server creation kit

[steuben]

like the blurb for the this fourm says, what would we like to see in the finished game, or atleast provided for it. and this is one of those things is again looking aways down the road. roughly about the time that planeshift goes mid to late stage beta.

basically a \"kit\" that would allow those who are so inclined to create their own worlds around planeshift. a collection of walkthroughs/guides and utilties that could be used to create new and varied worlds and challenges.

and yes, i am aware that, they are not currently looking for new servers. this is one of those looking down the road things.

[Black_rose]

it would be cool but do you mean like a make your own world??? or just a new server?? cause making uour own world would be cool to start a char off with nothing but tour selected items and then play mabye a special game with (death battle)

[steuben]

both, or either.

i was thinking along the line of 3rd party but connected servers, an authorized content kind of thing.

but the mechanics of building a server and a whole world are pretty much the same. all i\'ve really been able to glean is that some heavy duty 3d modeling software and a couple of purpose built utilities are required. not that i\'ve investigated it very hard.

but, your idea may be interesting. kind of an \"it was all a dream, or was it?\" quest. especially if the dream character is a different class from the main one, and is assigned at the start of the quest.

[lyistra]

One or two minor problems with the connected server ideas.  The biggest is cheating.  Mud portals are entirely possible, mmorpg portals as easily so, distributed computing as well.  However, it\'s fairly simple to set up a single sever so for a moment, for your character who doesn\'t have good stats on the other server, you level up substantially, then return to the other.  Depending on your level of expertise, you can fairly easily spoof an IP address, using a malformed packet or three, (and no, I won\'t go into detail on that save with server admins) and modify your character on the other server (a major problem)   So, you\'re left with the choice of nothing transferring between servers save the character (which more or less obviates the need for portals) or a system of \'trust\' certificates.  Which only work insofar as they\'re followed.    So the mud portals thing is a major danger as far as security goes for now, short of advanced encryption (rsa or some other system) on the server ends, synchronized so the things would be far harder to break, preferably with a strong random key.

At which point your transmission rate between them is slowed down, and you end up with a much higher possibility of lag between servers. (more into a probability)  There are several methods that can get around this, which again, is something I\'d likely only discuss with developers.  

As far as the possibility of a graphical development kit, it could be done, I suppose, but the server admins are still trying to get the rules worked out, and remove massive playability bugs.  Down the road, true, but I\'ve learned one thing about muds.  They are *never* done.  Ever.  And the moment you think they are, the place is dying.

So you have a choice of a larger development team, or a less complete product in the long run, but a good developement kit.

In the long run a development kit would be a good idea.  It would allow the much more rapid creation of areas, however, I\'d rather not have templated areas involved in an RPG game.  Again, a level of complexity issue. *smiles*  Good idea, really, hope they can implement it sometime, but I\'m not holding my breath for now.  And in some ways, I\'d prefer they kept that to themselves, as the nature of such a kit would likely require access to the rules modules in order to actually build NPC/Weapon/Armor/monsters and balance ti.

Though I suppose they could put out one with a loadable library module, XML or Ruby or some other scripting language module for holding data within it, so you can define your own rules system. (*chuckles* Still trying to figure out how to build an LPC backend for on-the-fly rules modifications)


another .02 tria.
(darned rats, gotta stop hitting them so hard in the head)

Lyistra

[steuben]

pretty much. like i said this is one of those things for one ps goes mid to late beta. so there is probablly a few years before it begins to become an issue. so i\'m holding my breath either, darn oxygen addiction.

and it isn\'t like it is a perfect idea. well yet anyways. given that there are atleast some obivous security/play problems with 3rd party servers. but these aren\'t to hard to work through. but there might be a better thread to do it in...

http://www.planeshift3d.com/wbboard/thread.php?threadid=9989&boardid=11&sid=09505c75b41387d7585a0330f64bafb4

i started it back in the days of mb, but the concept remains valid.

[lyistra]

Excellent points in that thread.  I can see a few ways around the server issue, mostly keeping the servers secured and in one location.  And another method using hard encryption to process the system.  There are several very secure encryption methods that could be used, the question is synchronization, and data loss.  Part of that could also be highly compressed for transmission, then encrypted, (as compression produces a fairly randomized stream anyway).    The further encryption at that point simply manages to make it far more dificult to spoof the server.

I\'m not saying it can\'t be done, just has a number of major caveats.  After all, they have the netwide distributed computing modules avaialble, and many other tools for the job.  Most of the work would likely be in the synchronization systems, and the overall arbitration unit that allocates users to specific servers, and the security systems.  Then... lots... and lots.. and lots of bugtesting.  Part of the reason I suggest synchronized systems, on the encryption, is it\'s far more difficult to break into, and further, it is even more difficult to crack into.  Especially if the server IDs themselves are transmitted encrypted.

Another couple of considerations.  I think it could be made to work, really, it\'s just that the servers would have to have an \'overlord\' system to make sure that all the rules modules, etc, were up to par.  And I know few administrators who would give access to a server with such a program, due to security risks involved, even in chroot jails.

Probably the best option would be a network distributed computing system, with each section of the network optimized for its purpose, and a dispatcher, but that perhaps is going too far.  And would require quite a few dedicated computing systems.

*chuckles* and stops beating the poor suffering horse.. not dead yet, and hopefully won\'t die.  Bandages the wounds as much as she can, then smiles.*  I\'m not attacking you, just offering options, and playing devil\'s advocate.  Or perhaps simply pointing out security loopholes, and potential remedies.