IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 17 October 2009 (all times are UTC)
| 00:04 | yawasare has joined #ltsp | |
| 00:04 | <yawasare> hi there
| |
| 00:04 | <alkisg> Hi :)
| |
| 00:04 | <yawasare> is anyone in the house good with ltsp?
| |
| 00:05 | hey alkisg?
| |
| 00:05 | you there
| |
| 00:06 | <alkisg> !ask
| |
| 00:06 | <ltspbot> alkisg: "ask" is Don't ask to ask a question, simply ask it, and if someone knows the answer, they'll respond. Please hang around for at least 15 minutes after asking a question, as not everybody constantly monitors the channel.
| |
| 00:07 | <yawasare> i have ltsp server and thinclient boot into it, i have graphic face but anytime i put username and password it says fail am sure the userand password work even when i try to login on the main server it works put its doesnt work on client help guys
| |
| 00:08 | <alkisg> Try to run these commands on the server:
| |
| 00:08 | sudo ltsp-update-sshkeys
| |
| 00:08 | sudo ltsp-update-image
| |
| 00:08 | <yawasare> ok
| |
| 00:08 | <alkisg> Then reboot the client
| |
| 00:09 | alkisg has quit IRC | |
| 00:11 | alkisg has joined #ltsp | |
| 00:27 | yawasare_ has joined #ltsp | |
| 00:27 | <yawasare_> alkisg thanks men is working
| |
| 00:27 | it allowed login in
| |
| 00:28 | * cyberorg is in awe of alkisg's magic :) | |
| 00:31 | <yawasare_> so for this can i set applications that users can open and stuff? alkisg
| |
| 00:31 | <alkisg> Sorry I was afk. cyberorg, heh :)
| |
| 00:31 | yawasare has quit IRC | |
| 00:32 | <alkisg> yawasare_, I didn't understand this, can you rephrase it?
| |
| 00:32 | If you mean to install programs for the users to open, sure...
| |
| 00:33 | <yawasare_> yea
| |
| 00:33 | well what am doin is am in canada and setting computer up for school in Ghana West Africa, am donating to a school
| |
| 00:34 | <alkisg> Go to system > settings > synaptic package manager and install whatever you want there
| |
| 00:34 | Nice!
| |
| 00:35 | <yawasare_> what i wanted to know is that is i set it up and have number of clients can set app for users like when they login they can only open thse apps
| |
| 00:35 | if i
| |
| 00:36 | <alkisg> Yes, when the users log in it's like they're sitting on the server
| |
| 00:36 | So they see the same programs etc
| |
| 00:37 | <yawasare_> yea i see all programs on the main server, but i dont what thom to see all the programs
| |
| 00:38 | <alkisg> Why? Can you give an example?
| |
| 00:40 | <yawasare_> ok i have kids and adult using, i dont want the kids have access to all the programs on the main server
| |
| 00:40 | <alkisg> If you mean to hide the menus, sure, you can do that, but the programs will still be there
| |
| 00:41 | Access is better managed on a lower level, e.g. a kid shouldn't be able to run "gparted" because he wouldn't be a member of the admin group
| |
| 00:41 | Not because it would be hidden from the menus
| |
| 00:47 | <yawasare_> hiding will do it
| |
| 00:56 | well thank you soo much alkisg
| |
| 00:56 | am out
| |
| 00:56 | <alkisg> You're welcome
| |
| 00:56 | yawasare_ has quit IRC | |
| 01:30 | AndrewLe1 has quit IRC | |
| 01:36 | AndrewLee has joined #ltsp | |
| 03:05 | alkisg has quit IRC | |
| 03:10 | Sarten-X2 has joined #ltsp | |
| 03:12 | Sarten-X has quit IRC | |
| 03:29 | siki has joined #ltsp | |
| 04:45 | Sarten-X2 is now known as Sarten-X | |
| 05:20 | pmatulis has joined #ltsp | |
| 07:59 | siki1 has joined #ltsp | |
| 08:00 | siki has quit IRC | |
| 08:32 | garymc has joined #ltsp | |
| 08:51 | kusznir_ has quit IRC | |
| 08:53 | kusznir_ has joined #ltsp | |
| 10:19 | Egyptian[Home] has quit IRC | |
| 10:21 | Egyptian[Home] has joined #ltsp | |
| 10:33 | zamba has joined #ltsp | |
| 10:54 | japerry has quit IRC | |
| 10:57 | garymc has quit IRC | |
| 11:20 | johnny has left #ltsp | |
| 11:20 | johnny has joined #ltsp | |
| 11:20 | johnny is now known as Guest86579 | |
| 11:22 | Psifidotos has joined #ltsp | |
| 11:40 | Psifidotos has quit IRC | |
| 11:45 | Egghead has joined #ltsp | |
| 12:22 | japerry_cat has joined #ltsp | |
| 12:33 | japerry_cat is now known as japerry | |
| 12:36 | rjune has joined #ltsp | |
| 12:43 | Guest86579 is now known as johnny | |
| 12:57 | Selveste1___ has joined #ltsp | |
| 13:06 | japerry_cat has joined #ltsp | |
| 13:14 | japerry has quit IRC | |
| 13:14 | japerry_cat is now known as japerry | |
| 13:21 | Selveste1___ has quit IRC | |
| 13:52 | silver_hook has joined #ltsp | |
| 13:52 | Selveste1___ has joined #ltsp | |
| 13:52 | <silver_hook> Hullo.
| |
| 13:53 | I think I was talking to someone here that I could try to get him a VM (at Cyberpipe) for testing or building LTSP on Gentoo?
| |
| 13:53 | Could that person identify itself ...I'll try to persuade the main admin in Cyberpipe if you still need it.
| |
| 13:54 | <Ryan52> silver_hook: sounds like it could be johnny.
| |
| 13:55 | johnny: ^---
| |
| 13:55 | <silver_hook> Could be...
| |
| 13:55 | johnny: Still need that Gentoo VM?
| |
| 13:55 | <johnny> sure
| |
| 13:56 | <silver_hook> johnny: It was you, right?
| |
| 13:57 | <johnny> yes
| |
| 13:57 | altho.. my testers seem to have disappeared
| |
| 13:57 | :(
| |
| 13:57 | knipwim, where are you man?
| |
| 13:57 | <silver_hook> johnny: What exactly do you need ...so I know how to sell it to our cheif admin?
| |
| 13:58 | He uses LTSP anyway, so it shouldn't be too hard ...but he's also quite a security guy, so he needs some assurance regarding access.
| |
| 13:59 | <johnny> just a place to build/rebuild ltsp chroots mostly. ..but if i don't have anybody to also test it.. it's pointless
| |
| 13:59 | <silver_hook> Well, I could still get you a VM and you can use it when you get someone.
| |
| 14:00 | Still easier then trying to get a VM when you find a tester, I suppose :P
| |
| 14:00 | <johnny> i suppose
| |
| 14:03 | i was hoping to find somebody who wanted to share a box that they owned and let me remote in.. one that also had possibilities of testing terminal :)
| |
| 14:04 | <silver_hook> Care to give me a contact, so I can patch you through to the guy, if he approves it?
| |
| 14:04 | pmatulis has quit IRC | |
| 14:44 | knipwim_ has joined #ltsp | |
| 14:46 | vagrantc has joined #ltsp | |
| 14:57 | silver_hook has quit IRC | |
| 15:00 | knipwim has quit IRC | |
| 15:05 | knipwim has joined #ltsp | |
| 15:16 | knipwim_ has quit IRC | |
| 15:20 | johnny has left #ltsp | |
| 15:20 | johnny has joined #ltsp | |
| 15:20 | johnny is now known as Guest17409 | |
| 15:50 | rjune has quit IRC | |
| 16:09 | jammcq has joined #ltsp | |
| 16:09 | <jammcq> hello friends
| |
| 16:10 | heh
| |
| 16:12 | rjune has joined #ltsp | |
| 16:12 | J45p3r__ has joined #ltsp | |
| 16:13 | J45p3r__ has left #ltsp | |
| 16:23 | benGREEN has quit IRC | |
| 16:41 | lucascoala has quit IRC | |
| 16:45 | pmatulis has joined #ltsp | |
| 16:46 | greenmachine has joined #ltsp | |
| 16:53 | lucascoala has joined #ltsp | |
| 17:17 | rjune has quit IRC | |
| 17:39 | Egghead has quit IRC | |
| 17:48 | zamba has quit IRC | |
| 17:48 | zamba has joined #ltsp | |
| 17:53 | zamba has quit IRC | |
| 18:00 | zamba has joined #ltsp | |
| 18:42 | cliebow has joined #ltsp | |
| 18:52 | Egghead has joined #ltsp | |
| 19:12 | GGD has quit IRC | |
| 19:21 | Egghead has quit IRC | |
| 19:24 | Selveste1___ has quit IRC | |
| 19:27 | lejo has quit IRC | |
| 19:27 | Egghead has joined #ltsp | |
| 19:28 | lejo has joined #ltsp | |
| 19:45 | benGREEN has joined #ltsp | |
| 19:46 | lucascoala has quit IRC | |
| 19:54 | otavio_ has joined #ltsp | |
| 19:54 | otavio has quit IRC | |
| 20:19 | cliebow has quit IRC | |
| 20:30 | Guest17409 has left #ltsp | |
| 21:01 | nubae has joined #ltsp | |
| 21:03 | pmatulis has quit IRC | |
| 21:12 | vagrantc has quit IRC | |
| 21:15 | vagrantc has joined #ltsp | |
| 21:18 | <sbalneav> Evening all
| |
| 21:32 | <nubae> gosh's quiet today
| |
| 21:32 | hi vagrantc
| |
| 21:33 | sbalneav,
| |
| 21:33 | <sbalneav> Hey nubae
| |
| 21:34 | <nubae> u know I was thinking.... sabayon con do with some telepathy
| |
| 21:35 | think about it... if u wanted to duplicate profiles....
| |
| 21:35 | what are the dbus bindings like for sabayon...
| |
| 21:36 | * nubae goes look see... | |
| 21:36 | johnny has joined #ltsp | |
| 21:36 | vagrantc has quit IRC | |
| 21:37 | <nubae> hi johnny
| |
| 21:37 | <sbalneav> nubae: There are none.
| |
| 21:37 | <johnny> hello nubae
| |
| 21:38 | <nubae> hm.... welll it wouold be easy enough to add a method like duplicate profile
| |
| 21:39 | and signals of sabayon instances running on the network
| |
| 21:40 | of course.... remote profiling would be absolutely awsome
| |
| 21:40 | for help desks....
| |
| 21:40 | F-GT has quit IRC | |
| 21:40 | F-GT has joined #ltsp | |
| 21:41 | <sbalneav> I'd like to focus more on getting my apply profiles by group patches accepted upstream, and fixing crasher bugs, rather than more feeping creaturism :)
| |
| 21:42 | <nubae> well, you're not so much adding new features
| |
| 21:42 | as expanding existing ones
| |
| 21:44 | though I guess if there are no dbus bindings at all for anything... it is creating stuff from scratch
| |
| 21:44 | shame....
| |
| 21:45 | I mean, just adding dbus bindings, even forgetting about telepathy for the moment would be an important step
| |
| 21:45 | <jammcq> hey guys
| |
| 21:46 | <nubae> hi jammcq
| |
| 21:47 | <jammcq> hey
| |
| 21:47 | <sbalneav> Hey jammcq
| |
| 21:47 | <jammcq> I guess sbalneav is ignoring me
| |
| 21:47 | oooh, there he is
| |
| 21:47 | Scotty !!!!!!!!!!!!!!!!!!!!!!!!!!!!
| |
| 21:48 | <sbalneav> hey hey
| |
| 21:49 | johnny has left #ltsp | |
| 21:50 | johnny has joined #ltsp | |
| 21:51 | <johnny> test
| |
| 21:51 | <nubae> wow, this blog looks awfully familiar to nubae.com: http://robot101.net/
| |
| 21:52 | <johnny> nubae, ?
| |
| 21:52 | <nubae> now... I dont remember if I copied him :-)
| |
| 21:52 | style I mean
| |
| 21:52 | <johnny> nubae, there are some neat bindings happening with dbus..
| |
| 21:53 | <nubae> sabayon=
| |
| 21:53 | ?
| |
| 21:53 | <johnny> can you see me? what is my nickname?
| |
| 21:54 | <nubae> johnny
| |
| 21:54 | <johnny> ok..
| |
| 21:54 | <nubae> ok lets back up here cause Im confused
| |
| 21:55 | does sabayon have dbus bindings?
| |
| 21:56 | <sbalneav> No.
| |
| 21:56 | <nubae> ok, now then.... what has neat bindings with dbus?
| |
| 21:58 | <johnny> javascript
| |
| 21:59 | <nubae> hmmmm... that's a little generic dont u think :-) thats like me saying, there are dbus bindings for python
| |
| 22:00 | <johnny> desktop applications are going to be written in javascript
| |
| 22:00 | <nubae> thats even more generic
| |
| 22:01 | <johnny> as a web developer.. it is nice to see this...!!
| |
| 22:01 | <nubae> surely the world isnt going to sporadically turn to writing all applications in javascript overnight
| |
| 22:01 | <jammcq> johnny: you talking about chrome ?
| |
| 22:02 | <johnny> jammcq, no.. altho .. the engine this new js development will be taking place under is based off webkit
| |
| 22:02 | <jammcq> what engine is that?
| |
| 22:02 | <johnny> oh.. sorry. .v8
| |
| 22:02 | <nubae> well, we went from neat things happening with dbus, apparently in javascript applications for the desktop, though it is not apparent which ones yet
| |
| 22:03 | <johnny> nubae, it wasn't possible to do desktop apps in js UNTIL dbus
| |
| 22:04 | <nubae> well that isnt quite true, there have been many inter process communication frameworkds
| |
| 22:04 | the world just settled on dbus
| |
| 22:05 | <johnny> nubae, but no javascript bindings for them..
| |
| 22:05 | <nubae> but more importantly... the particular methods and signals that dbus would be implementing in a javascript app are what would be of interest
| |
| 22:06 | ie... it would be different for every app...
| |
| 22:06 | so I'm curious as to which app... u are referring to that does neat things with dbus
| |
| 22:07 | its just curiosity... I'm not putting u on the spot here or anything
| |
| 22:08 | johnny has left #ltsp | |
| 22:09 | <nubae> http://arstechnica.com/open-source/news/2007/09/using-the-tomboy-d-bus-interface.ars
| |
| 22:09 | that's a pretty neat dbus implemenation
| |
| 22:09 | something like that for sabayon would be nice (hint hint)
| |
| 22:12 | hmmm.... did I offend johhny? :-(
| |
| 22:13 | johnny has joined #ltsp | |
| 22:13 | <johnny> uggh.. my transport went psycho
| |
| 22:13 | it kept logging me in and out
| |
| 22:13 | going to use empathy for now..
| |
| 22:14 | <nubae> ah... ok... I thought I had offended u somehow :-)
| |
| 22:14 | <johnny> nubae: oddjob, that's something neat that *uses* dbus bindings.. nothing at all to do with javascript tho
| |
| 22:14 | <nubae> did u get the tommyboy link?
| |
| 22:14 | <johnny> no
| |
| 22:14 | <nubae> http://arstechnica.com/open-source/news/2007/09/using-the-tomboy-d-bus-interface.ars
| |
| 22:15 | thats a really nice simple implementation
| |
| 22:15 | <johnny> iso are you wanting to know about about just any old thing that actually implements dbus bindings?
| |
| 22:15 | <nubae> uggggh.... windows 7 to run on QT 4.6, say it aint so....
| |
| 22:16 | <johnny> err
| |
| 22:16 | sorry..
| |
| 22:16 | anything that actually communications decently over dbus?
| |
| 22:16 | nubae: do you have d-feet?
| |
| 22:17 | <nubae> sure
| |
| 22:17 | its great
| |
| 22:17 | whats even nicer though, for telepathy work is telepathy-inspector
| |
| 22:17 | <johnny> yes.. i even have that installed..
| |
| 22:17 | i also have that streamer one
| |
| 22:17 | <nubae> I'm a python coder though, so I'm really only interesting in python stuff
| |
| 22:18 | which is why dbus is pretty cool
| |
| 22:18 | <johnny> gajim could use some help.. :)
| |
| 22:18 | <nubae> its like cheating
| |
| 22:18 | <johnny> nubae: the reason i care about dbus in javascript btw.. is as a way to communication for even just the simple things..
| |
| 22:18 | <nubae> well nothing like directly accessing the telepathy dbus signals and methods
| |
| 22:19 | <johnny> like communication with the desktop notification thing
| |
| 22:19 | <nubae> well the reason I mentioned it before was that till I understood dbus properly I couldn't understand collaboration
| |
| 22:20 | <johnny> oh..
| |
| 22:20 | <nubae> since what is really happening under the hood is
| |
| 22:20 | dbus p2p
| |
| 22:20 | dbus from one program connecting to dbus on another program on another computer on another network
| |
| 22:20 | pretty cool stuff
| |
| 22:20 | <johnny> and that's why ltsp should use it..
| |
| 22:21 | now that hal is disappearing.. things will become easier imo..
| |
| 22:21 | <nubae> well u have to be pretty specific with how u create the dbus tubes...
| |
| 22:21 | I mean... u need a purpose
| |
| 22:21 | <johnny> a purpose.. yes we have them :)
| |
| 22:21 | <nubae> there isnt just a generic... lets dbus tube everything :-)
| |
| 22:22 | <johnny> many
| |
| 22:23 | making it so the remote session actually knows what hardware is represented
| |
| 22:23 | so the stupid power button will work too :)
| |
| 22:23 | err shutdown button
| |
| 22:24 | <nubae> anyway, I mentioned to sbalneav that sabayon would be a great contender for dbus tubing
| |
| 22:24 | but since it has no dbus methods or signals
| |
| 22:24 | <johnny> i don't think it needs that many..
| |
| 22:24 | <nubae> they'd have to be created first
| |
| 22:24 | <johnny> what it should do..
| |
| 22:24 | <nubae> which is kinda like recreating the wheel
| |
| 22:25 | true
| |
| 22:25 | <johnny> i think i agree with what sbalneav is doing
| |
| 22:25 | <nubae> doesnt need much
| |
| 22:25 | duplicate profile is the main one
| |
| 22:25 | I mean share profile
| |
| 22:25 | the dbus method would be duplicate profile
| |
| 22:26 | with a signal showing if a computer running sabayon is on the network
| |
| 22:26 | <johnny> i think it is just too soon to even need soemthing like that
| |
| 22:26 | sabayon has a hard enough problem not trying to kill itself while doing normal tasks..
| |
| 22:27 | but not all of that is the fault of sabayon, probably not most even
| |
| 22:27 | building profiles is hard.. :)
| |
| 22:27 | and making that desktop environment not freak out on you.. also hard :)
| |
| 22:27 | <nubae> yeah well, like all these things they must be worked on in tandem
| |
| 22:27 | <johnny> like using a one keyboard laoyout but making a session for another
| |
| 22:27 | <nubae> but its far easier to add a couple dbus bindings now, than later on down the line
| |
| 22:28 | <johnny> why does sabayon specifically need that feature you're talking about?
| |
| 22:29 | i think you're doing it at the wrong level
| |
| 22:29 | <nubae> hmmmm, why?
| |
| 22:29 | <johnny> having it a level up the stack would be nice..
| |
| 22:29 | making this happen..
| |
| 22:29 | <nubae> all I want is to have sabayon on one computer copy its profile to another computer
| |
| 22:30 | <johnny> why not just have it so you can easily copy ANY profile.. or any file.. ?
| |
| 22:30 | <nubae> right
| |
| 22:30 | that already happens
| |
| 22:30 | but I think thats not specific enough for the user
| |
| 22:31 | thats like saying, well... they can use ftp
| |
| 22:31 | <johnny> just copying the file doesn't help
| |
| 22:31 | it's what you do with it that makes a sabayon profile anything other than just a zipped up clump of files
| |
| 22:31 | <nubae> right, needs to integrate into the remote sabayon
| |
| 22:31 | thats why we use dbus
| |
| 22:32 | <johnny> hmm? exactly how will the other computer know what to do with this group of files?
| |
| 22:32 | <nubae> telepathy
| |
| 22:32 | <johnny> how?
| |
| 22:32 | <nubae> I mentioned before, what a dbus tube is right?
| |
| 22:32 | <johnny> describe the work flow
| |
| 22:32 | <nubae> us understand that?
| |
| 22:33 | <johnny> i know what the tube is.. but what exactly are you going to do with it
| |
| 22:33 | <nubae> normally... dbus is limited to one computer
| |
| 22:33 | <johnny> descripe the work flow
| |
| 22:33 | <nubae> it cannot jump across the network
| |
| 22:33 | <johnny> well you can ssh tunnel it i bet..
| |
| 22:33 | <nubae> well... telepathy allows it to jump across the network at will
| |
| 22:33 | <johnny> i know that
| |
| 22:33 | <nubae> u dont need to do that
| |
| 22:33 | telepathy takes care of the entire communications framework
| |
| 22:34 | <johnny> you're missing the point
| |
| 22:34 | <nubae> u dont have to worry about any of the low level stuff
| |
| 22:34 | <johnny> describe exactly how you imagine this working?
| |
| 22:34 | <nubae> what do u want me to describe to you, how telepathy works? the dbus bindings I imagine sabayon having?
| |
| 22:34 | <johnny> no
| |
| 22:34 | <nubae> what exacttly?
| |
| 22:35 | ok, lets back uo
| |
| 22:35 | up
| |
| 22:35 | a bit...
| |
| 22:35 | <johnny> tell me what the user is gonna see :)
| |
| 22:35 | <nubae> ok
| |
| 22:36 | sabayon right now allows me to create groups of users
| |
| 22:36 | and make a profile for groups of users
| |
| 22:36 | <johnny> sure
| |
| 22:36 | <nubae> right?
| |
| 22:36 | ok... but if I wanted to use your profile
| |
| 22:37 | well, a profile u created
| |
| 22:37 | ud probably have to email it to me
| |
| 22:37 | I might need to place it in some special dir
| |
| 22:37 | unzip it...
| |
| 22:37 | and a whole host of other things
| |
| 22:38 | i dunno
| |
| 22:38 | <johnny> well.. are you familiar with how sabayon is deployed now?
| |
| 22:38 | in a non ltsp system..
| |
| 22:38 | <nubae> however... if there was a dbus method in sabayon that simply duplicates the profile, it then sends it to the sabayon on the other computer via a dbus tube
| |
| 22:39 | ie... program A uses the dbus method on program B
| |
| 22:39 | on computer B
| |
| 22:39 | <johnny> but why would it have to be sabayon?
| |
| 22:39 | <nubae> because its just an example
| |
| 22:39 | this can be done with anything of course
| |
| 22:40 | but I just thought it would have been neat for sabayon
| |
| 22:40 | <johnny> right now.. an enteprise deployment would consist of an imaged system that gets deployed to all computers. this image would contain an http url so when folks login, it downloads the profile and they get it
| |
| 22:40 | <nubae> but since there are no dbus bindings the discussion is moot
| |
| 22:40 | sure,
| |
| 22:40 | <johnny> hmm.. i don't use bindings for what you're talking about
| |
| 22:40 | <nubae> could all be automated
| |
| 22:40 | <johnny> i use bindings as in dbus connection methods in programming languages
| |
| 22:41 | <nubae> why not, its perfect for it
| |
| 22:41 | <johnny> not the exposed dbus interfaces of a specific program
| |
| 22:41 | <nubae> forget ldap
| |
| 22:41 | tap into telepathy
| |
| 22:41 | <johnny> because that's not what people say when they say dbus bindings
| |
| 22:41 | hmm? you're confusing me
| |
| 22:41 | why would they not use ldap.. they have to use it for the actual auth anyways
| |
| 22:42 | <nubae> bindings are just the way to interface with dbus in a specific language
| |
| 22:42 | ie...
| |
| 22:42 | <johnny> if they happen to use ldap..
| |
| 22:42 | <leio> hey, anyone gotten any memtest to work over PXE recently? I get 0104 spam with all versions of memtest86 and memtest86+ I've tested
| |
| 22:42 | <nubae> python-telepathy
| |
| 22:42 | are the dbus bindings for python for telepathy
| |
| 22:42 | I'd imagine there is also javascript-telepathy
| |
| 22:43 | its basically a way to cheat and not have to use C
| |
| 22:43 | thats how I look at it
| |
| 22:43 | <johnny> ok.. yesi get it
| |
| 22:43 | <nubae> now... telepathy allows u to network dbus
| |
| 22:44 | <johnny> i'm just not imaging a real use for profiles outside of a standardized networks
| |
| 22:44 | <nubae> and thats where it gets cool
| |
| 22:44 | <johnny> nubae: i get it
| |
| 22:44 | i just don't see why i'd want to use the app you described
| |
| 22:44 | <nubae> well because if u are in an enterprise
| |
| 22:45 | and for example, a new computer is given to u, u are a new employee or whatever
| |
| 22:45 | <sbalneav> forget ldap?
| |
| 22:45 | <johnny> i'd rather see telepathy talk to something else .. that looks for the profile rather than to sabayon..
| |
| 22:45 | <sbalneav> wha?
| |
| 22:45 | <nubae> and you have just been assigned to the design group
| |
| 22:45 | <johnny> nubae: then you'd just get the profile already
| |
| 22:45 | by telling sabayon to get it from $URL
| |
| 22:45 | <nubae> your computer turns on, sends out a dbus signal saying hey I'm new, need profile
| |
| 22:45 | <johnny> since they are already providing it at URL*
| |
| 22:46 | <nubae> so it looks up who u are
| |
| 22:46 | and gives u what u need
| |
| 22:46 | <johnny> sure.. i just think sabayon is the wrong component to do that in
| |
| 22:46 | <sbalneav> If you're in an enterprise, you're NOT going to forget LDAP, LDAP's going to be central to everything you do. And sabayon can retreive profile mappings from LDAP, and profile .zips from an HTTP address, so dbus doesn't need to be there.
| |
| 22:47 | <nubae> whatever happens under the hood is not important... if it communicates using ldap, haze, libpurple, xmpp
| |
| 22:47 | whatever
| |
| 22:47 | <johnny> nubae: but that already works ;)
| |
| 22:47 | <nubae> its not important
| |
| 22:47 | <johnny> without dbus
| |
| 22:47 | nubae: but have you heard of oddjob?
| |
| 22:47 | you can givei t a list of jobs to perform, and unprivileged users can perform them by talking over dbus
| |
| 22:48 | <nubae> Im not really talking about dbus here anymore
| |
| 22:48 | <johnny> now there's a user for your telepathy talking!
| |
| 22:48 | user*
| |
| 22:48 | <nubae> but more about telepathy and the way it interfaces with dbus
| |
| 22:48 | <johnny> use*
| |
| 22:48 | nubae: i understand that :)
| |
| 22:48 | i think it is great
| |
| 22:48 | and there are tons of uses.. even the usage you describe
| |
| 22:48 | <nubae> well its great for me, cause I cant code C
| |
| 22:48 | <johnny> i just think you're trying to do it at the wrong level
| |
| 22:49 | <nubae> well, I disagree, I see a huge potential for auto-creating profiles by detecting the user through telepathy
| |
| 22:49 | <johnny> at that point tho.. you could do the same with xmpp..
| |
| 22:49 | <nubae> total traversal of NAT
| |
| 22:50 | <johnny> directly i mean..
| |
| 22:50 | <nubae> well, whether u use xmpp, ldap, local link, libpourple
| |
| 22:50 | haze idle google talk
| |
| 22:50 | it dfoesnt matter
| |
| 22:50 | telepathy does it all
| |
| 22:50 | <johnny> sure.. but that's the most terrible part of telepathy
| |
| 22:50 | <nubae> actually, I thought tghat too in the beginning
| |
| 22:50 | <johnny> too much happens in the tubes that should be outside it imo
| |
| 22:50 | i agree with the tube concept
| |
| 22:50 | <nubae> but now I see why its been done this way
| |
| 22:51 | the low level stuff has been completely modularised
| |
| 22:51 | and its as simple as calling mission control and telling it which connection I want it to use
| |
| 22:51 | * johnny still doesn't even see pep in empathy.. | |
| 22:51 | <johnny> nubae: you're not telling me anything i don't know
| |
| 22:52 | <nubae> or to use the best one it thinks is good for the job
| |
| 22:52 | empathy =! telepathy
| |
| 22:52 | <johnny> i know
| |
| 22:52 | <nubae> empathy is a bad example of how telepathy has been integrated into a chat client
| |
| 22:52 | hastily done
| |
| 22:52 | to show something off
| |
| 22:52 | <johnny> i finally have the empathy that shares desktops.. but over tubes
| |
| 22:53 | <nubae> sure
| |
| 22:53 | its one use of dbus tube sharing
| |
| 22:53 | <johnny> so.. yes.. i get the tube concept
| |
| 22:53 | <nubae> in Sugar... every app does this
| |
| 22:53 | <johnny> nubae: why don't you just implement this tubes thing for ltsp ..
| |
| 22:53 | <nubae> which is really cool
| |
| 22:53 | <johnny> telepathy for ltsp
| |
| 22:53 | <nubae> well, where?
| |
| 22:54 | <johnny> share a desktop :)
| |
| 22:54 | <nubae> I dont see where d tubes and ltsp meet, can u explain?
| |
| 22:54 | that already exists
| |
| 22:54 | <johnny> a chroot with a local app that replaces ldm
| |
| 22:54 | and does that share desktkop thing
| |
| 22:55 | <nubae> hmmm not getting u
| |
| 22:55 | what would it do?
| |
| 22:56 | <johnny> so.. standard bootup happens, but instead of ldm, you directly start "sharing a desktop"
| |
| 22:56 | <nubae> btw... PEP support has to do with the server u are using
| |
| 22:57 | <johnny> i run my own server..
| |
| 22:57 | <sbalneav> Something that *IS* actually needed within LTSP is getting proper dbus support for added devices. Currently, we're handling this with udev rules, but this isn't the proper way. What SHOULD happen is when the device is plugged in locally, standard HAL stuff should happen, and there should be some kind of network dbus signal that should go from the thin client to the server to plumb the LTSPFS connection.
| |
| 22:57 | <johnny> nubae: note.. " network dbus signal"
| |
| 22:57 | <jammcq> sbalneav: isn't HAL going away?
| |
| 22:57 | being replaced by something else I think
| |
| 22:58 | <johnny> jammcq: those events can get dispatched over dbus
| |
| 22:58 | just by other programs now..
| |
| 22:58 | <nubae> sbalneav, Id need to understand that a little better, but it sounds to me like telepathy dbus tubes could help with that
| |
| 22:58 | <johnny> specifically.. most of the hal functionality is going to be provided directly by what is in udev, as well as new sysfs entries
| |
| 22:59 | so.. that functionality is going to be obsoleted by udev
| |
| 22:59 | <sbalneav> nubae: You seem to have a new hammer. Just remember: everything isn't a nail. :)
| |
| 22:59 | <johnny> sbalneav: .. but he does have a point.. tubes are NETWORK DBUS
| |
| 22:59 | that's what they do..
| |
| 22:59 | <nubae> sounds very zenish
| |
| 23:00 | still. we are talking here about system dbus
| |
| 23:01 | not session dbus
| |
| 23:01 | telepathy runs over the session dbus
| |
| 23:01 | <johnny> nubae: so what you would do.. is uhmm.. takeover the functionality of certain dbus exposed interfaces before they reach their target on the host system
| |
| 23:01 | and get sent to the client
| |
| 23:01 | and vice versa..
| |
| 23:01 | so.. if you click shutdown in your remote session
| |
| 23:01 | it shuts down your computer
| |
| 23:01 | and not the server
| |
| 23:01 | <nubae> yep
| |
| 23:02 | actually, now I see what u mean
| |
| 23:02 | <johnny> ok.. now go do that :)
| |
| 23:02 | <nubae> u could do this for a whole host of ltsp stuff
| |
| 23:02 | <sbalneav> johnny: Didn't say he didn't.
| |
| 23:02 | <nubae> we should first list them
| |
| 23:02 | before go do them
| |
| 23:02 | <johnny> nubae: of course :)
| |
| 23:03 | <nubae> so we have remote shutdown
| |
| 23:03 | what else?
| |
| 23:03 | <johnny> display brightness?
| |
| 23:03 | <nubae> sound
| |
| 23:03 | <johnny> oh.. for the vice versa..
| |
| 23:04 | <nubae> controlling sound would be extra cool
| |
| 23:04 | <johnny> you see the local network card connection information
| |
| 23:04 | in network manager
| |
| 23:04 | in your remote session
| |
| 23:04 | <nubae> well, if.... everyone gets a xmpp account along with their ldm account
| |
| 23:04 | <sbalneav> sound control does work; that's what pulse does.
| |
| 23:04 | <nubae> ....
| |
| 23:05 | or replace ldm account with xmpp account
| |
| 23:05 | <johnny> nubae: or xmpp server with pam auth
| |
| 23:05 | whatever fits to prototype it
| |
| 23:05 | <nubae> well, xmpp brings in presence
| |
| 23:05 | avatars
| |
| 23:05 | <johnny> avatars?
| |
| 23:05 | <sbalneav> How are you planning to have the ssh login?
| |
| 23:05 | <nubae> a whole host of nice neat things
| |
| 23:06 | <sbalneav> Specifically, the ssh login with password expiry.
| |
| 23:06 | <nubae> xmpp login
| |
| 23:06 | ssl
| |
| 23:06 | F-GT has quit IRC | |
| 23:06 | <sbalneav> "ssl" doesn't answer the question :)
| |
| 23:07 | <johnny> sbalneav: in one of his ideas.. you don't use ssh at all
| |
| 23:07 | <nubae> well, your question was somewhat vague
| |
| 23:07 | right no ssh
| |
| 23:07 | <sbalneav> ok. I'll be more specific.
| |
| 23:07 | <nubae> telepathy does the comms
| |
| 23:08 | <johnny> nubae: that's solving the problem in a different area tho
| |
| 23:08 | depends on which way you go tho..
| |
| 23:08 | <nubae> well... its just exploration right now
| |
| 23:08 | but imagine this....
| |
| 23:09 | user is presented with user.... (password optional)
| |
| 23:09 | if xmpp server present, connects
| |
| 23:09 | downloads profile
| |
| 23:09 | etc
| |
| 23:09 | if no xmpp server present, local-link
| |
| 23:09 | <sbalneav> We need to be able to handle standard login, using standard shadow-password functionality. This includes things like password expiry, password warning periods, and account suspention. We're currently doing this over ssh. We could also do this with PAM. Would you be implementing a full pam module using all this stuff, that would implement all current pam behaviours?
| |
| 23:10 | <nubae> but then of course no authentication.... local-link xmpp cant do autthentication without dirty hacking
| |
| 23:11 | I still dont understand what you are asking sbalneav
| |
| 23:11 | <sbalneav> Really?
| |
| 23:11 | Ok.
| |
| 23:11 | <johnny> sbalneav: i don't think we're into those kind of specifics atm
| |
| 23:12 | just how the system works generally
| |
| 23:12 | F-GT has joined #ltsp | |
| 23:12 | <nubae> well, I think u are asking about how telepathy works...
| |
| 23:12 | <sbalneav> No, I'm not
| |
| 23:12 | because I don't care.
| |
| 23:12 | <nubae> ok, then its a simple xmpp login
| |
| 23:13 | <sbalneav> I only care if a new technology implements the behaviours we already have.
| |
| 23:13 | <nubae> it does+
| |
| 23:13 | but in a much better way
| |
| 23:13 | <sbalneav> So there's a pam XMPP module?
| |
| 23:13 | <johnny> sbalneav: sure..
| |
| 23:13 | xmpp server pam modules that is
| |
| 23:13 | <nubae> it was designed from the ground up as a interprocess communications framework
| |
| 23:14 | <johnny> you auth to xmpp, xmpp auths to pam..
| |
| 23:14 | lots of servers have a pam module
| |
| 23:14 | <sbalneav> sbalneav@phobos:~$ aptitude search libpam | grep xmpp
| |
| 23:14 | sbalneav@phobos:~$
| |
| 23:14 | <nubae> auths to whatever you like
| |
| 23:14 | <sbalneav> I'd say there's a big bupkis there :)
| |
| 23:14 | <johnny> sbalneav: no.. it's server specific sbalneav
| |
| 23:14 | ejabberd for example has a pam auth module
| |
| 23:14 | <nubae> ejabberd for example
| |
| 23:14 | yeah+
| |
| 23:15 | jeez think about it
| |
| 23:15 | sugars whole system runs like this
| |
| 23:15 | <johnny> there is never gonna be an generic xmpp pam module..
| |
| 23:15 | <nubae> theyre whole authentication runs via telepathy alone
| |
| 23:15 | as in fact will gnome in the future too
| |
| 23:16 | <sbalneav> So what does the distributed auth then.
| |
| 23:16 | <nubae> xmpp servers
| |
| 23:16 | <sbalneav> ok.
| |
| 23:16 | <johnny> and most xmpp servers can auth to pam
| |
| 23:17 | <nubae> or if not present
| |
| 23:17 | local cached version of data
| |
| 23:17 | and thats where it gets kinda cool
| |
| 23:17 | <sbalneav> So we have ldap servers TODAY, and we have a pam-ldap module TODAY. Why shouldn't we just do remote auth to LTSP workstations with LDAP, today?
| |
| 23:17 | <johnny> sbalneav: sure..
| |
| 23:17 | you could do that
| |
| 23:17 | your xmpp can auth to ldap..
| |
| 23:18 | you can still use pam-ldap :)
| |
| 23:18 | * jammcq has been talking about doing remote auth for LTSP with LDAP for many years | |
| 23:18 | <nubae> the important thing is to forget about the low level stuff
| |
| 23:18 | <johnny> jammcq: that functinality is going to happen.. just in a different way
| |
| 23:18 | <nubae> it doesnt matter
| |
| 23:18 | its built into both gnome and kde
| |
| 23:18 | and its called telepathy
| |
| 23:18 | let it do what its good at
| |
| 23:18 | <johnny> nubae: i don't think you do a good job at describing it
| |
| 23:19 | <nubae> well, its very very hard to describe
| |
| 23:19 | its really a set of dbus bindings
| |
| 23:19 | a very very big set of dbus bindings that do almost anything u can imagine with communicatino
| |
| 23:19 | but modularised completely too
| |
| 23:19 | <johnny> but.. the point isn't to describe how telepathy works
| |
| 23:19 | <nubae> so its not monolithic
| |
| 23:20 | <johnny> the point is describe how an optimal system would work.. where that implementation might happen to use telepathy
| |
| 23:20 | <sbalneav> telepathy's JUST ANOTHER PIPE. The PIPE isn't what's important. WE'VE GOT NOTHING ON EITHER END OF THE PIPE :)
| |
| 23:20 | <johnny> i think we just said the same thing..
| |
| 23:20 | <nubae> thats actually where I disagree
| |
| 23:20 | <johnny> but in two completely different ways..
| |
| 23:21 | <sbalneav> You guys are arguing about using either copper or PVC, meanwhile we have no water supply, and no toilets :)
| |
| 23:21 | <nubae> u need to understand telepathy in its entirety to see how it can help
| |
| 23:21 | <sbalneav> Well, I think one of the things I'd like to work on this hackfest is getting ldap in.
| |
| 23:21 | <johnny> nubae: not really.. you could just describe how the sytem would work.. and how it fit together
| |
| 23:22 | <sbalneav> That way, LDM can auth to LDAP.
| |
| 23:22 | <johnny> over an ldap connection?
| |
| 23:22 | or still ssh.. but ssh talking to ldap?
| |
| 23:22 | <sbalneav> Then we create a pam module that takes the stored PAM auth token, and plumbs the ssh connection.
| |
| 23:22 | <nubae> oh wow....
| |
| 23:22 | why are we talking low level again
| |
| 23:23 | <jammcq> cuz scotty is trying to actually implement something that works
| |
| 23:23 | and at some point, you have to talk low level
| |
| 23:23 | <nubae> yes but we have already established that telepathy does all that
| |
| 23:23 | it authenticates
| |
| 23:23 | <johnny> nubae: telepathy doesn't do it alone
| |
| 23:23 | <sbalneav> NO IT DOESN"T
| |
| 23:23 | <nubae> in any way u want it to
| |
| 23:24 | <sbalneav> it's just a pipe.
| |
| 23:24 | <nubae> actuallly it does
| |
| 23:24 | I call mission control
| |
| 23:24 | <sbalneav> ok, so WHERE's the telepathy pam module?
| |
| 23:24 | <johnny> sbalneav: you don't need one
| |
| 23:24 | <nubae> mission control selects the connection manager I want
| |
| 23:24 | <johnny> that's the point
| |
| 23:24 | <nubae> it creates a connection
| |
| 23:24 | <johnny> you might need an xmpp server with a pam module yes..
| |
| 23:25 | <nubae> xmpp is what does all the magic
| |
| 23:25 | <sbalneav> ok, so the XMPP server and PAM module are doing the auth
| |
| 23:25 | NOT TELEPATHY.
| |
| 23:25 | <nubae> no
| |
| 23:25 | telepathy
| |
| 23:25 | which USES XMPP
| |
| 23:25 | <johnny> nubae: does it still send an xmpp login request in xml?
| |
| 23:25 | <nubae> yes
| |
| 23:26 | <johnny> then it's xmpp doing the uath
| |
| 23:26 | wrapped by telpathy perhaps
| |
| 23:26 | <nubae> but it can use haze
| |
| 23:26 | <johnny> but still xmpp doing it
| |
| 23:26 | <nubae> or libpurple
| |
| 23:26 | <johnny> but haze just uses something else
| |
| 23:26 | <nubae> or wilde
| |
| 23:26 | <sbalneav> Well, go ahead and implement it then :)
| |
| 23:26 | <nubae> ofr whatever u feel like
| |
| 23:26 | <johnny> nubae: WE KNOW
| |
| 23:26 | in in those cases.. it auths to msn, irc, aim, etc
| |
| 23:26 | and their $whatever does the auth
| |
| 23:27 | <nubae> well.. then why a i being told that telepathy doesnt do authentication
| |
| 23:27 | when it does?
| |
| 23:27 | <johnny> telepathy is just a generic interface to those specific auths
| |
| 23:27 | <sbalneav> because it's a pipe to send authentication information over
| |
| 23:27 | PAM does the auth
| |
| 23:27 | <johnny> telepathy is just a freindly wrapper around their auth
| |
| 23:27 | Bart__ has joined #LTSP | |
| 23:27 | <nubae> its a set of dbus bindings
| |
| 23:27 | <johnny> nubae: that talk to those services!
| |
| 23:27 | specifically at some point
| |
| 23:28 | <nubae> and its a whole host of interfaces
| |
| 23:28 | <johnny> it's just an abstraction nubae
| |
| 23:28 | it's not a thing on it's own
| |
| 23:28 | <nubae> right
| |
| 23:28 | right
| |
| 23:28 | <johnny> ok.. so we understand that..
| |
| 23:28 | <nubae> which is why its difficult to explain
| |
| 23:28 | <johnny> nubae: not really
| |
| 23:29 | you're trying to do two things at once here
| |
| 23:29 | describe how the system actually works for a user
| |
| 23:29 | <nubae> well... I cant even get the #telepathy guys to explain it properly
| |
| 23:29 | <johnny> while many us us need to understand how the pieces fit :)
| |
| 23:29 | <Bart__> Hi, sorry to interrupt, but can I attach a DVD burner on my thin client? So that the workstation can burn DVDs?
| |
| 23:29 | <johnny> Bart__: sorry.. not yet
| |
| 23:29 | in fact.. part of this discussion is to perhaps solve tha issue
| |
| 23:29 | at least on very fast networks
| |
| 23:30 | <Bart__> Oh I see. That was a quick. =)
| |
| 23:30 | <johnny> Bart__: it's somewhere in there
| |
| 23:30 | yanu has quit IRC | |
| 23:30 | <johnny> we just haven't hit the nail on the head yet :)
| |
| 23:30 | <nubae> u could if, for example, there were dbus bindings that brasero had
| |
| 23:30 | yanu has joined #ltsp | |
| 23:30 | * sbalneav bangs head on desk | |
| 23:30 | <nubae> u could then create a dbus tube to the thin client and call that
| |
| 23:30 | <johnny> nubae: there are.. but you still have to transfer the files back to the system
| |
| 23:30 | <nubae> yeah, thats why its a tube
| |
| 23:31 | its 2 way
| |
| 23:31 | or 3 or 4
| |
| 23:31 | <johnny> you're thinking at the wrong level of where these bindings are again.. and it is cluttering the message
| |
| 23:31 | <nubae> actually its p2p
| |
| 23:31 | <johnny> application level isn't hte point
| |
| 23:31 | <sbalneav> Bart__: No, but you could, potentially, run a DVD burner program as a local application.
| |
| 23:31 | <johnny> we're talking about adjusting the part that computer describes
| |
| 23:31 | <nubae> well, for the brasero thing I just sidetracked
| |
| 23:31 | wasnt on purpose
| |
| 23:31 | <Bart__> Johnny thanks for all your help, I was able to fix all the problems in my Intel Atom Clients now
| |
| 23:32 | <johnny> nubae: no.. thing is you're right..about the functionality that you are describing as good, just solving it at the wrong level
| |
| 23:32 | <nubae> look... a good example is vino/vinagre
| |
| 23:32 | <johnny> the important thing is adjusting what happens at the system and session level
| |
| 23:32 | nubae: we get it!
| |
| 23:32 | <Bart__> sbalneav, yeah, but I need to burn something, guess I have to find other methods
| |
| 23:32 | <johnny> it's just the wrong place to solve the problem
| |
| 23:33 | that we have
| |
| 23:33 | there are ways using telepathy to help us solve the problem.. i can see them
| |
| 23:33 | <sbalneav> Bart__: This for a lab?
| |
| 23:33 | <nubae> well, I'm not sure how session and system can communicate
| |
| 23:33 | <johnny> nubae: turnig it into a dbus proxy.. that's how
| |
| 23:33 | <nubae> cuase telepathy is totally session based
| |
| 23:33 | <johnny> i don't see that as a problem
| |
| 23:33 | i mean. .it might limit a few things we can do..
| |
| 23:33 | <nubae> I dont mean that
| |
| 23:34 | <Bart__> For office... Because we have videos, and my officemates, we just burn those files off our file server then we present them to customers
| |
| 23:34 | <nubae> there are already tons of dbus proxies for python in telepathy-python
| |
| 23:34 | <johnny> nubae: ! :(
| |
| 23:34 | nubae: let me describe what i'd like to see
| |
| 23:35 | at a level that helps LTSP that could potentially happen in a short amount of time
| |
| 23:35 | in a somewhat low level way.. and i want you to think about how you can solve it
| |
| 23:36 | <nubae> loook, I see in the simplest usage of telepathy.... remote control of whatever
| |
| 23:36 | <johnny> nubae: c'mon dude..
| |
| 23:36 | <nubae> what?
| |
| 23:37 | <johnny> hold on a sec
| |
| 23:37 | <nubae> using xmpp to do authentication is a wonderful idea, but it'll just never happen...
| |
| 23:37 | <johnny> looking for a good example
| |
| 23:37 | <nubae> so I'm being realistic
| |
| 23:38 | remote control of all aspects of thin client computers
| |
| 23:38 | <sbalneav> Bart__: I remember reading some time ago a tutorial for setting up a DVD burning stattion that used Samba. You "printed" the ISO to the server, and it burned the dvd for you.
| |
| 23:38 | <nubae> from sound/brightness/enable disable video
| |
| 23:39 | enable disable utube
| |
| 23:39 | stuff like that
| |
| 23:39 | <Bart__> Anything possible I will try. Where can I start looking for this tutorial? I've searched google, and none can find
| |
| 23:40 | <johnny> yes
| |
| 23:40 | <Bart__> Also, can you suggest ideas how to make my thin clients faster? I've tried LDM_DirectX = True. Will running local apps also make it faster?
| |
| 23:40 | <johnny> nubae: here's the remotte control i'm interested.. the specific thing we need implemented (imo)
| |
| 23:41 | that if we have say.. org/gnome/PowerManager/Brighness .. the one we see.. is not the one from the server.. but from the client
| |
| 23:42 | <sbalneav> Bart__: Well, what kind of thin clients do you have? And what exactly is slow?
| |
| 23:42 | <johnny> so.. something that would mix a set of server interfaces, with a set of client itnerfaces
| |
| 23:42 | so.. the user would see all the client monitor information
| |
| 23:43 | and i should be able to see both the remote storage, and local storage
| |
| 23:43 | if you can help us plumb those features.. the greatest problem to ltsp would be solved (imo)
| |
| 23:43 | <nubae> hey that would be cool
| |
| 23:43 | like a thin client inspector
| |
| 23:43 | <Bart__> Um, not really slow, but there's a .3 sec delay on mouse click, but with LDM_DirectX = false it's .7 sec, I have Intel Atom the dual core type, it's fast enough but I want to optimize it more, thanks.
| |
| 23:43 | <nubae> almost on a debug level
| |
| 23:43 | <johnny> sure.. that is one tiny thing you could do es..
| |
| 23:43 | but the important thing is that the you can override some of the exposed dbus interfaces by the ones from the client
| |
| 23:44 | or certain ones from the server
| |
| 23:44 | <nubae> hmm... u lost me there
| |
| 23:44 | give me an example
| |
| 23:44 | <johnny> uhm ?
| |
| 23:44 | i did
| |
| 23:44 | the power manager brightness thing..
| |
| 23:44 | <sbalneav> Bart__: You may want to look at stgraber's PPA. There's been some repairs to the X libraries that fix latency problems.
| |
| 23:45 | <nubae> i still dont get what the problem is there though?
| |
| 23:47 | <johnny> having the logged in section seeing org.freedesktop.DeviceKit.Disks.DeviceAdded
| |
| 23:47 | if i plugin usb flash drive.. :)
| |
| 23:47 | plugged in on the thin client
| |
| 23:48 | <nubae> by the way, telepathy (jumping low level again briefly) intends to wrap encryption around all the protocols it evenually will support
| |
| 23:48 | <Bart__> Ok what does stgraber's PPA mean?
| |
| 23:48 | <nubae> uding XTLS and jingle
| |
| 23:49 | <Bart__> How do I install it?
| |
| 23:49 | <nubae> oh yeah remote control flash player would be nice indeed
| |
| 23:49 | <johnny> ?
| |
| 23:49 | flash player?
| |
| 23:49 | i'm talking about hardware..
| |
| 23:49 | <nubae> i bet flash player alread has dbus bindings
| |
| 23:49 | oh
| |
| 23:50 | <johnny> i just need to talk to the dbus guys i guess
| |
| 23:50 | <nubae> sorry I wasnt reading that part, thoguht uwere taling to bart
| |
| 23:50 | <johnny> the nice thing about software that has come out recently..
| |
| 23:50 | <sbalneav> Bart__: https://edge.launchpad.net/~stgraber/+ppa-packages
| |
| 23:50 | <johnny> is that they are not using hard coded device strings and whatnot
| |
| 23:50 | and not having alsa.. but pulse..
| |
| 23:51 | <nubae> johnny, u can call it whatever u like in dbus
| |
| 23:51 | <johnny> huh?
| |
| 23:51 | <nubae> as long as it contains u're domain name backwards
| |
| 23:51 | the device
| |
| 23:51 | <johnny> uhm..
| |
| 23:51 | i'm referring to the fact that now they just listen in on signals for what to do
| |
| 23:52 | <nubae> no
| |
| 23:52 | <johnny> which means.. that if we can inject signals from remote
| |
| 23:52 | <nubae> signals and methods
| |
| 23:52 | <johnny> sorry.. send signals..
| |
| 23:52 | <nubae> but backup
| |
| 23:52 | what specficially are u taling about here
| |
| 23:52 | which device
| |
| 23:52 | <johnny> look.. i'm asking for your help on how to solve a most pressing problem for ltsp
| |
| 23:52 | no specific device
| |
| 23:53 | i just gave ou an example of the kinds of things i want to happen
| |
| 23:53 | a very specific example.. but only an example
| |
| 23:54 | <nubae> ok, i must be blind, please repeate the specific example
| |
| 23:54 | <Bart__> won't these packages break my client? do I need to start a new ltsp-build-client? what do you sugges sbalneav? thx
| |
| 23:55 | <johnny> nubae: lemme try again
| |
| 23:55 | <nubae> Bart__, backup
| |
| 23:55 | <sbalneav> Bart__: Are you running jaunty?
| |
| 23:55 | <nubae> as the saying goes
| |
| 23:55 | <johnny> if i plug camera into a thin client machine, and i plug a flash drive into the server
| |
| 23:56 | if i'm logged into a session remotely.. i should see the camera, and the flash drive
| |
| 23:56 | but not the way we do now
| |
| 23:56 | with ltspfs
| |
| 23:56 | and checking /media ..
| |
| 23:57 | <nubae> well, it doesnt have to be mututally exclusive
| |
| 23:57 | <johnny> no.. but it could be :)
| |
| 23:57 | you could set policy at the server level..
| |
| 23:57 | via policykit
| |
| 23:58 | <nubae> u really want to rewrite LTSP yet again?
| |
| 23:58 | <johnny> no.. it wouldn't involve rewriting ltsp
| |
| 23:58 | basically just a dbus proxy :)
| |
| 23:58 | <nubae> ok... stop there for a sec
| |
| 23:58 | <johnny> back to that again..
| |
| 23:59 | <nubae> for me a dbus proxy... is basically just a standin object in the lingo of your choice
| |
| 23:59 | for a dbus method or signal
| |
| 23:59 | <johnny> hmm?
| |
| 23:59 | nubae: .. the proxy would mix in the methods and signals.. and their targets
| |
| 23:59 | <nubae> ok... thats as simple as I can explain I think
| |