MMO Communities (was RE: [MUD-Dev] MMORPG Cancellations: The sky is falling?)

[J C Lawrence]

On Tue, 20 Jul 2004 02:58:04 -0400
Derek Licciardi <kressilac at insightbb.com> wrote:
> Sean Howard wrote:
>> "Mark Mensch" <mark at larping.net> wrote:

>>> I look at the impacts upon a community and society - including that
>>> within virtual space.

>> I do too, and once you reach a critical mass of people, the
>> communities break down into smaller separate communities. Society has
>> a self defense mechanism that way. I've lived in a moderately sized
>> city in Florida, a tiny ass village in Japan, and in LA, quite a
>> large city, and no matter where I go, I see the same social
>> trends. There are differences, but people are people. On the
>> internet, this is still true.

> This references the following law from the Laws of Online World
> Design.

>   Community size: (http://www.legendmud.org/raph/gaming/laws.html)
>   Ideal community size is no larger than 250. Past that, you really
>   get subcommunities.

> I'm still not convinced that a 250 person group is the maximum group
> size one can have.  I'm of the opinion that the group size is
> inversely proportional to the depth of the community but that doesn't
> mean that a larger group cannot exist.  A smaller group can have a
> deeper connection with one another but large groups can have a
> shallow, easier to maintain connection.

The 250 number has some grounding.  In particular it partially relates
to Dunbar's number.  Dunbar's number itself derives from an academic
paper by Robin Dunbar, an anthropologist at University College of
London, entitled 'Co-evolution of neocortex size, group size and
language in humans".  A pre-publication version of the paper can be
found at:

  http://www.arts.uwaterloo.ca/~acheyne/dunbar.html
  http://www.arts.uwaterloo.ca/~acheyne/dunbar.html

A decent point summary by David Smith of the notion can be found at:

  http://www.drsnet.org/preoccupations/2004/07/dunbars_number.html

Additionally Christopher Allen has carried some of the ideas forward WRT
online communities at:

  http://www.lifewithalacrity.com/2004/03/the_dunbar_numb.html

ChristopherA cites Ross Mayfield:

  http://radio.weblogs.com/0114726/2003/02/12.html#a284
  http://radio.weblogs.com/0114726/2003/02/10.html#a281

Whose articles are interesting as they bring aspects of network theory
into consideration.  Others which continue the network aspects:

  http://www.zylstra.org/blog/archives/001183.html
  http://www.zylstra.org/blog/archives/001078.html

Zipf curve distributions (power curve) have become almost the assumed
model for most online forms:

  http://www.shirky.com/writings/powerlaw_weblog.html
  http://radio.weblogs.com/0114726/2003/02/05.html#a268
  http://radio.weblogs.com/0114726/2003/02/06.html#a277
  http://modelingtheweb.com/
  http://radio.weblogs.com/0114726/2003/01/25.html#a242

And yet there are exceptions, specifically some as aspects of the
blogging world which is normally considered to necessarily follow zipf
distributions (look at the Technorati graphs), instead follow a log
normal distribution:

  http://www.utsc.utoronto.ca/~farrell/blogpaperfinal.pdf

Log normal distributions are associated with network growth models that
provide more room for link-poor sites to grow richer.  I'm still looking
for analysis from the vantage of small-world theory. [1]

Other interesting commentaries on Dunbar's number and its application
and implications as related to social groups:

  Clay Shirky: http://shirky.com/writings/community_scale.html
               http://shirky.com/writings/powerlaw_weblog.html

There's been confusion in recent threads over what "community"
means.  Many assumed definitions, few to no explicit definitions,
and little recognition of the fact.  The simplest definition of a
community is probably a group of people who share a common
self-recognised identity metric.  Howver this makes for a nearly
useless definition for the social forms we're interested in as the
binding can be tenuous at best.  What is the "Slashdot community",
the "EQ community", or even the "negro community" or the
"people-with-big-earlobes community" in terms of the intersection of
network graphs and identity mechanics?  Things become more
interesting when the bindings are rich and tight enough that there
is not only a shared identity, but there is shared recognition of
the fact that the identity is shared.  They are not only members of
the group, but everyone in the gorup knows the others to also be
members, and they recognise each other.  Dana Boyd used the word
MAXIMUM in her eTech Talk [2]:

  (Dunbar) found that the MAXIMUM number of people that a person could
  keep up with socially at any given time, gossip maintenance, was
  150. This doesn't mean that people don't have 150 people in their
  social network, but that they only keep tabs on 150 people max at any
  given point.

This is a fairly tight binding requirement.  Dunbar in fact is
careful to define what he means by "community", and restricts his
discussion to such tightly bound networks with strong inter-personal
relationships.

Yes, there are clans and guilds with hundreds of thousands of
members, but for the purposes of Dunbar's number and most of the
interesting analysis that isn't actually interesting.  The community
of people who wear blue jeans is also large, self-selective, and
prone to social network effects (cf Jordache jeans in the
mid-1980s)...  Especially in terms of player retension, in-game
social activity, value of communications media and group services,
etc etc etc in our games its is the smaller tighter social forms
that hold sway.

  [1] MUD-Dev itself typically hovers a bit under 200 active posters
  (more than 3 posts per month on average)

  [2] Dana Boyd's eTech Talk:
    http://www.zephoria.org/thoughts/archives/2004/02/11/my_etech_talk_revenge_of_the_user.html

--
J C Lawrence
---------(*)                Satan, oscillate my metallic sonatas.
claw at kanga.nu               He lived as a devil, eh?
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.
_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev