[MUD-Dev] MMORPG/MMOG P2P design

Crosbie Fitch crosbie at cyberspaceengineers.org
Tue Mar 4 09:31:58 New Zealand Daylight Time 2003

From: lynx at lynx.purrsia.com

> Didn't your earlier posts and the discussion on that site suggest
> that yo uwere looking at/after a trust/reputation scheme?  That
> would be a different type of design than a hierarchial authority
> system.

It would operate in addition. There is a hierarchical 'chain' of
responsibility/ownership. However, trust/reputation can be measured
irrespective of this hierarchy.

> It also accounts for the possibility that people might want to
> choose to play out different roles in a universe's management.
> This person might want to be the webmaster and do the webpages for
> the community website, which might also host the chats for the
> game.  That person might be content providing the invisible but
> extremely important database support.  Still another person might
> not have the resources to host any kind of 24/7 service but might
> have a lot of cool ideas for content, so he or she would want
> content creation tools.

Naturally, there'd be content creation tools, so yes, a different

> They all require different packages, and making the same software
> serve all uses only results in bloated software and a much bigger
> download time.

Ok, I'll unwind my generalisation a tad. Everyone gets the same
p2p/distributed system node component, irrespective of whether
they're a player, sharer, or what. There may be different front-ends
depending on platform. Content editing tools may also be separate.

> Sure you can!  If a game is 'winnable', then you want to
> eventually allow the game to be won and start over with another
> game so people can start fresh, especially so if the game is
> balanced such that the person who takes the lead becomes stronger
> than other players, and therefore more able to keep that lead.

I do not envision 'winnable' games, or games that are reset. Perhaps
games may become boring and players leave them, but this is
termination through attrition rather than deliberation. This is
mainly because such a global change as a reset seems very tricky to
implement (think of millions of PCs and the achilles heel of being
able to perform a reset at all).

> Different instances of the same game might be run by different
> people, creating a different feel.  One instance of Pern might be
> warm and social, focusing mainly on relationships between
> characters.  Another might be action-driven, with struggles for
> leadership and duels of words and knives.

Yes, but I'd call these different universes (one copied and tweaked
from the other). I'm just saying you shouldn't be forced to divide a
universe up into shards simply because of scalability problems.

> Sure, having one big world with lots of people in it is cool, but
> the small worlds can be fun too.  Not everything has to be big to
> be good.


> Well, think of it as working up a proof of concept then.  What's
> the smallest it can be and still demonstrate that a P2P MMOG is
> practical?  What can you do with it that will interest people in
> playing even if it isn't absolutely huge and filled with content
> that would take a dedicated team years to build?

It may not start off as a game. It could start off as a
collaborative content creation system. I am trying to get funding
for such a thing, but it's not looking promising.

> An interesting system, but I think missing the point here.  A
> virtual world is built by a dedicated team of creators and
> administrators, for the enjoyment of their players.  They may want
> the right to decide who should be accepted into their list of
> people with administrative access, and who remains merely players.
> They aren't necessarily going to charge people for access, but
> they do want to control access to the experience they have
> created, perhaps to ensure the experience is seen in the way that
> they intended, and not redistributed, altered, and copied, perhaps
> in a malicious or parodical context.

Yes, I'm thinking that for every universe there would only be one
user that had the ability to add original content and game scripts.
However, this user could choose whether to open things up or force
everything (other approved creators' work) to be funnelled through

Even so, there wouldn't be anything to stop someone making a copy of
the universe and altering the copy (parody or otherwise). Even so,
it's not so straightforward to persuade the players to move over (or
get any players at all).

> In a purely P2P system, content has to be distributed to all nodes
> in order to be used.  Per your suggestion above, all nodes running
> the same copy of the software, this suggests you are leaning
> toward such a distribution mechanism.

I think so. Though no node needs every bit of content at any given

> Digital art auctions cover recompense for the release of one's
> artwork, but they don't guarantee the integrity of the virtual
> world experience.


MUD-Dev mailing list
MUD-Dev at kanga.nu

More information about the MUD-Dev mailing list