Dear Denis,

I have tried your model and it looks really nice.

My opinion is that to answer your request you probably need to set up a clear research question first. E.g. which mechanisms for appointing taskable volunteers might lead to better results? Or another one - who is better in finding of new issues for different amounts of reporters and observers? And what about if the issues are changing and growing in space (like forest fire)? Then, depending on the problem, the proper way of improving the model features could be selected.

Another helpful thing might be a little bit of experimentation with the model - a good way how to approach this would be the BehaviorSpace function of NetLogo that allows you to execute predefined experiments automatically and save results. Then it is possible to analyse them to find more general patterns of behaviour.

Good luck, Viktor

Dear Victor,

Thx. for the tip with "BehaviourSpace" - I'll take a look at that.

As for the research question, I guess the main question is "how to get the most out of my volunteers without loosing them". Which probably means that I'll have to play with the motivation and algorithms which will maximize the volunteers motivation for the same amount of resolved issues (or other tasks). We'll have to calibrate this against real volunteers later.

Other obvious questions are for instance "how does trust and effect of  learning by doing affect the quality of information?" or "which types of tasking should we implement in the model and in the real world?"

Ok, then I agree that the motivation probably should be the next step.

I a little bit struggle with the concept of "issues". What should it be? I think that their types and importance might play a crucial role in the motivation as well.

Dear Denis,

I've never used NetLogo, but I find the idea to apply it to this domain very interesting. I have approached the differences between crowdsourcing and crowdtasking from the perspective of volunteers' involvement in disaster management efforts, trying to understand different forms of contribution and outcomes. Perhaps this paper can offer a few suggestions on different roles and levels of involvement: http://link.springer.com/chapter/10.1007/978-3-662-45960-7_19

Cheers,

Marta

Following the Victors suggestion, I have developed a second model which takes into account several aspects influencing the volunteers motivation. The result is available at http://modelingcommons.org/browse/one_model/4241

It's simple and a bit rough around the edges, but I think that it goes in the right direction. Volunteers can get tired and recover, and they can also be driven to despair by assigning them too tedious tasks.

I have even implemented a (very simple form of) long-term motivation evolution, which is even more interesting than the single-event motivation to volunteer managers.

By the way, a new version of the first model with greatly improved GUI and documentation is at http://modelingcommons.org/browse/one_model/4237#model_tabs_browse_info

Some tips, how to get the most out of your volunteers with no loosing them. In order to maintain your volunteers, try to give some rewards such as Status, Access and Power (SAP), they will be motivated, as per contributions, this will work

Keeping your volunteer campaign is most successful when you tap into players’ (customers) intrinsic motivators and you should determine the goal of your contributors and they value and then establish a suitable reward schedule.

These are my suggestions:

Status refers to the various levels or ranks to which players can aspire. Remember when you couldn’t wait to get into the next grade at the start of the school year? How about getting a promotion at work?

Access refers to opportunities that allow players to interact with a private or a special way with your company or service. For example, allowing select players to view content before others, or to enter an exclusive contest. There are numerous tactics for using access to separate and reward ranked players.

Power refers to special entitlements or rank that allow top players to be in positions of superiority over others. For instance, a highly ranked player who, based on his/her title, can donate virtual goods to help newer players in their quest. Power drives engagement, loyalty, revenue, influence, time spent, the fun, the ability to edit.

In your question there are two key words, from the first question it was about “How can one model the crowdsourcing/ crowd tasking?  And the second question is about "how to get the most out of your volunteers without loosing them"

These are the two terms; Crowdsourcing and volunteering:

Crowdsourcing: is the act of sharing a job or function that was once the domain of a specific agent with distributed masses through an open call e.g. Wikipedia.

Volunteering: participants are volunteers in the true sense of the word in that they: come and leave of their own free will, are unknown to the administrators before they volunteer – i.e., they can come from anywhere, not just from the administrators’ domains, and do not expect substantial compensation. By allowing anyone on the Internet to join as their contributors.

An addition to the previous mentioned techniques, these include technical issues that need to be addressed in making volunteer computing possible and effective. These technical issues can be classified broadly into:

Accessibility (making volunteer computing as easy, open, and inviting to volunteer as possible)

Applicability (making volunteer computing useful in real life), and

Reliability (making volunteer computing work in the presence of faults and malicious volunteers).