21:03:05] not really, we can get started now if people are ready [21:03:09] * Hellfire667 is ready [21:03:17] * Hyronymus is ready too [21:03:20] ready [21:03:20] * chirho sets mode: +v Hyronymus [21:03:25] * chirho sets mode: +v Arathorn [21:03:46] how generous Very Happy [21:03:47] Apparently GameBot is not, so: [21:03:49] Right then, who's got some progress to report and talk about? [21:03:51] * Hellfire667 sets mode: -v GameBot [21:03:56] * Hellfire667 sets mode: +m [21:04:07] ewwww, not my mode of choice ... [21:04:20] then why "voice" people? [21:04:24] * Hellfire667 sets mode: -m [21:04:48] so we can see who's actually here Smile [21:05:05] (for counting up afterwards) [21:05:10] ah. ok. [21:05:10] Anyway, shall we crack on? [21:05:12] Yes: [21:05:12] yup [21:05:12] As you all might have noticed, Epistax and I have worked on the UML diagrams of Zugspiel [21:05:29] Are the results promising? [21:05:33] Well... [21:05:40] Well.. the results are the UML. [21:05:46] The code needs quite a bit of work. [21:05:50] indeed [21:05:59] (Standards wise) [21:06:01] Chochma's coding style is really messy [21:06:02] From my POV, the results are the inferences we can draw from the UML Smile [21:06:27] OK, is it going to take a lot of work to clean it up? [21:07:04] to clean up his coding style, we could use a standard program... [21:07:08] but I haven't seen much of commenting in there [21:07:09] (the header files, that is) [21:07:12] Depends on how it's done. A lot of it is nitpicking. The UML program we are using is automaticly generating tasks for standards. [21:07:20] The external code he used is commented. [21:07:30] Such as "this is a bad name" or "incorrect capitalization", etc. [21:07:58] It might be worth adding here that I've still had no reply to my email yet, so I assume he's still away and hasn't necessarily heard what we've been doing [21:08:21] he has no clue you are looking in his code? [21:08:28] No. [21:08:50] I see [21:09:04] well, I put an email off last week ... [21:09:04] But looking into his code is ok: he shouldn't have posted it on his website if he didn't want that Wink [21:09:04] suggest he doesn't want to borrow it [21:09:18] * orudge is now known as orudge|away [21:09:30] It can still be used for reference (I would assume). [21:09:42] true [21:09:56] Epistax: In your opinion, would it be wise to use Zugspiel as a base to build TE on? [21:09:59] but does it pay to improve it then before you have clearity? [21:11:07] As a base, as in starting strictly from that code and modifying it, perhaps not. However the UML built from it might be. [21:11:16] I agree. [21:11:33] The whole reason for Zugspiel to exist, is to test Chochmah's pathfinding code. [21:11:50] As we can see in the demo on his website, that part works fine [21:12:03] (the program crashes when I try to close it) [21:12:09] (I did make the UML to match the code, meaning some errors exist in the UML). Perhaps that portion is available to us? [21:12:27] Probably. [21:13:15] I think we should start building from scratch, and use, if we are allowed to, Chochmah's pathfinding code. [21:13:21] Is there a licence document included in chochmah's source bundle? [21:13:21] just about to ask what you wanted to do [21:13:21] The UML diagrams we have now can aid us with that. [21:13:25] I'll check right away! [21:13:42] And I would point out "start by scratch" could mean grabbing a few parts of code from various places (that allow us to), such as for an SDL engine, sound, etc. [21:14:08] None. [21:14:41] I agree on the SDL part... We can use SDL for almost all aspects of cross-platform OS interaction. [21:14:52] sound, net, window system [21:15:06] and use OpenGL for the graphics. [21:15:36] are we certain about the choice for OpenGL? [21:15:53] Do you have a better alternative? [21:15:57] I'd support anything cross platform (not DirectX) [21:16:26] i would very much like TE to be open source as well [21:16:26] (For the record my system has XP and Mandrake 9) [21:16:37] no, but I mean shouldn't there be a proper research in what to choose [21:17:21] I'm taught that having a proper foundation under your decissions minimizes criticism [21:17:23] Arathorn: TE will be open source. That's the whole reason of using sourceforge and all. If we wanted it to be closed source, we could simply wait for Rein Wink [21:17:36] eh [21:17:40] what have I been saying [21:17:46] Hyronymus: true, but there isn't much research to do on 3d api's. [21:17:53] I've been confusing open source and multi platfoirm today [21:17:58] ah [21:17:59] sorry Razz [21:18:00] hehe [21:18:00] okay. [21:18:07] Well I hope for both! [21:18:11] So do I. [21:18:17] I know, Hellfire667. I won't make a problem of it [21:18:17] in which case, we decided it should be multiplatform about a year ago Smile [21:18:26] but proper documentations is valuable [21:19:25] the documentation you can make about the choice for OpenGL already is: supported by multiple platforms [21:19:27] and so is SDL. [21:19:41] It's not an issue at this point but I'd like to point out that we should avoid proprietary formats such as gif and mp3 and instead us such open formats as png and ogg. [21:19:53] * Hyronymus agrees [21:19:53] Hehe... If we only wanted to support multiple platforms, even Java would be an option [21:20:05] * Hellfire667 too [21:20:27] is jpeg proprietary? [21:20:36] I think it is, but not sure. [21:20:40] chirho, can you make a summation of what has been discussed sofar? [21:20:44] it might be, but it's not patented [21:21:14] Hyronymus: as in so far tonight? [21:21:23] Anyway.. Jpeg is lossy. PNG isn't. So PNG will give a better image quality [21:21:39] yes, chirho [21:21:49] I won't throw a vote for Java because that'd bump up the minimum system requirements by a couple hundred mhz. [21:21:58] (Sorry, should have said that earlier) [21:22:01] [15] Freely available source code for JPEG [21:22:05] That's why I said "only" Wink [21:22:13] Just found a FAQ on JPEG [21:22:25] post-UML discussion: probably not using the zugspiel code, though pathfinding stuff might be useful ... [21:22:26] * HEXerium has quit IRC (Signed off) [21:22:28] Ok so jpg's are free and open, but png's are preferable. [21:22:34] but as mentioned, JPEG isn't a s good [21:22:46] so we'll stick to png? [21:23:25] let's keep that choice open. [21:23:25] followed by cross-platform discussion: SDL/OpenGL, etc., and now data formats [21:23:38] Depending on the graphics we're using, png may not be possible, so jpg can be a 2nd choice. [21:23:47] why may it not be possible? [21:23:57] Not a clue. [21:24:42] would it be much trouble to have it support both? [21:25:08] what's the use of supporting both? [21:25:08] Might tnot be, but if it supports both, it'd likely all be png. [21:25:08] Probably not. [21:25:31] Hyronymus: flexibility. [21:25:50] The interface could always be there for user-custom graphics. [21:25:51] Hellfire667, that means supporting readme in txt, hmtl and doc too?! [21:25:53] There are people who prefer jpeg over png because jpeg usually gives smaller files. [21:26:13] Hyronymus: Not really [21:26:23] Apples and oranges ... [21:26:31] Hyronymus: indeed. But if we write the readme in for example texinfo format, we can automagically generate .txt .html .ptf etc. [21:27:10] it can be done Wink [21:27:31] As it would effect coding all over, do we want to use text modules such that different languages can be subbed in and out? (English, French, Spanish...) [21:27:31] but if size does mattters more then quality [21:27:31] *affect [21:27:34] Epistax: Yes. [21:27:38] * Hyronymus votes for text modules [21:27:52] Message catalogs are a very good idea. [21:27:59] I've never done it before. If no one else here has I could look into it. [21:28:17] sure. [21:28:32] If i'm not mistaken, Zugspiel uses a text module. [21:28:35] gettext is very good, though I'm not sure if Windows has an answer for it ... [21:28:43] Heck we could add a babelfish interface and offend a great number of people Wink [21:28:48] hehe [21:28:49] hehehe [21:28:57] babelfish doesn't support dutch Sad [21:29:28] Perhaps there is an answer on sourceforge.. [21:29:36] how about a structure like Civilization? [21:29:38] http://home.a-city.de/franco.bez/gettext/gettext_win32_en.html [21:29:44] Something that should be very much avoided is things like "printf(messages[1]);" [21:29:56] * Hellfire667 agrees [21:30:01] where you have all sorts of txt-files: civilopedia.txt etc [21:30:20] I love Civ's style. [21:30:26] that would be good yes [21:30:27] It's great [21:30:33] ery easy to hack for nerds [21:30:36] *very [21:30:38] otherwise you would have a file with a huge amount of text strings [21:31:01] only has advnatges I think [21:31:31] "Jackal.txt" for errors, naturally Smile [21:31:37] as long as you keep the string the same it reads all that's put behind it [21:31:39] hehe [21:32:01] Will we support Unicode characters? [21:32:02] So a text engine would be a task. [21:32:03] * Hyronymus wonders if string is the proper name [21:32:20] Unicode is prefered I think [21:32:32] multiple languages, multiple characters [21:32:36] Hellfire667: UTF-8 is the ideal situation, as long as the support is there [21:32:44] Hellfire: I've done japanese characters in VC++, but I'm not so sure there's a method that'll work in both W32 and *nix. [21:33:30] hmm [21:33:30] * HEXerium has joined #tempire [21:33:30] Hello, [21:33:30] Hello, HEXerium [21:33:35] Is this project going to be done in such a way that we'll be able to use GPL'd code? [21:33:37] I don't think we want to get bogged down in multiple encodings, so UTF-8 is an i18nally-safe option [21:33:44] Epistax: The idea is to issue TE under GPL, so using GPL'd code shouldn't be a problem [21:33:51] UTF-8 it is Smile [21:34:00] Sounds good! [21:35:23] * Hyronymus accepts UTF-8 too [21:35:39] So... [21:35:42] What on the todo list for next week? [21:36:04] * Research into text modules? [21:36:11] is there a todo-list on the forum (in a locked topic) [21:36:12] * Research into cross-platform api's? [21:36:40] Hyronymus: Not yet, though I might tack it onto something already stuck there [21:36:58] please make sure it is locked Wink [21:37:27] Do we want to commit some of this to CVS if only to appease the god of sourceforge %'s? [21:37:34] research in usubiltiy of png / jpeg [21:37:51] damn, don't look at my supposed-to-represent english [21:37:56] Epistax: submitting binairies into cvs is not a really good idea. [21:38:03] It's possible [21:38:16] Not binaries, let's say To do, for instance [21:38:20] but you can't simultaneously work on a file without fixing conflicts [21:38:22] Ah yes. [21:38:28] That's a good idea [21:38:41] chirho, no need for a todo on the forums then [21:38:51] just a link to the todo on SF [21:39:27] Be warned that SF CVS has the anonymous user delay. [21:39:27] Hmm... A todo in the cvs is indeed better: non-developers can't mess with our little list Wink [21:40:06] warning taken [21:40:18] though I suppose it should be easy to have a dev-docs module in CVS, and have a script check it out and put it somewhere ... [21:40:36] I could do that within 5 minutes... [21:40:38] Wink [21:40:38] Sigh. Looks like I'll have ot slap the guy with the MP3s ... [21:40:47] It'd be nice to define the roll CVS will play, as oppose to the forum. And then there's the forum on sourceforge we have. [21:40:55] Hellfire667: I could write such a script in my head, let along on a shell server Smile [21:40:57] did we discuss everything on the agenda, chirho? [21:41:40] Who has mp3's? I can ogg'm Wink [21:41:46] I have mp3's and an ogg converter [21:41:47] I would prefer people making them in ogg immedeately [21:41:47] although it refuses to work [21:41:52] wich one do you have? [21:42:14] lemme check (only run it once, got disppointed Happy [21:42:16] If we have any video at any point, I'd suggest ogm format. (Just thought I'd throw that out there) [21:42:23] About the role of CVS: [21:42:25] umm... [21:42:41] We could restrict CVS to code [21:42:44] CVS will definitely be the place for the source files, when we start coding... [21:42:52] that's a possiblity. [21:42:58] text-based docs can go in there as well [21:43:00] Then there's the question: the dependent datafiles. [21:43:00] I'd like to see the documentation in there as well [21:43:07] well: [21:43:12] yes documentation I agree [21:43:13] just a thought. [21:43:39] it depends on how "dependent" we are on them [21:43:39] Put the datafiles in a separate module [21:43:42] the source files in a separate module [21:43:50] and IDE related files (such as the .dsw and .dpr file of MSVC++) in a separate module [21:43:50] I suppose we could .cvsignore the datafiles ... [21:43:59] Arathorn: mp3toogg converter (original name) [21:44:04] ok [21:44:10] then, [21:44:15] i use dBpowerAmp [21:44:24] and it works? [21:44:24] hack into CVSROOT/modules and make a line like: [21:44:27] works from win exploror [21:44:32] very god [21:44:34] (sorry for the OT) [21:44:34] vc-src vc-src &src [21:44:38] you only need an ogg plugin [21:44:46] yeah sorry fot ot Wink [21:44:52] which will checkout the MSVC files and the sourcefiles if someone checksout vc-src [21:45:03] that makes sense [21:45:41] one word about sources, they doesn't compile with mingw Razz [21:45:41] same for (for example) a anjuta-src module (anjuta is a Linux IDE) [21:45:47] Jpl: there are no sources yet [21:45:47] Smile [21:45:49] yeah, that line kinda confused me [21:45:57] aw studio sport time, I'm sure you guys can enjoy yourselves without me Wink [21:46:04] if you use msvc, they will not compile Razz [21:46:13] * chirho DCCs Arathorn a VCR [21:46:16] we're old enough, Arathorn [21:46:18] Now Linux is different from windows (waits for the shock in the room to die down). Win32 executables are almost always expected, whereas linux are not. I guess we won't have to worry about this for a few months anyway so nm. [21:46:43] Executables are distributed separately [21:46:46] but do keep it in the back of you mind, Epistax [21:46:59] like Simutrans, chirho? [21:47:05] is he meeting over yet? [21:47:10] Metalcore424: not really [21:47:10] then we really need a decent site [21:47:11] no, not yet [21:47:13] ok [21:47:14] cool [21:47:24] I'm here now ^_^ [21:47:51] Hyronymus: Our sourceforge homepage can supply us with that for the moment. [21:47:52] ok [21:47:54] (http://tt2.sourceforge.net) [21:47:55] dang, phone again [21:48:06] For those not familiar with the SF file release mechanism, you check out the files you need from CVS, build them, upload them, and then they are available through the SF download system [21:48:10] (with the masses of mirrors, etc.) [21:48:17] what'd I miss? [21:48:23] Sourceforge really is a nice place. [21:48:35] Metalcore424: How's your scrollback? [21:48:52] So shall we take picks for the week's task? [21:49:00] okay [21:49:02] tasks [21:49:06] sorry, can't drop the phone right now... [21:49:10] fine, but I like summaries even better [21:49:12] ^_^ [21:49:23] well yes, I'd hope for a list around now [21:49:38] my task this week is learn C....or start [21:49:53] i started learning C some years ago [21:49:55] ok [21:50:01] still don't understand everything [21:50:07] * Research into gettext or similar module? [21:50:16] * Research into cross-platform api's? [21:50:24] * Research into PNG/JPG loaders? [21:50:47] i'd suggest the use of mingw for win32 enviroment.. [21:51:08] Ah yes we'll have many pieces of data to work with. We can have them all bunched up into one file, or sprawed across a directory structure.. Ah this doesn't matter this point. [21:51:16] mingw... [21:51:55] Jpl: if it has no runtime dependencies (DLLs, etc.) then we can build our binaries with mingw and distribute that way [21:51:55] minimalist gnu compiler for windows... [21:51:55] Jpl: good code would support both MSVC, BCC and mingw on win32 [21:51:58] reading.. [21:51:58] http://mingw.org [21:52:02] (cygwin has the dependency on cygwin1.dll) [21:52:19] no dependencies, only msvc runtime.. [21:52:40] Oh goody, using this we can write in fortran. [21:52:45] (Yes, a joke) [21:52:54] chirho: Even though MSVC has some strange "features" Wink, it probably is the best compiler for the win32 platform. [21:53:10] I'm using MSVC Introductory version [21:53:22] Although the Intel compiler gives a bit faster code [21:54:16] Hellfire667: Whatever. As long as the resultant binary can be run without having to grab additional DLLs, it doesn't matter for now [21:54:29] true [21:54:30] (well, no dependencies other than the libraries we use) [21:54:45] If we're hoping the same code compiles alright in Linux and W32, it's safe to hope the same code compiles under mutliple Windows compiles with minimal changes. [21:55:15] About Linux: do we want to use autoconf? [21:55:54] (I don't know enough about it) [21:56:02] well, the resultant shell scripts work nicely on any Unix system, and on cygwin [21:56:02] though if we did, someone would have to learn how to use it [21:56:18] Safe to save that for later, or should someone tackle it now? [21:56:26] But on the other hand: it doesn't aid us on win32. [21:56:36] hmm... they describe msys as cygwin replacement. [21:57:06] Hellfire667: win32 doesn't have quite so much subtle variation [21:57:43] hmm.. yes, that's true. [21:58:00] If someone wants to build it, we can always have a predefined makefile for some of those compilers [21:58:01] or appropriate project files included [21:58:08] Yep. [21:58:18] okay. [21:58:21] I've seen that by some places, such as some stuff just for Solaris. [21:58:58] That said, the configure script produced by atoconf should (theory) run on a Solaris system anyway [21:59:05] IF we'd make different modules for each platform, that would be fairly easy to accomplish [22:00:06] I've always assumed that Gnome/CDE/KDE, etc all look the same to the running program. Is that correct? [22:00:14] yes. [22:00:25] X = X, even if it's running Gnome or KDE [22:00:42] How about once XFree86 goes away? Smile [22:00:56] Smile [22:01:06] Well... If we're using SDL that won't be a problem. [22:01:37] Alrighty. So what do we want accomplished by/for the next meeting? [22:02:03] * Research into gettext or similar module? [22:02:03] * Research into cross-platform api's? [22:02:03] * Research into PNG/JPG loaders? [22:02:09] And * Research into autoconf? [22:02:36] heh [22:02:47] What about the UML? Would we like to do anything to that? [22:02:47] libpng? [22:03:02] Do we want them done by next week, or do we want some of these tasks to be longer-term? [22:03:14] umm [22:03:29] They ALL can be longer-term. It has to be done when we start coding. [22:03:47] ok I've come to the conclusion that lib???.org, where ??? is any three letter acronym, is a GNU library. Razz [22:03:57] In the mean time, we can make a model. [22:04:05] what about libXXX.org? ^_^ [22:04:05] Epistax: not really. what about libc? Smile [22:04:06] LOL Razz [22:04:09] what does xxx stand for? Smile [22:04:13] THREE letter acronym Wink [22:04:41] xenophobic xylophone xenon [22:04:42] xpensiveporn xpensiveporn xpensiveporn? [22:05:02] oh wait. seX seX seX. nm [22:05:06] * Hellfire667 DCC's xpensiveporn to Metalcore424 [22:05:07] enjoy [22:05:19] Dude he's 14. oh wait. enjoy. [22:06:00] *cancel* *cancel* *cancel* *cancel* *cancel* *cancel* *cancel* [22:06:05] whatever... BAck to topic Smile [22:06:08] very true [22:06:08] I'll look into autoconf. K? [22:06:09] sure [22:06:17] SF has a series of trackers for tasks, etc. Do we want to use them? [22:06:20] I can look into language techniques [22:06:30] chirho: Yes, please [22:06:32] It may be a good idea [22:07:29] So, do we want to fix the UML up? As in, deviate from the zugspiel code. [22:08:35] I'd like to see a "fresh" UML diagram. Not too detailed, just an overview of the program. We can discuss that diagram at a next meeting, and work it out in the weeks to come. [22:09:05] sure, so a high level view of the engine(s) interaction [22:09:15] (If we were working according to the ESA standard for software engineering, that would be the "logical model") [22:09:17] yes. [22:09:27] hehe use case diagrams Smile [22:09:33] bckz0rs [22:10:59] * chirho slaps SF [22:11:09] SF? [22:11:11] oh, right [22:11:11] They were kinda broken yesterday. [22:11:42] Sourceforge is down a lot lately. [22:12:15] Permission Denied [22:12:15] Access to this page is restricted (either to project members or to project administrators) and you do not meet the requirements to access this page. Please contact the administrator of this project for further assistance. [22:12:17] hmm.. (I clicked "add task") [22:12:19] Hmm... [22:12:22] Perhaps that's a priveledge that is dealt out? [22:12:44] That would probably be why I was slapping SF [22:13:05] Does Roll / Position bring with it accesss, or is it just a tag? [22:13:17] It's a tag [22:14:04] nothing more [22:14:16] I can dig up a list if people are particularly interested to see what's on the list [22:14:33] A list? of.. [22:14:41] roles that can be set [22:14:53] ahh gotcha. [22:15:07] I think everyone qualifies as developer [22:15:18] Unless someone on that list is art/music [22:15:29] you should be able to mess around with the task list now [22:16:10] I can get to the screen to post one, but I haven't tried actually posting. [22:16:19] \o/ [22:16:35] I'll add "my" task... [22:17:08] back [22:18:02] \o/ It works! [22:18:11] Aside from sourceforge fun, is there anything left? [22:18:23] I don't see it hellfire.. [22:18:40] strange! [22:18:46] Hellfire667: Did you really start it yesterday? Smile [22:18:51] Where did you file it? [22:18:56] he's very fast. [22:18:59] in the "research" subproject [22:19:01] Default task list view is only open tasks assigned to you (or something) [22:19:10] Hey... It's still march 13 over here! [22:19:15] gotcha [22:19:28] just that the start date is listed as 12-03-2004 Smile [22:20:03] ??? [22:20:19] 95453 Research into autoconf 2004-03-12 2004-03-26 0% [22:20:22] I know it's not necessairy yet and perhaps you can do it yourself but if you ever need a planning for this project you can always ask [22:20:24] excuse hte lack of tab damage [22:20:24] 95453 Research into autoconf 2004-03-14 2004-03-28 0% [22:20:26] lalalala [22:21:24] Hyronymus: Every hand is appreciated Smile [22:21:33] I'll try to add mine.. [22:21:56] just tell me when the time for a planning has come [22:23:17] right now I'm going to get some much needed sleep [22:23:26] I'm almost literally dead tired [22:24:42] They have drugs for that. [22:24:42] ok I added my task [22:24:44] dead tired is almost impossible [22:24:48] you'd pass out first, which would be sleep [22:24:48] I can see it! [22:24:50] I gave a different end date [22:24:58]