Skip to content

Meeting_2010 02 20

Ivan Spasov edited this page Jun 6, 2022 · 3 revisions

The 2010-02-20 meeting was held on Saturday, February 20th at 15:00 UTC on IRC.

Present were:

Minutes

Project management

  • (Unstable) Release
    • General
      • Auria: Even an unstable release will attract level designers. The latest development snapshot is more fun than the latest stable release.
      • bkralik: Agrees with Auria, favors new release.
    • data/AUTHORS has been created by mathnerd314 in 6333, based on DataFiles.
      • ohnobinki: This documentation is enough

        Did you look at it? Thats just a list with some pseudonyms, impossible to link to persons or email accounts without searching the mailing list, the wiki and svn log. The goal is to have all the information in one easy to find place in case someone need to contact the creator of some file. --WolfgangB 17:36, 20 February 2010 (UTC)

      • octo: Some information on the graphics would be good. → He'll look into it.

        I would say don't worry about graphics. All the graphics we use were contributed to the SuperTux project. It's not like the issue we had with sound files where the source was “various royalty free FTP sites and CD-ROMS”. I would say the same for the levels, but apparently we included some levels from supertux.info without properly verifying the licence. Once the source/licence of these levels are determined we should stop worrying about documenting each and every level and graphics file. Going forward we can do this if want, but there's no need to document existing files. - sik0fewl 14:27, 24 February 2010 (UTC)

  • Decide whether to finish up forest (ghost) world or to improve world 1 some more

Gameplay

  • Feedback for the Yeti and Above the Arctic Skies levels.
    • Is Yeti level too hard? SuperTux:Community Discussion#Vote
  • Decide which new enemies to implement/rewrite first (bombfish, snowman, ninja snowball, forest enemies such as owl/zeekling...)
    • Neither behavior, graphics nor code exist for those, according to grumbel.
    • grumbel: Not a high priority; fix existing badguys first.
    • E.g.: Use Mrs. Snowball more, remove auxiliary limiting blocks.

Source code

  • Split physics, drawing, etc. apart
    • Which design pattern to use?
  • Find some way to change object constants without recompiling (sprites, Yeti bounds, velocities of various things, …)

Next meeting date and time

  • Every Saturday at 15:00 UTC?

Log

16:01 < octo> Okay, it should be 15:00 UTC now, right? 16:02 < grumbel> yes 16:02 < octo> Hey grumbel :)  16:03 < octo> Let's give Mathnerd314 one more minute - I understand it's early in the morning for him ;) 16:04 < ohnobinki> oh, it would be 16:06 < octo> Okay, I'd say five minutes is plenty ;) 16:06 < octo> Hello everybody and welcome to the second meeting in 2010 ;) 16:06 < octo> As last week, I'll be writing minutes at Meeting`` ``2010-02-20 16:06 < octo> The first point on the agenda is: (Unstable)`` ``Release`` ``-`` ``now?16:08 < octo> Last week the majority was in favor of creating an unstable release soon, but to make sure licenses of the data files are documented properly first 16:08 < octo> The appropriate bug is #578 16:09 < octo> Unfortunately I was skiing until earlier today, so *I* didn't make any progress.. Has anybody else done anything in this direction?  16:09 < ohnobinki> I think there is now adequate documentation of the state of the licenses of data/ 16:10 < octo> ohnobinki: Do you mean data/AUTHORS? 16:10 < ohnobinki> yes  16:11 < ohnobinki> also, supertux is playable as it is 16:11 < octo> Yeah, the sounds look okay to me in that file 16:11 * MMlosh is somewhat here 16:11 < octo> Hey MMlosh :) 16:11 < MMlosh> just without keycaps on keyboard... 16:11 < octo> Some note on the graphics would be nice though 16:12 < octo> As far as I know the vast majority of the graphics has been done by grumbel, right? 16:13 < ohnobinki> yup 16:14 < octo> Does anybody know how many graphics were done by someone else and who that was? 16:15 < ohnobinki> I think not :-/ 16:15 < ohnobinki> and checking the log for details doesn't really seem feasible 16:16 *** mabu [[email protected]] has joined #supertux 16:16 < octo> Okay. I'll add some note to that file and we can see at the next meeting if that sounds okay to everybody. 16:18 < octo> Okay, the next point is: Decide`` ``whether`` ``to`` ``finish`` ``up`` ``forest`` ``(ghost)`` ``world`` ``or`` ``to`` ``improve`` ``world`` ``1`` ``some`` ``more16:18 < octo> Does anybody know what that is supposed to mean? 16:18 < MMlosh> how many of us are there anyway? 16:18 < MMlosh> I am not sure if 3=enough to resolve anything 16:19 < octo> I'm currently at four, though grumbel is very quiet ;) 16:20 < ohnobinki> yeah, I can't really help with that question 16:21 < octo> But, yes, I agree that it'd probably best to take decisions to the mailing list 16:22 < ohnobinki> yeah, I can't help with #3 either 16:22 < ohnobinki> and it seems that #4 (meeting times) actually matters.... 16:22 *** yetanothername [[email protected]] has joined #supertux 16:23 < grumbel> Well, from my point of few me basically have to delete almost everything and start from scratch when it comes to ghost-world/world1 16:23 * MMlosh has heard something whether do dump forest world and improve ice world OR finish forest world to make it releasable 16:23 < ohnobinki> ok, that's what I thought ;-) 16:23 < grumbel> I don't consider forest world releasable, not even close 16:24 < ohnobinki> then should the next release just be of iceworld in its current state? 16:24 < ohnobinki> *next unstable release 16:24 < grumbel> I don't even consider it improvable, some parts are salvageable, but that it 16:24 < octo> It's maybe not releaseable in a stable kind of way, but since the world can be played and completed I don't see why we shouldn't release it as an unstable version 16:24 < octo> grumbel: What's so horrible about it? 16:25 < grumbel> octo: basically everything 16:25 * MMlosh is unable to give any example of everything 16:26 < ohnobinki> I can see that the forest world doesn't seem to have any goal to it other than to collect keys 16:26 < octo> I actually think the files are mostly nice, most levels are nice or at least decent.. 16:26 < octo> I don't like the new badguys much, though 16:28 < octo> Okay, let's try the other way around: What exactly is there to improve in World 1? 16:28 < MMlosh> worldmap 16:28 < grumbel> Again, basically everything. 16:29 < octo> We can't do everything at once, therefore some smaller steps would be helpful 16:29 < grumbel> At the very very least all levels need to be up-scaled to work with larger resolutions 16:29 < grumbel> but thats really just scratching the surface, I'd prefer to completly redo most of it 16:29 < octo> Okay, so at least xy tiles high? 16:29 < Auria> If I may allow to speak, even though I'm not a SuperTux regular contributor : I think level design and programming are two separate things, totally. SMC has non-programming users make levels, and the programmers focus on fixing bugs and adding features. I think it works well this way. A problem with ST is that it looks dead, so it's not very attracting for users. If programmers can make a development release, that _might_ attract more usrs to do l 16:30 < ohnobinki> auria: your message was truncated at ``more usrs to do l'' 16:30 < octo> Auria: Thanks for your input :) Due to length restrictions your line got cut off after _might_`` ``attract`` ``more`` ``usrs`` ``to`` ``do`` ``l…16:30 < Auria> levels 16:31 < Auria> ends ther ;) 16:31 < ohnobinki> :-) 16:31 < octo> Auria: Oh, okay ;) 16:31 < octo> Auria: Yes, I agree. A new release doesn't have to be very polished.. 16:31 < grumbel> Auria: at this point in time I consider it a *good* thing. The more level builders we have, the harder it will be to improve the core 16:31 < ohnobinki> so we have too many people touching src/ and not enough touching data/... 16:32 < octo> It should just show some activity to attract new contributors 16:32 < Auria> well indeed contributions from everyone is very touchy. with STK we have many more contributions than _good_ contrinutions ;) 16:32 < bkralik> I am only friend of MMlosh, however i have same feeling as Auria... Supertux need a new release, most of users play only old repository/non-svn verison, which is not new... 16:33 < Auria> but here and there come a few gems that are very helpful to the devs 16:33 < octo> I don't think SuperTux should be a game engine only, I'd prefer a complete game with game engine AND levels 16:33 < Auria> IMO, the latest SuperTux unstable is more fun than the latest stable... which is what most people get from their package managers 16:34 < Auria> (yes, even if not fully polished) 16:34 < grumbel> octo: I consider game engines a waste of time, as you can't design an engine without a game 16:34 < grumbel> octo: thats of course not to say that some parts of supertux should be better separated and easier reusable 16:35 < octo> grumbel: If I understand your concerns corretly, you fear that level designers start to use unfinished / unwanted features which we then can hardly change / remove 16:35 < octo> grumbel: Is that roughly your point? 16:36 < octo> I would switch to a scheme where we create unstable releases much more often.. 16:36 < octo> Release one new unstable version every two months, for example.. 16:37 < octo> Release 0.4.0 in early March, 0.5.0 in early May and so on 16:37 * MMlosh would prefer unstable preview too... every second user now compiles SVN version.. that's wrong 16:37 < octo> Bugs in the development releases are fixed in 0.4.1, 0.4.2, ... 16:37 < ohnobinki> then 1.* can be milestone 3? 16:38 < octo> Once we feel that a release is getting to a state we like and which reflects the design document well, we'll create 1.0.0 as stable release 16:38 < Auria> anyway, to close my point, I think SuperTux currently has enough featrues to make interesting levels, so why wait before trying to make better levels? IMO that's what ST needs most atm, not new features 16:38 < octo> Then we continue with development releases 1.1, 1.2, 1.3, ... 16:38 < Auria> </Auria's_opinion> ;) 16:40 < grumbel> octo: Its not a fear, its a fact. Thats why supertux has made zero progress over the last few years 16:41 < octo> grumbel: So which features would you like to remove? 16:41 < MMlosh> octo, do not ask that 16:41 < octo> or change 16:41 < grumbel> Auria: supertux needs *less* features 16:41 < Auria> grumbel: ok, so we somewhat agree ;) 16:41 < grumbel> Its currently full of tons of half done stuff, basically nothing works completely 16:41 < Auria> I haven't checked the level editor in ages, last time I wasn't able to even get it to run 16:42 < Auria> but that may have changed 16:42 < grumbel> Auria: it runs, but its not exactly as comfortable as it should be 16:42 < MMlosh> Auria, I am not sure... I cannot test on Apple's operating systems 16:42 < Auria> MMlosh: was on Ubuntu 16:42 < grumbel> octo: basically all forest stuff should be removed 16:43 < octo> grumbel: I'm afraid you can't sell that to users 16:43 < MMlosh> Auria, sorry... I guess it was somebody else who hat troubles on macs 16:43 < grumbel> octo: its the only way to make progress, in its current state I consider it a complete lost cause 16:43 < ohnobinki> grumbel:who is going to write the replacement levels after forest world is removed? 16:44 < grumbel> ohnobinki: the point is to cut the thing down to a point where you can build something again 16:45 < octo> grumbel: So what forest`` ``features are we talking about here? The tiles? Some game objects? Levels? 16:45 < grumbel> ohnobinki: who builds stuff in the end is another issue and not really that important, as the important part is bringing it back into a stage where you can work on it again 16:45 < ohnobinki> would releasing a copy in its current state truly create enough problems with users expecting forestworld to stay around? 16:45 < grumbel> octo: all of them 16:46 < octo> *sigh* 16:46 < ohnobinki> you could release an unstable ``forestworld'' version and then in the next commit chop the forestworld away 16:46 < grumbel> ohnobinki: thats more or less the plan 16:46 < ohnobinki> there just should be some way to make sure people don't expect forestworld to stick around 16:46 < octo> I think cleaning up the broken / half implemented / unwanted features one at a time is much easier than throwing everything away just like that 16:46 < ohnobinki> grumbel: ok, so unstable release _now_ is a go 16:47 < grumbel> ohnobinki: I could do without, but if somebody else is doing it I won't stop them 16:47 < ohnobinki> ok 16:48 < octo> I have to admit I still don't understand the horrors of the forest world 16:49 < ohnobinki> grumbel: ok, so mathnerd314 has access to the berlios download mirror? or who does? 16:49 < octo> Is is the forest world in general, i.e. its look and feel? Because I have never experienced problems with, say, switches or flying platforms 16:50 < grumbel> ohnobinki: no idea, member list on berlios should show that 16:50 < grumbel> octo: you are not looking close enough. 16:50 < octo> grumbel: Then enlighten me 16:51 < grumbel> octo: tiles are abused, graphics don't fit together, level design is mostly ugly, code isn't pretty either, scripting is used in the wrong places, ... 16:51 < grumbel> even the loading times suck 16:52 < MMlosh> and Ice loading times do not? 16:52 < octo> That sounds more like the levels and the design should be improved.. I was always kinda fixated on the implementation.. 16:54 < octo> But ugly graphics can be changed at basically any time 16:54 < grumbel> MMlosh: they aren't good either, but forest world seems a lot worse due to larger levels I guess 16:54 < octo> Improvements on levels doesn't influence other levels - at least not neccesarily 16:54 < grumbel> octo: its not ugly graphics, its ugly everything 16:54 < octo> grumbel: There I agree: The Forest levels are too big 16:54 *** ced117 [~ced117@opensuse/member/ced117] has quit [Quit: Quitte] 16:55 < grumbel> its specifically the world loading time, not the level loading time, which is large as all levels have to be loaded to query their name (curse of text based level format) 16:56 < grumbel> octo: worldmap sucks, levels sucks, enemies suck, level design sucks, there nothing level once you clean up 16:57 < octo> grumbel: So you like the overall theme? I.e. Tux in a Forest is generally okay with you? 16:57 < grumbel> SuperTux isn't the only FOSS game with this issue, its a common problem, you get 50 pieces of crap, instead of 1 polished one 16:58 < grumbel> octo: it kind of was my idea to begin with, but I am not all that comfortable with the forest theme, I prefer the ice theme 16:58 < grumbel> but that something that could probably be fixed with better graphics 16:59 < octo> I like the general idea of having several worlds as a long term goal 17:00 < grumbel> thats the goal, but some themes just fit better then others 17:01 < grumbel> but as said, thats probably just a design issue, not a problem with the theme itself 17:02 < octo> Okay 17:03 < octo> I still think that improving the Forest world incrementally is easier than starting from scratch 17:04 < grumbel> Its easier to draw something on a blank canvas then one that is already filled with stuff 17:05 < octo> Then we can always replace the Forest with a Forest II world when that second world is done 17:05 < MMlosh> yes.. why not... 17:06 < grumbel> I don't want to care about backward compatibility or have all those graphics in the way 17:07 < MMlosh> and forest world seems nice to me, ice world is a bit ... all the same everywhere. Perhaps because there is not much scripting? 17:07 < grumbel> MMlosh: scripting sucks 17:07 < grumbel> MMlosh: and thats why ice world should be redone 17:08 < Lazure> when you go to the forest world, it's almost like you're playing ad ifferent game. everything changes, you have slopes suddenly, you can go up and down suddenly, it's no longer linear. 17:08 < grumbel> i.e. add verticality to levels, more interesting caves, etc. 17:08 < Lazure> more tiles and tilesets for ice world would be nice 17:08 < Lazure> only one map at all uses that extra ice tileset that was made 17:09 < octo> Actually that has changed a bit in the lest few days 17:09 < octo> Look at the (new) Mysterious House of Ice, for example 17:09 < octo> Or The`` ``Escape17:09 < octo> s/lest/last/ 17:11 < octo> Okay, so as far as I understand, (big) changes are required on tilesets and levels before we can call something finished17:11 < grumbel>    http://pingus.seul.org/~grumbel/tmp/md5/6c54813834f98bdf9abd9d7564ccff05-galapix-screenshot-0000.png 17:11 < grumbel>    http://pingus.seul.org/~grumbel/tmp/md5/677e174d88b851844d7fc73570667109-galapix-screenshot-0001.png 17:11 < grumbel>    http://pingus.seul.org/~grumbel/tmp/md5/c72a01dde4b0bd614600f82a41559b2a-galapix-screenshot-0002.png 17:11 < grumbel>    http://pingus.seul.org/~grumbel/tmp/md5/c7d6262a3f24ce3c497139f58c13702a-galapix-screenshot-0003.png 17:11 < grumbel>    http://pingus.seul.org/~grumbel/tmp/md5/66b2b53b8ae6e12148bef83c42b3f946-galapix-screenshot-0004.png 17:11 < grumbel>    http://pingus.seul.org/~grumbel/tmp/md5/9b6a1bc3acc77093b7ab2764ab1014da-galapix-screenshot-0005.png 17:11 < grumbel>    http://pingus.seul.org/~grumbel/tmp/md5/c58c334c7bc2e3e825c67b9274b490b6-galapix-screenshot-0006.png 17:11 < grumbel> already posted those on the last meeting 17:11 < octo> But creating an unstable release can be done before that has been done. 17:12 < octo> grumbel: What's new in #4? 17:12 < grumbel> the background 17:13 < octo> Thought so.. Looks nice :) 17:13 < octo> I like #2, too 17:13 < octo> #5 looks like Above`` ``the`` ``Arctic`` ``Skies17:14 < octo> Mathnerd314 wanted feedback on that, but unfortunatly he doesn't appear to be here 17:14 < grumbel> some tiles for 6 are in svn since today 17:14 < MMlosh> yeah... 2,4 are nice 17:14 < Lazure> wow, i like a lot of those screenshots 17:15 < Lazure> i'm assuming people are making tiles for those if not already? 17:16 < octo> I'd say be skip Feedback`` ``for`` ``the`` ``Yeti`` ``and`` ``Above`` ``the`` ``Arctic`` ``Skies`` ``levels. since Mathnerd314 isn't here and go to the next point 17:16 < octo> That would be Decide`` ``which`` ``new`` ``enemies`` ``to`` ``implement/rewrite`` ``first`` ``(bombfish,`` ``snowman,`` ``ninja`` ``snowball,`` ``forest`` ``enemies`` ``such`` ``as`` ``owl/zeekling...)17:16 < octo> As far as I understand, those are enemies of the design document 17:16 < octo> Is someone working on graphics or code for one of those? 17:17 < Lazure> i like the new above arctic skies level with the tileset and parallaxing 17:17 < octo> Lazure: Prefix that with Mathnerd314: so he gets a highlight ;) 17:17 < Lazure> Mathnerd314: i like the new above arctic skies level with the tileset and parallaxing 17:18 < octo> Mathnerd314: I find the many layers a bit confusing. Especially the nearest layer is irritating to me 17:19 < octo> Mathnerd314: Other than that the level looks great :) 17:20 < Lazure> Mathnerd314: i like the effect of those nearest layers tho. I guess they could be made darker in color tho 17:20 < octo> So, any comments on those new badguys? 17:21 < octo> If we wanted to implement them, which order would we use? Wait for the graphics, than code behavior or the other way around? 17:21 < grumbel> we have neither finished graphics nor code nor behaviour 17:21 < grumbel> the important part is the behavior 17:22 < octo> Yep.. 17:22 < octo> Is there something on those in the wiki? 17:22 < grumbel> anyway, those don't have high priority, first the existing enemies should be fixed 17:23 < grumbel> not much more then graphic and a very short description 17:23 < octo> grumbel: Okay, which badguys are those? 17:23 < octo> Okay 17:23 < grumbel> http://supertux.lethargik.org/wiki/Milestone_2_Design_Document/Enemies 17:23 < octo> All of those should be changed / fixed? 17:24 < octo> Mrs. Snowball, for example, behaves as advertised ;) 17:24 < grumbel> no, but if they already work they should be documented 17:25 < grumbel> mrs.snowball for example should simply be used 17:25 < grumbel> as we have lots of level with classic snowball and blocks to stop them from falling of the cliff 17:25 < octo> Mrs.`` ``Snowball 17:26 < octo> Actually, in Above`` ``the`` ``Arctic`` ``Skies we have Mrs. Snowball *and* limiting blocks, Mathnerd314 17:28 < Lazure> yeah, those limiting blocks are kinda silly 17:28 < Lazure> gives tux a piece of land to sit on to avoid being attacked altogether 17:29 < octo> Okay, any more comments on those badguys and their possible implementation one day? 17:29 < Lazure> i would like to see more levels utilize some of these enemies, and possibly remove the 'hacks' placed into the levels to 'fix' the existing ones, such as those limiting blocks where they were not intended. 17:30 < Lazure> if a map was intended to not have a block sticking up on both ends of every platform, they should rightfully be removed in favor of the snowball being replaced by mrs.snowball 17:32 < octo> Okay, anyone eager to discuss the source code points or can we postpone that to the next meeting? 17:35 < octo> I'll take that as a no.. 17:36 < octo> What about a new date for a meeting? Any opinions or should I ask the mailing list? 17:37 < grumbel> octo: about source code and stuff: my plan would be: 1) remove everything unneeded 2) fix lots of bugs in existing stuff 3) start enhanching/rebuilding snow levels 4) add new enemies 17:38 < grumbel> at 2) we might also change graphics API a bit and do other changes such as better extensibility 17:38 < grumbel> but its really stuff that has to wait till we do a big fat cleanup of everything unneeded 17:38 < octo> grumbel: Well, I have a hard time figuring out what unneeded features are 17:39 < grumbel> everything that isn't needed ;) 17:39 < octo> .oO( Why did I know that? ;) 17:40 < octo> There's a bunch of stuff I don't know how to use, but I don't know if that makes it unneeded.. 17:40 < grumbel> for most part its just all forest, as nobody has spend much time improving iceworld 17:40 < octo> The Decal object, for example 17:40 < grumbel> decal is very important, its just not very well integrated into the editor 17:40 < Lazure> i still believe we should make a toggle for auto-run 17:40 < grumbel> its basically just a graphic that you can stick somewhere in the level for decoration 17:41 < octo> grumbel: What does it do? 17:41 < octo> grumbel: Oh, okay 17:41 < octo> grumbel: So just some graphics that can be used without turning it into a tile or tiles first? 17:41 < grumbel> so that you don't have to create a new badguy for each piece of decoration 17:41 < grumbel> yes 17:41 < octo> I see 17:42 < grumbel> the glowing stuff and the crytsals in http://pingus.seul.org/~grumbel/tmp/md5/c58c334c7bc2e3e825c67b9274b490b6-galapix-screenshot-0006.png would be done with decal 17:43 *** claymore_ [[email protected]] has joined #supertux 17:45 < octo> Okay, so what about you tell me three things that are unnecessary and I'll try to remove at least one until next week? 17:46 < grumbel> octo: we have to first get the devel release out 17:46 < octo> grumbel: Okay, that's fine by me, too 17:46 < grumbel> and then decide on how we branch/fork/whatever the old stuff out so it doesn't get lost 17:46 < Lazure> there's another problem that needs to be fixed majorly 17:46 < Lazure> i find if i die enough times, the game crashes 17:46 < grumbel> I wouldn't mind a libsupertux-extra-stuff or whatever 17:47 < Lazure> doesn't matter if i'm running it in a VM or running it in real ubuntu, it just randomly crashes after i die a few times. 17:47 < grumbel> as long as its not part of the core and I can work on the rest of the game without ever having to touch extra stuff 17:47 < octo> Lazure: Can you create a stack backtrace for us? 17:47 < Lazure> a what? 17:47 < Lazure> i'm not a coder, i don't really know that stuff lol 17:48 < octo> Lazure: Okay, maybe a coredump is easier ;) I'll tell you in a minute 17:48 < octo> grumbel: I don't quite understand what you try to do.. Why move stuff into a lib? What stuff precisely? 17:49 < grumbel> octo: all the stuff that gets removed 17:49 < octo> Yes? Why put it in a library instead of removing it? 17:50 < ohnobinki> because people here seem to like that stuff 17:50 < grumbel> If somebody wants to keep it and wants to invest time to keep it in a external library I wouldn't stop them. 17:50 < grumbel> I wouldn't want to invest the time myself 17:51 * MMlosh cannot imagine badguys locked in separate library... 17:51 < octo> I don't think that'd be a good thing 17:51 < grumbel> shouldn't be that hard, they just have to register themselves in the factory 17:51 < MMlosh> and they you have to import their graphics... 17:51 < ohnobinki> hopefully the factory abides by the equal employment opportunity standards 17:52 < grumbel> you can already map new graphics into the data/ without changing the data/ directory itself 17:54 < MMlosh> and that will work? sounds a bit strage 17:59 < octo> I still don't have a clear picture what kind of stuff would get removed. But since the answer is probably everything again, :q 17:59 < MMlosh> definitely 17:59 < octo> … I'm not sure whether it's worth asking 18:00 < grumbel> octo: everything that isn't in the M2 design document 18:07 < grumbel> in the end work should always start on the M2 document before it gets turned into code

Category:Development

Clone this wiki locally