Chat log from the meeting on 2019-06-04

Latest revision as of 12:25, 4 June 2019

[11:04] Ubit Umarov: gezzz got a thing i submited to FS jira looked at
[11:04] Ubit Umarov: and has a nice chat with Beq about that
[11:04] Arielle.Popstar @grid.hgluv.com:8302: was it a positive look?
[11:05] Ubit Umarov: https://jira.phoenixviewer.com/browse/FIRE-19734
[11:05] Ubit Umarov: that will impact day--- also Gavin.Hird
[11:07] Gavin.Hird @grid.xmir.org:8002: maybe
[11:07] Ubit Umarov: it will, its on ll base code
[11:07] Gavin.Hird @grid.xmir.org:8002: don't have the SL group code that FS has
[11:08] Ubit Umarov: it also does "double" or "triple" requests
[11:08] Ubit Umarov: all on "same" ms timestamp
[11:08] Gavin.Hird @grid.xmir.org:8002: I'll check it out and make a JIRA of it
[11:09] Ubit Umarov: beq is basicly going to paste that on ll jira
[11:09] Gavin.Hird @grid.xmir.org:8002: I bet
[11:09] Ubit Umarov: no need to bet.. she will :P
[11:10] Gavin.Hird @grid.xmir.org:8002: Dayturn has a JIRA too, though not particularly public
[11:10] Ubit Umarov: she actually tells that on that fs jira
[11:10] Ubit Umarov: well its a old issue.. you can see the date
[11:11] Ubit Umarov: ansariel just seen opensim and totally ignored it
[11:11] Gavin.Hird @grid.xmir.org:8002: aha
[11:11] Ubit Umarov: ( or other devs )
[11:11] Ubit Umarov: and it is a SL issue also lol
[11:11] Ubit Umarov: poor servers
[11:11] Ubit Umarov: its a big flood at login if you have a large notifications/invites list
[11:13] Ubit Umarov: and bad code.. it askes group profile per notification, even for the same group
[11:14] Gavin.Hird @grid.xmir.org:8002: issue noted, so I'll check it out
[11:15] Ubit Umarov: but well a report on fs jira with word "opensim" from 2016 did got a big push into resolve :)
[11:17] Ubit Umarov: well and those notifications are the sum of notifications for all grids
[11:17] Ubit Umarov: since they are stored on disk on a single location
[11:17] Ubit Umarov: ( i was told.. )
[11:18] Gavin.Hird @grid.xmir.org:8002: This is caused by the persistent notifications being loaded from disk and the setup of the UI around them.
[11:18] Ubit Umarov: yeap
[11:18] Gavin.Hird @grid.xmir.org:8002: I don't understand that
[11:19] Gavin.Hird @grid.xmir.org:8002: because if they were stored on disk, they should already be delivered to the user
[11:19] Gavin.Hird @grid.xmir.org:8002: it is not like the viewer stores anything while not running
[11:19] Gavin.Hird @grid.xmir.org:8002: or not being logged in
[11:20] Arielle.Popstar @grid.hgluv.com:8302: too bad my s/l groups didnt work in opensim
[11:20] Ubit Umarov: dunno.. did not went look on viewers :)
[11:21] Ubit Umarov: well i do have 3 notifications here
[11:21] Gavin.Hird @grid.xmir.org:8002: maybe it sent the request multiple times over UDP in lack of ACK in hope at least one should reach the server
[11:21] Ubit Umarov: guess they are stored until user takes some action on then
[11:22] Gavin.Hird @grid.xmir.org:8002: if you get a notification in a session and just ignore it, it will not reappear in the next session
[11:22] Ubit Umarov: so you do need to be online to get them, but they stay there until you do something about it
[11:22] Ubit Umarov: it seems to
[11:22] Gavin.Hird @grid.xmir.org:8002: so I cannot think they are persisted anywhere
[11:22] Ubit Umarov: well they do
[11:23] Gavin.Hird @grid.xmir.org:8002: in which file are they persisted?
[11:23] Ubit Umarov: no idea
[11:23] Gavin.Hird @grid.xmir.org:8002: there you have it
[11:23] Ubit Umarov: just its a global file
[11:24] Ubit Umarov: err the fact i don't know only means that, and nothing about file existence :p
[11:24] Gavin.Hird @grid.xmir.org:8002: I got that, hehe
[11:24] Ubit Umarov: just i didn't look why would i having good viewers devs looking also??
[11:24] Ubit Umarov: :)
[11:25] Gavin.Hird @grid.xmir.org:8002: to my knowledge no such file exist
[11:25] Ubit Umarov: but at login you do get the notifications you did ignore last time, at least in fs
[11:26] Ubit Umarov: little thing at right top side
[11:26] Gavin.Hird @grid.xmir.org:8002: then they do something special
[11:26] Gavin.Hird @grid.xmir.org:8002: could be feature of the FS bridge
[11:27] Ubit Umarov: now the double/triple requests happen all the time we open a group chat etc
[11:27] Ubit Umarov: Beq told me its a historical issue on how ui works
[11:28] Gavin.Hird @grid.xmir.org:8002: as I said, it could be a primitive safeguard to ensure the request actually reach the server over UDP
[11:28] Ubit Umarov: that is not how lludp works
[11:29] Ubit Umarov: ( and all have dif sequence numbers, so server does sent 2 or 3 replies for same thing )
[11:29] Ubit Umarov: just a total waste
[11:29] Gavin.Hird @grid.xmir.org:8002: but if it has to do with the rendering of the UI xml, then tough luck to fix that in a hurry
[11:30] Ubit Umarov: beq passed that to ll :p
[11:30] Gavin.Hird @grid.xmir.org:8002: I can see in some log entries that it asks repeatedly for certain info
[11:30] Ubit Umarov: well thats a bug
[11:30] Ubit Umarov: i did add several traps a server side on that
[11:31] Ubit Umarov: map itens was another severe case
[11:31] Gavin.Hird @grid.xmir.org:8002: OZ said int he last dev meeting they have big problems with the renderer. Every time they make a change for EEP, it falls over with unexpected results, and they have to resort to trial and error to make things work
[11:31] Ubit Umarov: also parcel information.. a fs version did sent a request every time a mouse moved a micrometer over a parcel
[11:31] Gavin.Hird @grid.xmir.org:8002: that was fixed
[11:31] Ubit Umarov: hope so
[11:32] Ubit Umarov: i still did not removed the protection code
[11:32] Ubit Umarov: ( it just kills fs map :p )
[11:32] Ubit Umarov: well after a abuse, and for a timeout period
[11:32] Gavin.Hird @grid.xmir.org:8002: https://bitbucket.org/dayturn/dayturn/commits/71b11deeba275799b13edca4e12414de93a7001a
[11:33] Gavin.Hird @grid.xmir.org:8002: it was a fix Cinder submitted to SL in 2016
[11:33] Ubit Umarov: "only send when ShowLandTooltips is enabled" ... and parcel is not the same should be added
[11:34] Ubit Umarov: in fact that last part was enough for anycases
[11:35] Ubit Umarov: uff ansariel on that told me why you care about servers flood...
[11:35] Ubit Umarov: duhhh bc im a servers dev?? lol
[11:35] Gavin.Hird @grid.xmir.org:8002: :-)
[11:35] Andrew.Hellershanks @hg.osgrid.org: EEP?
[11:35] Ubit Umarov: think thats on fs jira somewhere :)
[11:36] Gavin.Hird @grid.xmir.org:8002: EEP is the new enironmental functionality in the SL viewer
[11:36] Andrew.Hellershanks @hg.osgrid.org: ok
[11:36] Ubit Umarov: yes more problems on horizont for us
[11:36] Ubit Umarov: lets see
[11:36] Gavin.Hird @grid.xmir.org:8002: where you can have environmental settings down to the parcel level
[11:36] Gavin.Hird @grid.xmir.org:8002: and
[11:36] Gavin.Hird @grid.xmir.org:8002: you will have roofs it does not rain through
[11:36] Andrew.Hellershanks @hg.osgrid.org: oh, nice.
[11:37] Ubit Umarov: for now im considering a thing 100% to ignore
[11:37] Gavin.Hird @grid.xmir.org:8002: plus the sun describes a path on the sky like the natural sun and not just a simple arch
[11:37] Ubit Umarov: parcel weather is just nonsense
[11:37] Bill.Blight @hg.osgrid.org: Ubit is going to need some heavy medication to make it through that one
[11:37] Gavin.Hird @grid.xmir.org:8002: hehe
[11:37] Ubit Umarov: :)
[11:37] Ubit Umarov: well need to wait for working viewers to see
[11:37] Gavin.Hird @grid.xmir.org:8002: let LL get through the mess first and stabilize the code
[11:38] Gavin.Hird @grid.xmir.org:8002: there is a project viewer
[11:38] Gavin.Hird @grid.xmir.org:8002: and the EEp server side is distributed to all of SL simulators
[11:38] Ubit Umarov: any viewer for opensim MUST keep working with current enviroment
[11:38] Gavin.Hird @grid.xmir.org:8002: I believe it will completely break Lightshare
[11:38] Ubit Umarov: yeah i think it will also
[11:39] Gavin.Hird @grid.xmir.org:8002: so FS will struggle with that
[11:39] Ubit Umarov: well for sure ill not add per parcel control EVER
[11:39] Ubit Umarov: at most per region control
[11:39] Gavin.Hird @grid.xmir.org:8002: at most? :-)
[11:40] Andrew.Hellershanks @hg.osgrid.org: I can think of a case where per parcel control might be useful but it would be a rare case.
[11:40] Ubit Umarov: yeap was if any suport :)
[11:40] Bill.Blight @hg.osgrid.org: I can think of many per-parcel uses
[11:40] Ubit Umarov: we do not need to have parcels as complete and independent as SL does
[11:40] Gavin.Hird @grid.xmir.org:8002: say you have a store on a parcel that needs to be in the Christmas mood
[11:40] Ubit Umarov: regions costs are a lot less
[11:40] Bill.Blight @hg.osgrid.org: caves that get dark, swamps that are foggy, so on and so forth
[11:40] Gavin.Hird @grid.xmir.org:8002: then it is easier to dress up that parcel with light, weather and all, than do the entrire region
[11:41] Ubit Umarov: caves and swaps are per region thing not necessary parcels
[11:41] Gavin.Hird @grid.xmir.org:8002: or if you want to simulate daylight conditions at a geo location
[11:41] Bill Blight: I'd disagree, you don't want to make your whole region a swamp,
[11:41] Gavin.Hird @grid.xmir.org:8002: hehe
[11:41] Ubit Umarov: we already have sound settings dependent on parcels
[11:41] Bill Blight: maybe only half the region a swamp
[11:42] Ubit Umarov: bad karma...
[11:42] Gavin.Hird @grid.xmir.org:8002: a laggy swamp
[11:42] Bill Blight: maybe
[11:42] Bill Blight: but valid uses
[11:42] Ubit Umarov: something else is needed to define zones.. not parcels
[11:43] Ubit Umarov: ( parcels already have 2 much potential conflicting settings )
[11:43] Ubit Umarov: but as i said.. need to wait and see
[11:44] Ubit Umarov: seems there is also a special control prim or something...
[11:44] Bill Blight: I mean yeah I see more usage in SL because land is so expensive you would want multiple environments on one region ..
[11:45] Ubit Umarov: yeap
[11:45] Gavin.Hird @grid.xmir.org:8002: right, but if you wanted to implement particle based rain (weather) across an entire region, it would be very laggy
[11:45] Ubit Umarov: there parcels are totally independent things
[11:45] Ubit Umarov: ppl even place things around them to set own enviroment etc etc
[11:46] Gavin.Hird @grid.xmir.org:8002: of course LL complicates things in that Envo settings have permissions and can be sold
[11:46] Gavin.Hird @grid.xmir.org:8002: they see a market for it
[11:46] Ubit Umarov: yeah i bet they do :)
[11:46] Ubit Umarov: damm mesh avatars... grrrr
[11:47] Ubit Umarov: well opensim viewer need to work as is, if possible with a way to test the new stuff
[11:48] Gavin.Hird @grid.xmir.org:8002: I am not going to try and implement EEp any day soon
[11:48] Ubit Umarov: not sure ll will make their viewer incremental as that, like they managed to so so far
[11:48] Gavin.Hird @grid.xmir.org:8002: maybe some of the improved shaders can be adopted
[11:49] Ubit Umarov: yeah guess a lot of that is good stuff like updated shaders..
[11:50] Bill Blight: the good I see in this, is maybe finally we will see an updated render engine in the viewer, but what that means for us , not sure .. I just know that the current renderer is sadly outdated ..
[11:50] Ubit Umarov: nahh fear it will be just another set of hacks
[11:51] Bill Blight: most likely , but it is nice to dream once in a while
[11:51] Ubit Umarov: :)
[11:52] Ubit Umarov: well when will the meeting start ??
[11:52] Ubit Umarov: andrew ?
[11:53] Andrew Hellershanks: You were talking about a number of issues. I thought it already had. I was letting you continue the discussion you had started.
[11:53] Ubit Umarov: :)
[11:54] Andrew Hellershanks: A day or two ago I received an email from the Firestorm bug track system that a patch I provided to the viewer has been accepted.
[11:54] Ubit Umarov: ( oh dear.. )
[11:54] Ubit Umarov: (hihihihi )
[11:54] Gavin.Hird @grid.xmir.org:8002: what did it do?
[11:54] Ubit Umarov: the map one?
[11:54] Gavin.Hird @grid.xmir.org:8002: the patch
[11:55] Andrew Hellershanks: The patch improves searching for regions via the map. They found out it also improves searching in SL so the patch is going to be included to the SL version of the viewer and not just the opensim version.
[11:55] Bill Blight: NICE
[11:55] Arielle.Popstar @grid.hgluv.com:8302: congrats Andrew
[11:55] Ubit Umarov: yeap that search is just a pain
[11:56] Andrew Hellershanks: I had already patched the OS code to improve the map search but it also needed a viewer side change.
[11:56] Bill Blight: Very cool
[11:56] Gavin.Hird @grid.xmir.org:8002: contribs are always welcome
[11:56] Ubit Umarov: we still have issues on that
[11:56] Ubit Umarov: specially with HG
[11:57] Andrew Hellershanks: If you had regions called "My Sandbox" and you search for "Sandbox" you wouldn't found the "My" version. Now you will be able to find it.
[11:57] Ubit Umarov: we had a similar issue elsewhere
[11:57] Ubit Umarov: now fixed
[11:58] Andrew Hellershanks: Good.
[11:58] Andrew Hellershanks: The hour is almost up. Any last minute topic before we call it a day?
[11:58] Ubit Umarov: like it there was a region sandbox, you could never find a "my sandbox"
[11:58] Ubit Umarov: or something like that
[11:58] Andrew Hellershanks: I think I just said that. :)
[11:59] Gavin.Hird @grid.xmir.org:8002: :-)
[11:59] Ubit Umarov: well but was somewhere else
[11:59] Andrew Hellershanks: np
[11:59] Ubit Umarov: ( and not ncessary that case )
[11:59] Ubit Umarov: bc i did fix that, not you :p
[11:59] Gavin.Hird @grid.xmir.org:8002: https://bitbucket.org/dayturn/dayturn/pull-requests/
[12:00] Gavin.Hird @grid.xmir.org:8002: or rather https://bitbucket.org/dayturn/dayturn/pull-requests/new
[12:00] Andrew Hellershanks: If anyone is interested, the bug report is FIRE-23591
[12:00] Ubit Umarov: dan was the one reporting that case to me
[12:00] Ubit Umarov: but that map search still needs more work
[12:00] Andrew Hellershanks: wow. I filed that in February. Four months. Not too bad I suppose.
[12:00] Ubit Umarov: mb your patch on viewers will allow us to improve
[12:02] Ubit Umarov: yeah one of several things i filed in 2016 got a answer yesterday :)
[12:02] Andrew Hellershanks: Yes, it will help. The patch to the OS code was something I applied last July.
[12:02] Ubit Umarov: fact is that basicly anything with the word "opensim" on fs jira was just ignored
[12:03] Andrew Hellershanks: Good. They may be slow but at least there is movement.
[12:03] Ubit Umarov: or got that automatic answer ...
[12:03] Ubit Umarov: now there is one dev that being also a SL user, does look to opensim issues
[12:03] Ubit Umarov: or looks more.. whatever
[12:03] Andrew Hellershanks: Good to know.
[12:03] Gavin.Hird @grid.xmir.org:8002: patch applied to dayturn, compiling right now
[12:04] Ubit Umarov: that one called ansariel, not only did not care about opensim, but was a lot agresssive against opensim
[12:04] Ubit Umarov: at least its what we could see on jira answers
[12:05] Ubit Umarov: in fact in one case or 2 i had to talk with jessica, and at least in one case jessica had to step in to clear the waters..
[12:05] Gavin.Hird @grid.xmir.org:8002: right
[12:06] Ubit Umarov: to reduce all the agresssion to opensim users
[12:06] Andrew Hellershanks: Too bad that has to happen. I still see signs of a lot of OS users who are using the FS viewer.
[12:06] Ubit Umarov: well that was the reason i did want to share with you my talk with Beq yesterday
[12:07] Gavin.Hird @grid.xmir.org:8002: supporting opensim in the viewer is a lot of additional work
[12:07] Gavin.Hird @grid.xmir.org:8002: hence the negativity
[12:07] Ubit Umarov: and guess will be a lot worse in future
[12:08] Andrew Hellershanks: One of these days we may have our own OS specific viewer and not have to rely on other 3rd party viewers.
[12:08] Leighton.Marjoram @grid.outandproud.life:8002: fingers crossed that happens at some point @Andrew
[12:08] Andrew Hellershanks nods
[12:09] Ubit Umarov: SL can do deep changes we can't follow because they are a single grid
[12:09] Andrew Hellershanks: Work did get started on a viewer but I think it has stalled. I hope it will pick up again soon.
[12:09] Gavin.Hird @grid.xmir.org:8002: LL pushed a massive code update to the Linux repo
[12:09] Ubit Umarov: we are lucky they want to have several regions code versions up at same time etc
[12:09] Gavin.Hird @grid.xmir.org:8002: said they had some hope to get a reduced functionality version up to run on Debian
[12:09] Ubit Umarov: so a lot of things is still negotiated per region
[12:10] Ubit Umarov: see for example the ViewerAssets cap
[12:10] Ubit Umarov: that is the same url for the entire region.. fs had it set only on login region and that was fine for sl
[12:10] Ubit Umarov: and deadly for us
[12:11] Ubit Umarov: i mean same url for entire grid
[12:12] Ubit Umarov: it does make sense at SL to move most of that to be set only at login as fs did by mistake ( now fixed by liru patch if i remember )
[12:12] Ubit Umarov: etc etc
[12:12] Arielle.Popstar @grid.hgluv.com:8302: use this maybe? https://github.com/diva/Grider
[12:13] Gavin.Hird @grid.xmir.org:8002: 10 years since last update?
[12:13] Arielle.Popstar @grid.hgluv.com:8302: yes i always wondered why she stopped work on it
[12:13] Ubit Umarov: yeah that seems very old and outdated
[12:13] Arielle.Popstar @grid.hgluv.com:8302: never found anything as to why
[12:14] Arielle.Popstar @grid.hgluv.com:8302: it would have resolved a lot of isses with hypergrid
[12:14] Ubit Umarov: its a variant of gridproxy ?
[12:14] Ubit Umarov: wel could solve some rotuer issues also
[12:14] Arielle.Popstar @grid.hgluv.com:8302: and no matter if outdated, point is whether the concept is something worthwhile
[12:15] Ubit Umarov: potentially
[12:15] Andrew Hellershanks: The description of that project isn't clear as to what is its purpose.
[12:15] Arielle.Popstar @grid.hgluv.com:8302: have to ask diva why they stopped work on it
[12:16] Arielle.Popstar @grid.hgluv.com:8302: or insitituting to opensim
[12:16] Ubit Umarov: textures udp only :)
[12:16] Gavin.Hird @grid.xmir.org:8002: Arielle, I got voice working on that PI
[12:16] Arielle.Popstar @grid.hgluv.com:8302: it performed well
[12:16] Arielle.Popstar @grid.hgluv.com:8302: in spite of all the inventory loading
[12:17] Gavin.Hird @grid.xmir.org:8002: hehe
[12:17] Arielle.Popstar @grid.hgluv.com:8302: which is something i was going to bring up here
[12:17] Gavin.Hird @grid.xmir.org:8002: it will be less bad the next time as it creates a load of records in assets
[12:17] Arielle.Popstar @grid.hgluv.com:8302: Gavin mentioned how my inventory was taking over on his console when i hg'ed over
[12:17] Ubit Umarov: inventory is not assets
[12:17] Gavin.Hird @grid.xmir.org:8002: only the inventory items are zapped
[12:18] Ubit Umarov: and HG is a speciall mess :p
[12:18] Arielle.Popstar @grid.hgluv.com:8302: well is it maybe Osgrid's HG 1.0
[12:18] Ubit Umarov: even suitcase
[12:18] Gavin.Hird @grid.xmir.org:8002: particularly visitors from OSG with rich inventories creates a storm of transfers
[12:19] Ubit Umarov: yeap Bill had issues with that on opensim party
[12:19] Arielle.Popstar @grid.hgluv.com:8302: but you said they would not be able to move to a HG 2.0?
[12:19] Gavin.Hird @grid.xmir.org:8002: your patch compiled Andrew, so I should leave to see if it works as expected
[12:20] Andrew Hellershanks: cool.
[12:20] Andrew Hellershanks: You can let me know on IRC or at the next meeting.
[12:20] Ubit Umarov: well even normal in grid inventory still needs more work
[12:20] Ubit Umarov: i think its a bit better now..
[12:20] Gavin.Hird @grid.xmir.org:8002: There might be an updated version out this week if I can get all the new libs sorted
[12:21] Ubit Umarov: full download if it is now faster no ?