SauerWorlds

by CrazyTB on 04/19/2007 04:14, 35 messages, last message: 04/29/2007 21:55, 5273 views, last view: 05/16/2024 14:17

I was thinking about something that might be called "SauerWorlds". The idea is to mix coopedit with wiki, and add some touch of mmorpg.

The idea is to have a place that should store maps ON SERVER, and that should have some user authentication. Then, each map would have permissions (what users are allowed to enter/read this map, and what users are allowed to edit this map).

This way, a group of people might join to coopedit some map, but they don't need to be online all at once. Also, they can make some changes alone, and others will still be able to see the changes. I think this should help coopedit a lot.

Of course, there should also be some way of navigating those maps. So, a main "worldmap" (which could look like a city with streets, buildings, rooms) would have teleporters to other maps.

For example, an user might tell his friends: "Meet me at 4th ave, number 13, room 7. The map we will be editing will be available there."

Of course, this "world" can be abused somewhat easily, so there must be some way of avoiding abuses.

   Board Index   

#1: ..

by CrazyTB on 04/19/2007 04:16

BTW, there should be no more "sendmap". Since the maps should be stored on server, a /getmap would always get the most recent version.

Or, maybe, a /sendmap would mean "save this map on server now"

reply to this message

#2: ..

by Der/Nalia on 04/19/2007 04:18

Some RPG tips (I actively owned and programmed an RPG for 8 months)

Words that should not be in the Title of an RPG:

Worlds
Online
World
Mystic
Mirage
Destiny
Fantasy
Twilight

Reason: EVERYONE uses the above words

reply to this message

#3: Sauerville++

by c0rdawg on 04/19/2007 04:53

Sounds a bit like Sauerville...

reply to this message

#4: ..

by ..devoid on 04/19/2007 09:00

Speaking of abusing, wiki structures is safe from vandalism because they take backups. So a backup system is needed if vandalism becomes an issue.

reply to this message

#5: Re: ..

by CrazyTB on 04/19/2007 13:34, refers to #4

Yes, but would it be easy and feasible to implement a history of previous map versions?

It is easy for wikis because every page is just text, so it is easy to store previous versions and show differences between them.

This is why user authentication can be useful, so we only allow authorized people to edit the map.


And thinking about a name... Could be something like Mystic SauerWorlds Online beta (because "beta" is a must! every project nowadays have a "beta" in name!) ;-)

reply to this message

#6: Sauerville--

by Osbios391 on 04/19/2007 18:31

The last map update on Sauerville was my \"garden\" on 2007-01-17.

I dont think this project has any big future.

But I realy like the idea of CrazyTB. You just need somebody who (can) make this stuff. :/

BTW: For a new name you could continue the concept of sauerbraten.
Example: Kartoffelbrei (Ger.: mashed potatoes) ;)

reply to this message

#7: ..

by metlslime on 04/19/2007 21:14

I've been excited about this idea since the Cube days, and while Aard isn't interested in pursuing it, it would be cool if someone out there did.

I think the main tech needed is user accounts and user permissions assigned on a per-sector basis, where a sector is a chunk of world space of appropriate size. Sectors could be owned by one user, several users, or be public for all to edit.

Also, if it's not obvious, all functionality, including administrator functions, user login, and sending/receiving geometry updates from the server, should be handled in-game. If there's a version control system (useful if you choose to make some sectors public) then that should be in-game also.

One downside I see is that, if everyone can copy cubes from other people's sectors, then the incentive to be really creative and impress people is diminished because once you make something cool, everyone can copy it. Of course, the internet is like that too, and many people have webpages with the same animated gif or whatever. Maybe everyone will have a different concept of cool. Maybe people will get a reputation just by being known as the source for the most copied cubes -- a statistic like that could even be linked to a user. So maybe it'll turn into a 3D version of myspace instead of a 3D version of wikipedia.

Another approach is to be able to set "copy" permissions per sector as well, so users can disable copying on their sector by people that aren't owners, or something. But I think some copying is a good thing, and borrowing ideas and raw material from other artists is part of what makes an art community better than being an artist in a vacuum.

reply to this message

#8: okay...

by metlslime on 04/19/2007 21:31

i guess there might be some other technical hurdles to deal with, now that I think about it more.

1. the lighting model is one obstacle to realtime modification... will we need to keep relcalcing the entire level every time someone edits their giant tower or adds a huge radius light?

But it's tied up with another issue, which is that lights and geometry from one sector should probably not affect the lightmaps in any other sector -- otherwise you have an issue where people don't really have full artistic control of what their sector looks like, or they have to erect huge walls to protect their lighting. So if we say "each sector is lit independently" than it solves the control problem, and makes the lightmap updating a local problem that requires a quick lighting patch instead of a slow one that requires the entire world to be considered.

You might want to add a global sun light that does affect every sector, just to make it feel like a coherent place. And maybe you'd want to give neighbors some sort of shared buffer zone that they both can edit, which lets people create seamless transitions, in both lighting and geometry. It would requires some trust, but not as bad as trusting anonymous strangers.

2. There is also a problem where abusers can create extremely detailed geometry with expensive shaders and thereby kill everyone's framerate. But then again, even well-meaning people could create something that looks really cool but hurts framerate, so maybe there just needs to be a general solution to framerate problems, like having a distance culling or aggressive LOD system on farther away sectors.

reply to this message

#9: Re: ..

by Aardappel_ on 04/19/2007 21:40, refers to #7

curious you say I am not interested, since I have something exactly as you describe in my global project list. I guess if decline to do it right now, it sounds like I am not interested. It is just more work to do it right than you might think.

I think if this is done right, you'll quickly run into limitations of amount of geom in one sauer map, and amount of players on one server. Hence this should allow a system of linked sectors that are both independent maps and servers, and allows seamless transition.

Which would quickly turn it into a Cube-MMO. And if its an MMO, there'd better be something fun to do besides just editing geom.

reply to this message

#10: ..

by Julius on 04/19/2007 21:40

Yeah that is a cool idea. I have been toying with the idea of proceduarly and user created mmo worlds too for some time. IMHO that is the only way a open source MMORPG could really work.

The biggest problem with your idea is that it totally breaks the game aspect of SB and turns it into a virtual editable world (like second life?).
But I guess it could be doable pretty quickly with the code we already have in SB.

But I think a system where the world creation is actually part of a real game would be much more productive and a lot more fun too.

It would need a proper background story, like a world that is under constant attack by some time/space fabric dissolving extra dimensional aliens or someting, and the players have to extent the little that is left of the universe by waiving time and space...while fighting the aliens with big guns :p

It would need a different way of editing though, where more stuff is done procedurally and editing feels less like.. well "3D modeling". And it would need to be integrated into the gameplay mechanics (like power/money etc to build).
And it would be cool if building something is a certain way, like realisticly pysically stable, or is some abstract certain extendable rule following way, it would actually increase the "game value" of the structure. In that way there would be a certain puzzle/tetris like gameplay mechanic as another layer of the game.

Sorry for the long rant, and no I haven't got a playable alpha version of that, nor even started with it ;)

reply to this message

#11: Re: okay...

by Eivets on 04/19/2007 22:11, refers to #8

I have a similar sauerbraten mod project running since a while. Because of two little childs I don't have that much time at the moment, but it's progressing at some pace. Kind of the beginning of a (m)morpg. There are user accounts, multiple connected maps, simple quests, persistent online editing of the world etc.

The capability to edit a map can be limited per map and also in principle to parts of a map, although this is not so useful, as you agree.

Most admin tasks are not in the game itself, but you can instead use a web browser to connect to the server and can do most administration (like creating and placing npcs etc.). Producing a useful gui just many more gui elements and that's not my main focus.

As you said, the biggest limitation for online editing is that recalculating the lightmaps takes quite some time, and the lightmaps are currently also not sent, but the client needs to calculate them (on user demand). I experimented with background calculation of the lightmaps (works quite well), but this either takes very long or ruins the frame rate.

If someone is interested, take a look at www dot radig dot com/sauerbraten
There are some premade accounts and I can create accounts on demand (Self creation of accounts is almost finished).

reply to this message

#12: Dark Cloud... ?

by tentus_ on 04/19/2007 22:32

Wouldn't it make sense to have some terrain between each editable area, so that they
A) Don't clash visually,
B) Have some buffer room for lighting,
C) Give some breathing for distance occulling.

If you have an area that is designated as "yours" that is... lets call it 32 units wide, then 4 units of public area should surround it so that your gloomy lighting can get smoothed out for the 64 unit wide desert next door. Depending on the code available, these 4 units could make a relatively smooth transition from a gloomy swamp to a bright dry desert with little trouble- it might even be possible to have each user area output the brightness of their areas so that the public terrain automatically tries to match those edges and make a gradient in between. Even if such a feature was impossible, it'd just take a couple of hills to look decent.

If we're going to have an MMO like Julius described, could we have predefined continents with unique currencies? I've been working on some projects that involve changing currencies, and it's really engaging if laid out intelligently. Not to mention that at no point in human history have we all had the same currency (caveman bartering does not count as currency).

reply to this message

#13: Re: ..

by metlslime on 04/19/2007 22:53, refers to #9

Aard: Well, it's something we've talked about before. I didn't mean to say that you don't find the idea interesting, what I meant was that you were not interested in pursuing it. If that changes, great!

As for the technical considerations, I know that on top of simply creating a feature, there is a lot of work required to make it work in a good way and actually have the seamlessness of integration, and the simplicity of UI that is required for it to be worth doing in the first place. So a feature like "user permissions" or "version control" actually means a lot under the hood. Plus the unstated assumption that the tech must scale to support many users and large worlds.

Julius: i was imagining second life or snow crash's "metaverse" as a model becuase of its suitability to what already exists -- a good in-game editor, and the ability to edit in a shared space over a network.

The idea of having editing as part of actual gameplay is great, but requires actually designing a new game, including rules for how people can edit, and some sort of coherent design about how the editing actually benefits the player, and some sort of idea of how the game world won't look like an eyesore when people are editing with game goals rather than aesthetic goals in mind.

This last question especially suggests editing on a level completely different from what is currently there, where players edit in a very rough way and the game generates suitable artwork. Lighting is another issue in that the lighting is not realtime. Anyway, the challenges are surmountable but this is not "low hanging fruit" compared to the second life/metaverse idea.

reply to this message

#14: Re: okay...

by RealNitro on 04/19/2007 23:08, refers to #11

The dreams I had about a Sauerbraten MMO were much like the ones by Julius. Money to buy rough goods (units of wood, stone), and being able to build my own house, earn money building other people's houses (temples, castles), etc. But that's an MMO, which appears to be very hard to do.

I like this idea too, but the reason I probably wouldn't 'play' this (if it's supposed to be played) is that I'd rather not build 'maps' but build a city for a community.

Small note: Linden Lab is planning to Open Source their server-code...

reply to this message

#15: ..

by Eris on 04/20/2007 03:58

Why not a game without money? And without property? That would be innovative.
I've seen enough MMO medieval simulators... they all end up in hoarding.

IMO

reply to this message

#16: ..

by Eris on 04/20/2007 04:01

Hey im not done yet!

I though this might suit the topic discussed:

http://pr0t0n.homeip.net/~robertj/sauerville/

I dont quite understand the whole point of it haha

reply to this message

#17: ..

by CrazyTB on 04/20/2007 04:11

To keep it simple, and more "feasible" in short-term, I was thinking in user-permissions only for entire maps, and not sectors. The "main world map" could be read-only, with just "rooms" with teleporters to other people "miniworlds" (from which they can share their maps).

Of course, a sauer-like MMO that combines Second Life or even Active Worlds (this one is an old 3D MMO game...) would great, but for now I would like to see just some sauer-wiki (which could eventually evolve into a sauer-mmo).

User login could be done by simple (small) changes in sauer client code, or even by typing some commands at chat window (commands that can be mapped to some GUI sometime), maybe just like IRC login.


BTW, all these cool ideas about a big sauer world where everyone would be able to edit some sector remembers me a lot some (cool) projects I found at Internet:
http://www.mrwong.de/myhouse/
http://www.icontown.net/
http://www.kennethfejer.com/isocity/
http://www.pixeljoint.com/thejoint/
http://www.x-panded.com/pixeldam/indexpixelmoon.html
http://www.x-panded.com/pixeldam/

reply to this message

#18: ..

by Geartrooper on 04/20/2007 05:31

why not have a judging system to accept structures that are already lit. maybe a sub world for construction and a top tier for approved submissions.

robertj on irc did something similiar a while ago with sauerville. if you can find him he would be a good technical help. I haven't seen him around for a while though.

reply to this message

#19: Good idea

by rcdracoaway on 04/20/2007 17:16

Neat Idea, and so far the easiest way I see of implementing this is to use the largest map size and make seperate ares, make a central area, then make \\\\\\\"doors\\\\\\\" that warp you to a new area, edited or not.

reply to this message

#20: Re: Good idea

by Eris on 04/23/2007 07:07, refers to #19

I could be cool if none messed anything up for each other... I have allready my castle ready =)

reply to this message

   Board Index