handy, but maybe next week
[17:09]	xstorm Radek:	hi Rex
[17:09]	Andrew Linden:	Welcome Neuro
[17:09]	Sindy Tsure:	any idea how high server-1.25 is going to go? 1.24 went up to .10 or so
[17:10]	Rex Cronon:	hi xstorm
[17:10]	Rex Cronon:	hi everybody
[17:10]	Sindy Tsure:	hey rex
[17:10]	Moon Metty:	1.25 was terribly unlucky
[17:10]	Andrew Linden:	Sindy, the #release team does not want to deploy another patch release for server-1.25
[17:10]	Simon Linden:	We're hoping for no more 1.25 versions
[17:10]	Rex Cronon:	hi
[17:10]	Moon Metty:	hey Rex :)
[17:10]	Simon Linden:	Hi Rex
[17:10]	Andrew Linden:	the expectation is that such a patch release would delay server-1.26
[17:10]	Arawn Spitteler wonders if 1.25 will be upgraded once 1.26 has taken its place
[17:10]	Andrew Linden:	and no one want's that
[17:10]	Rex Cronon:	:)
[17:10]	Sindy Tsure:	cool
[17:11]	Andrew Linden:	The last thing on my agenda is a quick poll...
[17:11]	Yukinoroh Kamachi:	Since last year in April, it's been impossible or hard to script any artificial no-gravity movement, anything on the table to adress that issue ?
[17:11]	Andrew Linden:	I was talking to a fellow Linden at lucnh and he was wondering what would happen if we made a login poll for all SL residents:
[17:11]	Sindy Tsure:	you would get complaints
[17:12]	Arawn Spitteler:	Wasn't it a samplinig, previously?"
[17:12]	Sindy Tsure:	but LL gets complaints no matter what you do (some of which is ok)
[17:12]	pickle Flagon:	being a linden resident a login poll soounds good
[17:12]	Andrew Linden:	"Which would you prefer? (a) faster deployment of features? or (b) more stability improvements?
[17:12]	Rex Cronon:	depends o what the pool is about:)
[17:12]	Andrew Linden:	of course that poll isnt' going to be done
[17:12]	Sindy Tsure:	oh my.. trying to start a religious war?
[17:12]	Yukinoroh Kamachi:	b
[17:12]	Neuro Linden would bet on b :)
[17:12]	pickle Flagon:	(b)
[17:12]	Andrew Linden:	but I figured I would ask everyone here to see what they would say.
[17:12]	Aaron23 DeCuir:	it alternates andrew
[17:12]	Rex Cronon:	can there be "(c) boths" :)
[17:13]	Rex Cronon:	"(c) both"
[17:13]	Sindy Tsure would also like 'both'
[17:13]	Aaron23 DeCuir:	the truth is - both in sequence
[17:13]	Sindy Tsure:	can i suggest a different poll?
[17:13]	Aaron23 DeCuir:	first add feathers
[17:13]	Andrew Linden:	Well, when trying to maximize one the other tends to suffer
[17:13]	xstorm Radek:	i hope the login poll is better then the one i get much of the time asking if i tell others and i keep saying yes to every one i can :-)
[17:13]	Aaron23 DeCuir:	oops - features*
[17:13]	Moon Metty:	it's always a balance between fixes and new features, it changes over time
[17:13]	Andrew Linden:	so there is no "both", unless you're saying they need to be balanced somehow... differently than they are balanced today.
[17:14]	Sindy Tsure:	ask instead how people want to weight the two options.. 50%/50% fixes/features or 0/100 or 25/75 or whatever..
[17:14]	Arawn Spitteler:	Devide into teams, with one team providing up to date server software, and the other team producing more stable server software.
[17:14]	Andrew Linden:	Sure Sindy, if you have a poll for us here, go ahead and state it.
[17:14]	Charlette Proto:	hi everyone
[17:14]	Sindy Tsure points a few lines up in the chat
[17:14]	xstorm Radek:	hi Charlette
[17:15]	Rex Cronon:	hi
[17:15]	Sindy Tsure:	heya charlette
[17:15]	Charlette Proto:	sorry I'm late
[17:15]	Andrew Linden:	Well, it is kinda hard to define what 50% means between those two... but perhaps you mean "50% of LL resources"?
[17:15]	Sindy Tsure:	instead of asking a binary this/that, ask how much of the time LL should put into features.. 0%, 25%, 50%, 75% or 100%
[17:16]	Moon Metty:	even a bug that isn't very serious will have to be fixed at some time
[17:16]	Sindy Tsure:	of SW engineering resources
[17:16]	Andrew Linden:	Anyway, I was curious to see what the trend would be here. Looks like option (a) (faster feature deployment) loses out.
[17:16]	Sindy Tsure:	it's the same question as you said but with allows people to resolution
[17:16]	Arawn Spitteler:	Budgetting different sides of the studio
[17:16]	Moon Metty:	yes, i think so too
[17:16]	Opensource Obscure:	a) for me tho. :D (sorry for delay)
[17:17]	Sindy Tsure:	not for me - sl is as happy as it's been since i joined.. i'd like to see the feature set expand some
[17:17]	Charlette Proto:	I find a decen roadmap (constantly appended and prioritised) is the best way to make progress based on informal forums
[17:17]	Moon Metty:	with more bugs out of the way, it will shift to b)
[17:17]	Sindy Tsure:	except svc-22
[17:17]	Andrew Linden:	Ok, that is all I had on my agenda. The table is open if someone has an issue or questions.
[17:17]	xstorm Radek:	what about doing a beta server again to do a test run first ? and try it on just some people ?
[17:17]	Aaron23 DeCuir:	yes, - it should alternte - a,b,a,b,a,b
[17:17]	Yukinoroh Kamachi:	I want to talk about hovering buoyancy and llSetBuoyancy
[17:17]	Arawn Spitteler:	Beta Server Parties, to see how the software performs under stress.
[17:17]	Yukinoroh Kamachi:	I guess I could be nicknamed Mr. Buoyancy
[17:18]	Moon Metty:	i just came from aditi, my benchmark runs 50% faster on the new 64 bit server :D
[17:18]	Ardy Lay:	I agree. When you can see over the bug pile it's time to start bringing in new toys.
[17:18]	Andrew Linden:	xstorm, are you talking about the "Early Adopter" program we had for the Havok4 transition?
[17:18]	xstorm Radek:	yes
[17:18]	Simon Linden:	Moon - do you know what server version you ran it on?
[17:18]	Neuro Linden:	Moon: that upgrade process should be finished RSN on agni
[17:18]	Charlette Proto:	meeting aganda and a digest of previously tabled issues helps to actually deliver anything beyond small tweaks and fixes
[17:18]	Moon Metty:	well, it changed while i was there
[17:18]	Andrew Linden:	Woot! We expect some improvements on 64-bit, but I wasn't aware of any known benchmarks that show it.
[17:19]	Moon Metty:	but it's a 1.26 server
[17:19]	Moon Metty:	without the altered scheduler
[17:19]	Arawn Spitteler:	That's an idea, let Sim Owners pay for the upgrades, and see which way they really swing, when it effects their popularity
[17:19]	Ardy Lay:	1.26.0.110077 ?
[17:19]	Moon Metty:	i think so Ardy
[17:19]	Andrew Linden:	Neuro, what do you mean by RSN?
[17:19]	Neuro Linden:	Hearing about 'real world' (if you'll pardon the pun) performance improvements is awesome
[17:19]	Simon Linden:	Some old performance work is finally going out in 1.26
[17:19]	Andrew Linden:	oh, "real soon now"
[17:19]	Neuro Linden:	Andrew: Real Soon Now :)
[17:19]	Neuro Linden:	yeah
[17:19]	xstorm Radek:	lol
[17:19]	Neuro Linden:	we're blasting through hundreds of servers each night
[17:20]	Charlette Proto:	good sign Simon (UXIG work goes into a drawer)
[17:20]	Andrew Linden:	Ah, perhaps some of the performance increases have to do with Simon's work therein.
[17:20]	Moon Metty gave you Mono speed-03.
[17:20]	Andrew Linden:	Neruo, perhaps you have some details of the 64bit upgrade process to share?
[17:20]	Simon Linden:	The big win was in our internal xml parser ... which will show up as faster rezzing, smoother region crossings and other cases were new data comes into a region
[17:21]	Arawn Spitteler thought Yuki's issue settled some time ago, but it seems to've opened again: Have you the URL for the Jira, Yuki?
[17:21]	Neuro Linden:	not too much, we're laying the foundation for 64-bit everywhere by putting a more modern, 64-bit aware version of the OS onto all the simulators and other machines
[17:21]	Yukinoroh Kamachi:	yes
[17:21]	Charlette Proto:	smoother region crossings? you must be talking 1.26 not what I see
[17:21]	Yukinoroh Kamachi:	two bugs actually
[17:21]	Yukinoroh Kamachi:	http://jira.secondlife.com/browse/SVC-1792
[17:21]	Neuro Linden:	the main problem up until now is that the servers have 64-bit capable CPUs, but they were running a 32-bit OS
[17:21]	Yukinoroh Kamachi:	this one concerns llSetBuoyancy
[17:22]	Yukinoroh Kamachi:	http://jira.secondlife.com/browse/SVC-2013
[17:22]	NonSmokerGirl Tulip:	oops my apologies for squishing
[17:22]	Neuro Linden:	once that upgrade process is complete, the path is open for 64-bit simulator software
[17:22]	Sindy Tsure:	simon, i had a question on the xml stuff. does that help when rezzing something from inventory, too?
[17:22]	Yukinoroh Kamachi:	and this one is about buoyancy when you just float in normal hover
[17:22]	Arawn Spitteler suggests NonSmokerGirl not click that box
[17:22]	Simon Linden:	Sindy - yes, it should
[17:22]	Sindy Tsure:	sweet
[17:22]	xstorm Radek:	a 64 bit system and server is much better in many ways
[17:22]	Andrew Linden looks at SVC=2013
[17:22]	Neuro Linden:	I'd need to double check the status, but the playbook has us completing that process this week
[17:23]	Sindy Tsure:	etchification?
[17:23]	Sindy Tsure:	can residents tell if a region's been done?
[17:23]	Yukinoroh Kamachi:	I am very concerned about these two bugs as it broke a 3 months worth development on my side,
[17:23]	Neuro Linden:	and in talking to prospero (sindy: yes), barring any reason to throw another 1.25 release out there, we can press to 1.26
[17:23]	Andrew Linden:	Neruo I thought I saw discussions about the 64bit upgrade pushing out to next Friday.
[17:23]	Yukinoroh Kamachi:	every single day I'm wishing for a move, but nothign moves
[17:24]	Neuro Linden:	Andrew: I may have missed that; there has been some issues with servers in san francisco that stalled us there
[17:24]	Neuro Linden:	and we have change control policies that generally stop us making production changes over a weekend, so yes, that delay may be necessary
[17:24]	Neuro Linden:	(our own self-imposed policies, i should add)
[17:24]	Andrew Linden:	Ok, SVC-2013 is very hard to fix. It is not on my short list of things to tackle this quarter.
[17:25]	Andrew Linden reads SVC-1792
[17:25]	Sindy Tsure:	latest published etchification schedule: http://status.secondlifegrid.net/2009/02/18/post507/
[17:25]	Yukinoroh Kamachi:	half of SVC-1792 could be fixed right away IMO.... put a check in whatever code drags the avatar to the ground, check the value of buoyancy
[17:25]	Neuro Linden:	i love when our residents know more about our schedules than I do :)
[17:26]	Arawn Spitteler teaches classes that way
[17:26]	Sindy Tsure is trying to save you time so you can work on that bear!!
[17:26]	Neuro Linden:	i should outsource that, huh :)
[17:26]	Sindy Tsure:	lol
[17:27]	Andrew Linden:	SVC-1792 is not about llSetBuoyancy() so much as it is about how Buoyancy is broken when used on avatars (via attachments)
[17:27]	Sindy Tsure:	oh.. is that related to the other bug where gravity is getting applied twice/frame or something..
[17:27]	Andrew Linden:	SVC-1792 may be a dupe of SVC-2013. Likely, but I can't be sure without looking into it.
[17:27]	Yukinoroh Kamachi:	andrew : it's not
[17:28]	Yukinoroh Kamachi:	there's another force that drags the avatar to the ground, perhaps to make sure they stick to the ground or something, when they land
[17:28]	Sindy Tsure:	http://jira.secondlife.com/browse/SVC-2441 maybe?
[17:28]	Andrew Linden:	Well, it is related to avatar motion and it sinking when you wouldn't expect it to. Based on what I know about the problem(s) they seem very related.
[17:28]	Yukinoroh Kamachi:	that force is applied even when you are supposed to flaot with llSetBuoyancy, but it shouldn't
[17:28]	Arawn Spitteler uses llSetBuoyancy(1.0) to hover at sea level
[17:28]	Simon Linden:	If I remember some of that, there's some ugly problems about when our code applies force to keep an object buoyant vs. when the Havok physics engine does gravity
[17:29]	Simon Linden:	It can occasionally get one step out-of-sync, and you get the drift
[17:29]	xstorm Radek:	they float in to the air lol
[17:29]	Andrew Linden:	correct, and as a matter of fact we have an avatar-special buoyancy that is supposed to prevent the avatar from sinking when hovering... but sometimes it does not work right.
[17:29]	Charlette Proto:	does this say that we really don't have a roadmap and digest of the issues/bugs/improvements being worked on
[17:30]	Yukinoroh Kamachi:	o If you wear the item while standing and not moving; although the script has cancelled any remaining speed and buoyancy equals 1, at destination, the simulator will drag you to the ground.
[17:30]	Yukinoroh Kamachi:	this part could be fixed easily I think
[17:30]	Andrew Linden:	I've gone over the math and have inserted debugging info around where that code takes effect... the problem is due to some weird (undocumented) effect from when Havok tries to "deactivate" the avatar body (at least, that is my theory from what I found)
[17:30]	Yukinoroh Kamachi:	because it's caused by a code that'll drag the avatar to the ground after he's landed
[17:30]	Yukinoroh Kamachi:	probably a code to ease walking up slopes or something
[17:31]	Yukinoroh Kamachi:	put a check on it, check the value of buoyancy and don't apply that code if it's not null
[17:31]	Rex Cronon:	it would be interesting to walk on walls:)
[17:31]	Neuro Linden:	hurrah for undocumented effects!
[17:32]	xstorm Radek:	is this part of the same problem i have been seeing people sink below the ground when not doing any thing till they get stuck and crash ?
[17:32]	pickle Flagon:	i have a spider that can walk on walls
[17:32]	Simon Linden:	Both Andrew and I have spent time on it ... if it was easy, we'd have fixed it by now :(
[17:32]	Yukinoroh Kamachi:	rex : I have given up working on anything physical... I am mad about my 3 months worht developmetn that are loss
[17:32]	Andrew Linden:	Ok, your vote for SVC-2013 and -1792 has been noted. I'll see if I can find the time to look into that again in some reasonable amount of time.
[17:32]	Rex Cronon:	u might be able to use np and fake it
[17:33]	Andrew Linden:	The good new is that there was some internal discussions kicked up again about FPS-style client-side prediction...
[17:33]	Rex Cronon:	np=non physical
[17:33]	Andrew Linden:	if we ever try to accomplish that, then the avatar control code will get a lot more scrutiny and we might actually find some resources to focus on the bugs there.
[17:33]	Yukinoroh Kamachi:	I can tell you anyway that until I find a job in RL here in Japan, I am willing to wear out this chair every Friday morning
[17:34]	Sindy Tsure:	not sure i understand.. thought the client already did prediction.. no?
[17:34]	xstorm Radek:	:-)
[17:34]	xstorm Radek:	sounds nice
[17:34]	Andrew Linden:	Sindy, not really for the avatar. Not "pre-emptive prediction".
[17:34]	Sindy Tsure:	oh.. ok
[17:35]	Charlette Proto:	what about something like a fuzzy system logic > a force is acted upon or else nothing happens instead trying to constanltly counteract different forces acting on objects
[17:35]	Simon Linden:	For example, when you push the arrow to go forward, there's a message sent to the sim, it sends updates, then you see yourself move
[17:35]	Andrew Linden:	that ^^^^ si the current way it works
[17:35]	Yukinoroh Kamachi:	In my city, there was a major problem with a dam, some people built their cottages when the dam leverl was unusually low... as a result the city cannot set the dam level at the normal height now... see what I mean ? it's time to fix 1-year old issues until too many people develop with the new settings
[17:35]	xstorm Radek:	you can turn on updats to see that happen
[17:36]	Arawn Spitteler'd like Yuki to look at SVC-2931
[17:36]	Sindy Tsure:	any news on maybe moving camera stuff to the client too?
[17:36]	Sindy Tsure:	phone... scuse.. brb
[17:37]	Yukinoroh Kamachi:	I'd very much like the Lindens to just test my two bugs with the test scripts I've taken the time to put in the bug description
[17:37]	Yukinoroh Kamachi:	looking at it > arawn
[17:37]	Andrew Linden:	Hrm... no new news on that Sindy. I thought we had some people on the viewer side all fired up about it a while ago, but I haven't heard much about it.
[17:37]	xstorm Radek:	but this comes right back to having a beta server to do a pretest on befor more people get there work going down the tubes
[17:38]	Yukinoroh Kamachi:	the whole need of teleporting chairs would lessen if avatar rotation was saved with landmarks
[17:38]	Andrew Linden:	xstorm, the aditi grid provides some beta exposure, but I agree that it is often late and not long enough
[17:39]	Yukinoroh Kamachi:	built a mall, set the LM at the entrance, but sinve av always face north-east, every single customer was leaving the mall on arrival
[17:39]	Charlette Proto:	pickle grab the vacant chair on the right of the zebra (if you like)
[17:39]	NonSmokerGirl Tulip:	oh sure, use me as a landmark
[17:39]	pickle Flagon:	ty very much
[17:40]	Sindy Tsure:	turn left at the zebra?
[17:40]	Andrew Linden:	Yukinoroh, someone was working on that problem I think... I saw some discussions about that go by either email or IRC chatter.
[17:40]	Andrew Linden:	That is, getting the avatar to point the way it is wanted on TP arrival
[17:40]	NeoDrake Xue gave you Neo's Secrets Bear!!.
[17:40]	Simon Linden:	That would be a nice feature
[17:41]	Yukinoroh Kamachi:	I had to use a TP chair in a skybox to fix that... but it was messy, people tped there and thought there was nothing so they just left
[17:41]	Sindy Tsure:	a tp always points you NE? i've never noticed that on people coming into my place..
[17:41]	xstorm Radek:	that may fix a number of bugs later on
[17:42]	Sindy Tsure:	maybe a "welcome! turn around please!" sign would help
[17:42]	Charlette Proto:	the whole thing of chair/lift based TPs is a joke and always has been
[17:42]	Ardy Lay:	Does using a "landing point" face arrivals a specic firection?
[17:42]	Ardy Lay:	direction
[17:42]	Yukinoroh Kamachi:	Not mentioning that scripts at that height stop working after some time, this includes sensors and timers
[17:42]	Arawn Spitteler:	Have them arive southwest of the directory?
[17:42]	Charlette Proto:	script based direct TPs are a must
[17:42]	Sindy Tsure:	the landing point setting just has a location, ardy.. in about land in the options tab
[17:43]	Rex Cronon:	scripts with sensor stop working ant wht height?
[17:43]	Rex Cronon:	at what*
[17:43]	Yukinoroh Kamachi:	well, I have a landmark giver in my TP skybox... and every few weeks I have to go reset it
[17:43]	Sindy Tsure:	that's come up a bunch of times here, charlette! i always whine about griefer potential, though.. it's trickier than it sounds, i think
[17:44]	xstorm Radek:	a Landmark with rotaion number do not sound to hard to add to a line of code
[17:44]	Yukinoroh Kamachi:	so really, fix old bugs instead of working on new features
[17:44]	NonSmokerGirl Tulip:	griefers=croc-bait
[17:44]	Andrew Linden:	xstorm, I think adding look-at direction to the internal TP protocols is the hard part.
[17:44]	Sindy Tsure:	sadly, i have no crocs in my pool..
[17:45]	Arawn Spitteler's box is better than crocs
[17:45]	NonSmokerGirl Tulip:	:)
[17:45]	Sindy Tsure:	what did the box do again?
[17:45]	xstorm Radek:	well you have a set of code that tells it to be right side up
[17:45]	xstorm Radek:	and compes points for the av to see
[17:46]	Ardy Lay:	Hatbox
[17:46]	xstorm Radek:	why not have map point be a part of the LM ?
[17:46]	Andrew Linden:	yes, the right-side up code is a built-in-default. I think the NE direction is just an initialization default on arrival.
[17:46]	Arawn Spitteler:	That box is simple enough for basic scripting. I used it as something for touch_end
[17:47]	Andrew Linden:	In order to accept a specified look-at direction we'd need to ADD info to existing protocols. some of our protocols are kinda hard to add stuff too
[17:47]	Andrew Linden:	I'm not sure that TP is using the old rigid protocols. It has been a while since I've worked on TP code, and I know a lot has changed there.
[17:47]	Charlette Proto:	yah back to the Andreyko Soyuz rocket kodebase
[17:48]	Arawn Spitteler:	A lot of coding could stand to be more fluid
[17:48]	Sindy Tsure:	more fluid? so.. would that be a fix or a feature?
[17:48]	Andrew Linden:	Meanwhile, adding info to landmarks also requires code changes in lots of places, and we have to support legacy landmarks for a long time.
[17:48]	Simon Linden:	I think that's the problem with the landing point code as well
[17:49]	xstorm Radek:	but can this not be used later on to help when people start building more in the sky then they do not such at at 13000 and up ?
[17:49]	xstorm Radek:	do now
[17:49]	Andrew Linden:	I didn't understand that question xstorm. Was that directed at me?
[17:50]	Andrew Linden:	Where did the 13k number come from?
[17:50]	Arawn Spitteler's working around WEB-814 just after class: We already have high altitude TP poinnts, that aren't supported by Events. What should concern the H-4 team, is that the design philosophy didn't make it a concern for the H-4 Team. Some new philosophy needs be dreamed up.
[17:50]	xstorm Radek:	if you add direction to the landmark it can help in building in the sky
[17:51]	Yukinoroh Kamachi:	well to talk more about my land issue with that
[17:51]	xstorm Radek:	i build very high
[17:51]	Sindy Tsure:	there's a jira for the event thing.. talking about it on the help island channel a few nights ago
[17:51]	Yukinoroh Kamachi:	the SIm owner requested that all parcels have an entrance to the streets
[17:51]	Charlette Proto:	back to the same thing no roadmap or am i a design addict
[17:51]	Sindy Tsure:	you build over 4km up?
[17:51]	Arawn Spitteler:	Over 768 up
[17:51]	xstorm Radek:	yes
[17:51]	Yukinoroh Kamachi:	my parcel being in the north-west 1/4 corner, I made my entrance point near the center
[17:52]	Simon Linden:	Charlette -- I'm not sure what you're asking for, but I don't know of any public info that discusses what we plan on doing in the future
[17:52]	Sindy Tsure:	thought LL raised the limit from 768 to 4096 last year.. maybe as part of h4
[17:52]	Charlette Proto:	why is it that all team work with onthespot bug fixing and nobody takes on the development direction/vision onboard
[17:52]	Andrew Linden:	Ah yes, I see your problem with the NE default Yukinoroh. Not all arrival points will be looking the optimal direction if you are lining all sides of the streets.
[17:53]	Charlette Proto:	sorry I'll keep it to myself
[17:53]	Andrew Linden:	Charlette, you are looking for a roadmap? To what? server features?
[17:53]	Sindy Tsure:	hey, arawn.. touch this box!
[17:53]	xstorm Radek:	or in the sky you may endup with a off set that drops you off target
[17:54]	Arawn Spitteler examines Linden Roadmap: Wander aimlessly, which will bring you to any point within the square root of the number of your steps
[17:54]	Andrew Linden:	I don't think LL has ever supplied a long-term roadmap that we put much confidence in. No one here has such a roadmap. I think most planning goes out three to six months. Then it all gets too fuzzy .
[17:54]	Charlette Proto:	absolutely, and the codebase, if one was to contribute the whole JIRA would have to be read and each issue alalysed before one would give up
[17:55]	Andrew Linden:	There are lots of stuff we want to do that is out beyon 6 months.
[17:55]	Yukinoroh Kamachi:	any plans for open-sourcing the server code soon ?
[17:55]	Yukinoroh Kamachi:	this could give some old bugs a nice push
[17:55]	Andrew Linden:	But we are very confident in saying that anything we plan to do beyond 6 months is probably not going to be the most important thing that needs to get done when we get there.
[17:55]	Charlette Proto:	yeah I know fix this bug, change this default and the whole thing is just getting out of hand - all teams are doing this
[17:55]	Andrew Linden:	Er... I'm very confident...
[17:55]	Sindy Tsure:	with the growth SL had over the last few years, that's sorta understandable.. was shouts of "10k online at once is too many" when i joined..
[17:56]	xstorm Radek:	lol
[17:56]	Yukinoroh Kamachi:	much growth is due to bots though
[17:56]	Andrew Linden:	Yukinoroh, there as been almost zero discussion about open-sourcing the SL server internally, for the last year or so
[17:56]	Yukinoroh Kamachi:	it's hard to find real people
[17:56]	Simon Linden:	There are teams working on more long-term projects to solidify the asset system, for example. It's not just immediate bug fixing
[17:56]	Sindy Tsure:	llnet!
[17:56]	Charlette Proto:	good point Yuki
[17:57]	Charlette Proto:	free accouts are good for numbers but nothing else
[17:57]	Yukinoroh Kamachi:	the traffic system should be scrapped
[17:57]	Andrew Linden:	The open-source SL simulator was a topic of discussions when Philip and Cory were directing the company. It seems to have lost proponents.
[17:57]	Sindy Tsure:	yeah.. except buying stuff.. lots of basic accounts do that
[17:57]	Opensource Obscure:	I disagree Charlette but i dont think this is ontopic
[17:57]	Yukinoroh Kamachi:	word of mouth is what makes good avertising in this internet era
[17:57]	Yukinoroh Kamachi:	scrap the traffic system and you'll get rid of most bots
[17:57]	xstorm Radek:	Free account are starting to hurt the servers and making problems with sim owners
[17:57]	Simon Linden:	We're going to keep free accounts, in any case
[17:57]	Andrew Linden:	I think the traffic system should be scrapped.
[17:58]	Moon Metty:	hey, i have a free account
[17:58]	Sindy Tsure:	if i had to pay to try sl, i would not have tried it
[17:58]	Andrew Linden:	But 'search' should be much improved.
[17:58]	Rex Cronon:	come on. no all free accounts r bad
[17:58]	Neuro Linden:	Bear in mind that while we haven't been making noises about open sourcing the simulator software, there has been a lot of work to open up the protocols themselves
[17:58]	xstorm Radek:	then give free account less things like take away building from free accounts
[17:58]	Yukinoroh Kamachi:	I have a free account too
[17:58]	Yukinoroh Kamachi:	but I make 16,000L$ a month profit
[17:58]	Sindy Tsure read of grid-to-grid IMs last week.. using some sort of proxybot
[17:59]	Neuro Linden:	have a look at http://wiki.secondlife.com/wiki/MMOX
[17:59]	Charlette Proto:	even if you juast had a deposit and not monthly fee you would reduce the whole problem to insignificant
[17:59]	Arawn Spitteler:	If Traffic were to count only premium accounts, at least the bots would start paying.
[17:59]	Rex Cronon:	for a few bad apples u want to punish everybody els:(
[17:59]	Andrew Linden:	Indeed Neruo, and I have to admit some of the prerequisites for open-soruceing the SL simulator are being worked on.
[17:59]	Andrew Linden:	For example, pulling stuff like money transactions out of the simulator code and moving them to be web services.
[17:59]	Charlette Proto:	not sure Arwn bots are like spam - charg for everything and they disappear
[17:59]	xstorm Radek:	if you make that much on a free account then you can pay the 72 a year and stop living off the ones paying for a account
[18:00]	Charlette Proto:	charge*
[18:00]	Opensource Obscure:	this is going to happen soon isnt it?
[18:00]	Opensource Obscure:	(pulling stuff like money transactions out of the simulator code and moving them to be web services.)
[18:00]	Yukinoroh Kamachi:	you want to avoid "money grow on trees" dreamers to move to other grids huh ;)
[18:00]	Sindy Tsure:	not gonna happen, charlotte
[18:00]	Charlette Proto:	precisely
[18:00]	Andrew Linden:	Opensource, I don't know the status of that project, and I've been meaning to ask the developers involved.
[18:00]	Opensource Obscure:	thanks.
[18:01]	Andrew Linden:	The project has been a long time coming, much longer than we expected.
[18:01]	Opensource Obscure:	do you know which visible changes will users see - if any?
[18:01]	xstorm Radek:	no i see it as griefers coming in on free accounts just to take away my time that i pay for when the griefers fill up the AR's
[18:01]	xstorm Radek:	taking time for more importent things
[18:01]	Andrew Linden:	I don't think there will be many visible changes. The only ones I would expect would be to see more web interfaces to your SL money and inventory.
[18:02]	Sindy Tsure:	that doesn't mean all free accounts are griefers, xstorm.. far, far more than 1/2 the grid is not mainland
[18:02]	Andrew Linden:	There shouldn't be any visible changes in the usual SL client interface.
[18:02]	Charlette Proto:	yep and governance is so stretched they have resourted to dishing out suspension based on bogus ARs
[18:02]	Andrew Linden:	Ah, we seem to have run out of time (for me anyway).
[18:02]	xstorm Radek:	this is no longer a working system but a baby sitting sim for to many free account that are no more then a bunch of leaches
[18:02]	Andrew Linden:	You are all welcome to stick around and continue discussions
[18:02]	Sindy Tsure:	drat.. forgot to steal andrews watch battery again..
[18:02]	Andrew Linden:	but I'm going to have to go very soon.
[18:03]	NonSmokerGirl Tulip:	leaches? L(
[18:03]	NonSmokerGirl Tulip:	:(
[18:03]	Moon Metty:	ok Andrew, thank you
[18:03]	Sindy Tsure:	oh.. did i hear we can tell if a sim's been etchified?
[18:03]	Moon Metty:	and Simon of course :)
[18:03]	Yukinoroh Kamachi:	well, at least the most common crash-orbiting griefing has been gone, that I'd like to raise my hat
[18:03]	NonSmokerGirl Tulip:	yes thanks andrew
[18:03]	Sindy Tsure:	and, if so, how?
[18:03]	Arawn Spitteler:	Oh, something we should have, is PRIM_ROT_LOCAL, for the llSetLinkPrimitiveParams
[18:03]	Neuro Linden:	sadly there's no way for a resident to detect that
[18:03]	Andrew Linden:	Neuro, is it possible to tell if a sim host has been etchified?
[18:03]	Rex Cronon:	i think that bots used for trafic r the leaches:)
[18:03]	Sindy Tsure:	ok.. ty
[18:03]	Neuro Linden:	etchified sims run the same simulator code as non-etchified ones
[18:04]	Sindy Tsure:	is the update going by host number?
[18:04]	Neuro Linden:	it's going by server
[18:04]	Arawn Spitteler:	Griefers have more personality, than the Camp-Bots
[18:04]	Ardy Lay:	They should all be done by Monday morning I think Vektor said.
[18:04]	Sindy Tsure:	k. ty
[18:04]	Simon Linden:	Neuro - do you know if we've seen any changes in sim crash rates on etch?
[18:04]	Sindy Tsure:	and ty andrew & simon!
[18:04]	Arawn Spitteler:	What's Etchification?
[18:04]	Yukinoroh Kamachi:	perhaps something you could point out about bots
[18:04]	xstorm Radek:	yes and there are to many bots stepping all over the TOS
[18:04]	Neuro Linden:	Simon: good question, I haven't looked at that yet
[18:04]	Simon Linden:	Thanks everyone for coming
[18:04]	NonSmokerGirl Tulip:	thank you all for having us
[18:04]	Neuro Linden:	I certainly haven't seen as many crashes while on duty watching the grid
[18:05]	Rex Cronon:	bye everybody
[18:05]	Yukinoroh Kamachi:	hm. n/M
[18:05]	Sindy Tsure:	arawn: http://forums.secondlife.com/showthread.php?t=307304
[18:05]	Arawn Spitteler hopes Neuro can be a regular at this happy camp
[18:05]	Neuro Linden:	but that data is clouded by the mass of starting and stopping regions while the RRs are going on
[18:05]	Simon Linden:	Arawn - etichifiation is an upgrade to the OS the sims run on
[18:05]	Andrew Linden:	Perhaps Neuro holds regular office hours?
[18:05]	Neuro Linden:	and my regular shift happens while the RRs are going on :)
[18:05]	xstorm Radek:	sorry but i do not like the griefers using what we pay for
[18:05]	Neuro Linden:	Andrew: I don't, but I'm considering it
[18:05]	Simon Linden:	You can always visit other ones, like I do :)
[18:05]	Neuro Linden:	Prospero used to run office hours as a DNOC team member, but he's now moved to the release group, so no DNOC members run office hours atm
[18:05]	Opensource Obscure:	oh, please do it Neuro, i'd be a regular!
[18:06]	Neuro Linden:	Simon: indeed :)
[18:06]	Ardy Lay:	This table is a real shin-banger. :-)
[18:06]	Neuro Linden:	thanks Opensource :)
[18:06]	Arawn Spitteler'll set his box temp, before it gets returned.
[18:06]	Neuro Linden:	I really should, get some use out of my lodge in Longfellow
[18:06]	Sindy Tsure:	cya, folks!