LainOS Developers Meeting #3 May 3rd, 2003 irc.eggheads.org (#LainOSDev) Topic: "Technology Testing/3D UI Demos and Other Various Things" [19:03] [Neovanglist] first lets address darth's suggestion [19:04] [Neovanglist] should LainOS be strictly workstation focused, or should we take into consideration server support [19:04] personally, I don't see the point in adding server support. [19:04] the features that we'll be adding aren't especially useful to servers in any way. [19:04] you could argue about biometric authentication, but in reality, any server worth its weight is already physically secured. [19:04] [Neovanglist] I agree [19:05] cutting out server functions not only makes it easier on us, but lets us really focus down. [19:05] [Neovanglist] well another thing [19:05] [Neovanglist] is that if we are shooting for serving [19:05] [Neovanglist] security becomes a much much larger issue [19:05] also, all of the extra features are going to consume resources; hardly a good thing for servers. [19:05] [Neovanglist] indeed [19:06] [Neovanglist] alright, so that settles that (unless anyone else has any comments) [19:06] [Neovanglist] next, we need to start making hard progress on LainWM [19:06] have we decided on a definite feature set? [19:06] [Neovanglist] no, but the problem is we aren't getting enough public feedback to get a good idea base [19:07] oh, just to make things more complicated, I wanted to throw this out. [19:07] http://www.visualthesaurus.com/online/index.html [19:07] I really like the interface, and I could see it becoming usable as a desktop. [19:08] but instead of having it always in 3d space, have it such that you hit a button (say win) to access 3d space. [19:08] otherwise, it always focused on a window. [19:08] (the window being where the word would be in the current implementation) [19:09] [Neovanglist] yea, I see where you're going [19:09] [Neovanglist] but it's still very linear [19:09] [Neovanglist] in a way [19:09] something with a desktop like xcruise (but more organized, and click to navigate) with a 2D workspace for documents would be interesting [19:09] [Neovanglist] you move from word directly to word [19:09] what if our design let you move from any word to any other word? [19:09] xcruise? link? [19:10] one sec.. [19:10] what I find amazing is that these guys did it with basically a web-browser. [19:10] LainWM would need to be more complex, but think of how little resources visual thesaurus is taking up. and what few hardware requirements there are. [19:10] http://xcruiser.sourceforge.net/ [19:11] [Neovanglist] ahh I've used that before [19:12] [Neovanglist] the thing is [19:12] with some more attention to organizing where things go, and making things more visible, it could become very usable [19:13] [Neovanglist] hrmm [19:14] [Neovanglist] ok lets keep things flowing [19:14] [Neovanglist] the thing I don't see though [19:15] [Neovanglist] is how to impliment what visualthesaurus does in a WM way [19:15] it's a java plugin into a browser. [19:15] people have already written window managers that let you use webpages as backgrounds, so that's a start. [19:16] we'd have to find some way for the WM to "user applications as webpages" in a sense, [19:16] and have the WM load these apps as webpages where the words are now. [19:16] as for organization, we could have the applications clustered together, somewhat like what windows XP does. [19:17] [Neovanglist] yea, but is that really the most elegant, efficent way to do that? [19:17] but again, unless we want to write it from the ground up, we'd have to find the source somehow, and I doubt that a commercial company is going to give away source. [19:17] I think we agree that MS's implementation of a 3D workspace, sucks, right? [19:17] [Neovanglist] yes [19:17] that is, the long hallway with picture frames on the walls? [19:17] [Neovanglist] yea, it's retarted [19:17] truth be told, I'm not happy with 3dna's implementatin either. a virtual house? [19:18] you can't see behind you, so that's a waste. [19:18] earlier, Xuric told me to repeat: [19:18] I'm all for an animated desktop... has to be fluid though [19:18] also, anything we substitute for the 'taskbar' has to be SIMPLE and EASY... [19:18] the idea of digging with transparency can get tedious when you have 9 windows open" [19:18] I like my idea of digging, but Xuric is right; once you have a lot of windows, it's more of a nuisance than a help. [19:19] [Neovanglist] well [19:19] [Neovanglist] here is the thing [19:19] [Neovanglist] if programs are in program groups [19:19] [Neovanglist] the different levels of depth (the things you dig to and from) [19:19] [Neovanglist] are on rings a certian distance from the middle [19:19] [Neovanglist] each app isn't op top of the other [19:20] [Neovanglist] it might be on the depth level in front of it and to the right [19:20] [Neovanglist] so if you wanted to go to ring 2, app 3 [19:20] hey [19:20] [Neovanglist] you'd hit like alt+2+3 [19:20] [Neovanglist] hey there minddog [19:20] you'd have to remember which ring/depth each app was on? [19:20] [Neovanglist] not really [19:21] they're all visible (to some degree) simultaneously? that is, the whole ring system? [19:21] [Neovanglist] hit alt, and a box pops up with the groups, hit the group number (2) and next to the other box a box comes up with the list of apps in that ring [19:21] [Neovanglist] well they don't have to all be visable at once, but you would for example hit alt, and it would zoom out a bit [19:22] [Neovanglist] showing more of the would around that window [19:22] http://mrl.nyu.edu/~perlin/zoom/SiteTour.html and http://mrl.nyu.edu/~perlin -- the zoom section may be of interest [19:24] [Neovanglist] hrnn [19:24] [Neovanglist] *hrmm [19:24] [Neovanglist] and welcome minddog [19:24] [Neovanglist] glad you could join us [19:25] thanks, just listening =) [19:25] [Neovanglist] fell free to talk and contribute [19:25] [Neovanglist] *feel [19:25] [Neovanglist] well here is what I was thinking [19:26] [Neovanglist] we can talk, and debate these things all we want, but we really won't know how well/what works until we try [19:26] [Neovanglist] so lets start some technology tests, and do some experimentation [19:26] I hope you mean small scale. : ) like flash demos? [19:26] [Neovanglist] collect data on what does and doesn't work, what needs to be improved/rethought, and apply it to LainWM [19:27] [Neovanglist] I mean any scale, flash demos, OpenGL apps, 3d renderings [19:29] [Neovanglist] from that we can better picture what LainOS should be [19:29] sounds good. [19:29] [Neovanglist] alright, cool. I'll write up guidelines for it tonight and post it on the stie [19:29] [Neovanglist] *site [19:30] [Neovanglist] also [19:30] [Neovanglist] we need to get some PR going [19:30] [Neovanglist] too few people know about us, and what we are doing [19:31] I almost want to say that maybe we should hold back on announcing our presence. [19:31] [Neovanglist] I've thought the same thing [19:31] a year ago, when I first heard about LainOS, it was all over the place. [19:31] EVERYONE was talking about the people who integrated X into the kernel. [19:32] a few days ago, I did another quick search, and I turned up three pages of people saying *cough*vaporware*cough* [19:32] I'd rather we have something to show them (however primitive), and then announce ourselves. [19:32] it seems that a lot of people are confused about how much/little progress has been made, and I'd rather not add to that confusion. [19:32] [Neovanglist] then we need to start working on getting some solid concepts out in code [19:33] absolutely. [19:33] * malenfant has never heard of LainOS before now [19:33] i think running the tech tests, and getting feedback is going to help tremendously in getting our asses into gear [19:33] [Neovanglist] yea, me too [19:33] [Neovanglist] lets do this [19:33] [Neovanglist] do you know any opengl/sdl? [19:34] malenfant, I think I first heard of LaiOS last june (ironically, months before I saw Lain, so I REALLY had no clue what LainOS was supposed to be) [19:34] no opengl. [19:34] honestly, I'm not much of a coder. but I'm a fast learner, and I've got the summer before me. [19:34] [Neovanglist] check out this site http://nehe.gamedev.net [19:34] [Neovanglist] it's a great into to graphics programming on all platforms [19:34] [Neovanglist] from Windows to Mac to Solaris [19:34] aight, I'll look into it later tonight. [19:34] [Neovanglist] and everything in-between [19:35] I know only a little SDL... I'm very familiar with traditional GUI programming [19:35] [Neovanglist] that will help tremendously [19:35] [Neovanglist] I'm experienced with SDL and opengl, but no so with GUI programming [19:35] [Neovanglist] that is if you'd like to contribute to us [19:36] * darthchung shakes fist threateningly at minddog [19:36] I'll have to see more about what the project is about, and if time permits... so far, sounds interesting :) [19:36] [Neovanglist] cool, stay in touch [19:36] [Neovanglist] what I can start doing is some simple opengl demos [19:37] [Neovanglist] handling blank "would be" windows in a 3d space [19:37] that was my plan, but I was going to do it in flash [19:38] [Neovanglist] I think OpenGL is a better idea because it will allow us to more directly impliment what we learn into LainWM [19:38] [Neovanglist] and get some code out to show people [19:39] I'll see how much time I've got; the next two weeks are finals season for me. [19:39] [Neovanglist] same for me [19:40] [Neovanglist] well the next meeting is two weeks from today [19:40] [Neovanglist] may 17th [19:41] aight. [19:46] what else do we want to touch on tonight? [19:47] [Neovanglist] well thats mostly it [19:47] * Neovanglist thinks [19:48] [Neovanglist] nope, I think thats it [19:48] [Neovanglist] anybody? [19:48] alright, so our goals for the next two weeks are to put together some demos to get feedback [19:48] (and of course to think of features to put into the demos.) [19:49] [Neovanglist] yup [19:49] aight; I'll see you guys in two weeks then. [19:49] good luck on your finals. [19:51] [Neovanglist] you too [19:51] I will try to drum up some time to do some more SDL, and do a few things with that... also perhaps look at DirectFB