[MUD-Dev] [BIZ][TECH] worlds.com gets patent

John Robert Arras johna at wam.umd.edu
Sun Apr 29 10:35:40 New Zealand Standard Time 2001

----- Original Message -----
From: "Kwon Ekstrom" <justice at softhome.net>
Date: Sat, 28 Apr 2001 19:58:21 -0600

> From: "Jessica Mulligan" <jessica at mm3d.com>
>> At 12:05 AM 04/26/01 -0700, Bruce wrote:

>>>   http://www.delphion.com/details?&pn=US06219045__&s_clms=1


> I just took a look at this, and I agree that it won't stand if
> challenged.

> Bruce later mentioned that it would take time, energy, and money
> which alot of MU* based games don't have.

>  <snip>1. In a system for interaction between a plurality of users
>  in a three-dimensional, computer-generated graphical space</snip>

> Shall we say... Duke3d, Quake, Starsiege, Mechwarrior, (etc, etc,
> etc) Very generalized and it touches the basics of gaming since
> computers started speaking with each other.

> Shall we say... VRML?

> Anyway, lets just say that enforcement would bring several
> industries down on top of them.

Delurking here for a moment. The thing about these patents that is so
insidious is that you have to take the claims as a whole. Yes, 99.9
percent of their patent has been done before, but they have one piece
that might be new:

  "determining a maximum displayable avatar count for the target

  "determining a total avatar count for the server,"

  "when the total avatar count is greater than the maximum displayable
  avatar count for the target client, limiting the number of avatars
  processed by the target client to the maximum displayable avatar
  count, wherein the step of limiting is performed at the target
  client; and"


  "The method of claim 1, wherein the step of limiting is a step of
  limiting the number of avatars to a number N, the step further
  comprising the steps of:"

  "identifying excluded avatars, wherein an excluded avatar is an
  avatar to be excluded from interaction with the local avatar
  regardless of the relative positions of the excluded avatar and the
  local avatar; and"


  "determining which avatars other than the excluded avatars are
  closest in the graphical space to the local avatar."
So, their idea is to limit the number of avatars a client can see by
capping the number based on the client and server capabilities. I
think this even overrides relative locations or LOS. For example, I
assume that if a client can handle 10 avatars, and 11 people walk into
your LOS and strike up a conversation, that client will only be able
to see and interact with 10 people. So, then if the 11 other people
are shifting positions slightly, avatars might be blinking in and out
of existence right before your eyes.

Needless to say, I'm not terribly impressed with this method of
clipping what a client can see. However, does anyone here who has
experience making 3D worlds think this is a good idea? Maybe I just
don't know enough to realize that this is a good idea, since it looks
pretty stupid to me as evidenced by the example in the previous

The main problem is that the company got a patent for a very general
well-known idea, except one part used a very picky, specific and
nonobvious (probably because it's not very good) algorithm. Then what
they will do if they want to enforce it is broaden the claim about how
the avatars are limited to a claim that their patent covers any method
of determining what avatar information should be sent.

They wouldn't even have to win, since it would be up to the defendant
to prove that they don't use this particular method of determining
what avatar information gets sent. That would mean you would have to
explain to a judge why your method differs from theirs, i.e. why their
avatar cap is different from what you do. Good luck. :( That's the
problem, since the only part that other games might not use is the
part where there's a hard limit placed on the number of avatars that
can be seen.

That's how you go from a patent which has a trivial new component to a
patent covering a broad well-known category of programs.


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

More information about the MUD-Dev mailing list