Crowds

As part of research into Large Scale CVEs, we are looking at trying to incorporate and support crowds of users. This page presents an overview of some of the ideas and concepts.

Contents

• Why crowds?
• What are crowds?
• Spatial Crowds
• Taking advantage of the nature of CVEs
• Non-spatial crowds.
• Some examples.

Why Crowds?

• Why should we consider crowds. Well, possibly the first and most obvious reason, is that if we are trying to create social spaces, then crowds will exist, just as they do in everyday life. And in real life, crowds help to provide atmosphere, entertainment, they are a core of society and if we wish to make CVEs enjoyable and 'inhabitable' to many users, then they will be a core of the CVE also.
• Secondly, if we are talking about collaboration, then we assume that means there are a number of people. So, we may wish to question if there is any benefit to be had in working as a crowd. I tend to think of current systems as supporting many 1:1 relationships, as opposed to a 1:M or M:1 or M:M relationship.
• Thirdly, we can consider crowds, and the use of crowds, from a system perspective. If we are looking to develop a system that supports upwards of 1000s of users, then the concept of a crowd might provide a natural abstraction in terms of system saving resources. In other words, they could be a part of the scalability issue of CVEs.

Using crowds, we can then investigate new means of navigation, collaboration and issues of consistency, persistance and subjectivity, and examine the emergent properties. Evaluation criteria can be defined in a number of ways, but the relationship of how crowds behave in the real world as opposed to a CVE is one of the more interesting ideas.

What are crowds?

In terms of defining a crowd, we initially use a basic definition used in Crowd Psychology literature.

In other words, people become a crowd, when they assume a focus of interest, or as we have decided to call it, an Object Of Interest (O.O.I.). An O.O.I. might be a single artefact or a composite of artefacts and users. For example, a particular painting in a museum might be an O.O.I, or the playing field, teams, umpires and ball in a football match might be a composite O.O.I.

We also look at crowds from 2 different perspectives, spatial and non-spatial.

Spatial Crowds

Spatial crowds fit in with the concepts presented by the Spatial Model. There is some spatial relationship and binding between members of a crowd within a world. Within the current implementation of CVE, the concept of regions, abstraction and secondary sourcing support the creation of spatial crowds.

Taking advantage of the nature of CVEs

Spatial crowds are all very well, but the question remains can we take advantage of what CVEs offer us in our use of crowds?

If we return to our definition of crowds, we can say that there are effectively two unwritten rules that we assume. One is spatial proximity, the other is time. However, we can bend these rules using a CVE, and doing so opens up the possibility of creating Non-Spatial Crowds. This in turn may offer us an insight into alternative means of collaboration. However, it also means that we then need to consider issues such as consistency, persistance and subjectivity.

Non-spatial crowds.

Non spatial crowds allow us to develop the idea of crowds of people based on their interest in something (an O.O.I. perhaps!) or each other, or any particular (non-spatial) relationships which they may have. For instance, a non-spatial crowd might be a specific team in a football match (their interest relationship is belonging to a specific team) whereas a spatial crowd in the same example, might be all the members of all teams on the pitch at the same time.

One of the elements for defining non spatial crowds, is that we define an interest space as opposed to a normal cartesian coordinate space. Users and artefacts populate this space, and their proximity to each other is represented by an 'interest relationship'.

We can then produce a connectivity graph, which represents our interest space, users and artefacts, and their relationship.

Some Examples

Team Games.

The members of a football team could form a crowd, but members of the opposition should not be a part of the same crowd. If we followed a spatial concept, this would (potentially)not be possible. If a member of the opposition was sufficiently close, then they would enter the crowd spatial boundary. However, if we use a non spatial approach, the interest relationship is defined as team membership, and we can redraw our crowd boundary.

Travel Agent.

The user enters a world where people are showed around by a travel agent. People may travel as visually one crowd (to the non member), but their audio awareness is dependant on their interest. Based on this, while in a crowd, the user will hear the tour in the language of their choice. If they make no choice, or do not pay the agent, they will be allowed to partake in the tour, but will not receive any audio from the agent.

Virtual Museum.

Members of the public are shown around a gallery filled with artefacts. The artefacts represent specific O.O.I, but can also be used to represent logical O.O.I. In this instance, a user may wish to choose to express an interest in the "work" of the creator of a certain sculptor, or express that their interest is in painting as opposed to sculpture.

The tour of the gallery takes place as crowds, and is determined by registering interest upon entering the gallery, in the registration room. If no specific interest is registered, a user joins the crowd for the full tour, otherwise the tour is tailored to the users interests.

These then are just a few examples. There are many more that can be explored.

Links

The following are links to related pages:

MASSIVE2/CVE

The Spatial Model.