Transmitting directives among participants during evaluation in a purely random fashion cannot ensure that trolls, unscrupulous initiators, or a virtual gang of cooperative (or uncooperative) participants won’t attempt to manipulate an exchange's outcomes in their favor. Ideologi's directive sorting algorithm, MAIA (Mediation Agent for Iterative Analogy), will ensure that the distribution methodology of directives is weighted by any and all historical contact between participants and initiators during past evaluation phases within an exchange as well as past exchanges.
Here is a sample of some of the information that ideologi will assess to determine how directives are transmitted among participants...
- Bell curve percentile ranking that the participants have attained in previous exchanges
- Average score that the group has received from other participants in each phase of evaluation in previous exchanges.
- Average score that the group has given out to other participants in each phase of previous exchanges.
- Average number of exchanges that all participants have previously joined.
- Average number of times that each participant has played with the same initiator in previous exchanges.
- Average number of times that each participant has played with each other in previous exchanges.
- Average score that participants have given each other when they’ve played each other before in previous exchanges.
- Contact information of participants (e-mail, addresses, surnames, etc).
In order to optimize the level of consensus regarding the outcome of each exchange, MAIA will also have the ability to increase or decrease the number of times a particular directive is distributed within each phase of an exchange based on the level of deviation in the scores it received in the previous evaluation phase. So, for example, if one directive receives almost identical scores from each of the participants who reviewed it in the previous phase, then the number of participants determined to review it during the next phase could be reduced from the standard number of participants evaluating it. The system will then calculate the number of ions received by that participant based on prorating their score based on a standard number of participants. Conversely, if the scores received by a particular directive during an evaluation phase has an excessively high level of deviation, then the number of participants assigned to review it during the next phase will be higher than the standard number. In either case, the total number of evaluations made per participant will be equal to the number of directives evaluated per phase as established by the initiator of the exchange.
The order in which each sample of directives is listed for evaluation by participants is based on the order in which they were submitted. If they choose to edit their submission before the beginning of the exchange process, then the moment the edited directive is submitted will be identified as the actual submission date. The order of directives will then be adjusted accordingly.
If a participant does not complete evaluating and scoring their allotment of directives within the time allowed in the latest phase, then their own directive is automatically rejected for that phase and all of the Ions accumulated by her in that exchange will be eliminated. Any directives left unevaluated shall automatically receive the median of scores they received by other participants evaluating their directive during that phase of evaluation. Then a new phase will take place, and so on, until the last phase has been completed.
The following is a rough sketch of a communication ethics policy ("CEP") which will be adhered to within exchanges initiated and operated by Ideologi Foundation. It's based on a philosophical theory of ascending consequential reciprocity. Ultimately, the CEP will be defined by an ongoing series of exchanges coordinated by ideologi incorporated and open to all participants.
2. A participant submits a directive that could cause imminent harm to another person, animal, or species, or threatens Ideologi’s ability to operate. The exception for any of these scenarios is if the accused participant can provide an argument that her exchange is for the purpose of protecting a group, species, and biosphere or serve the greater purpose of aiding the participants of ideologi or protecting it from imminent harm, respectively.
3. A participant identifies themselves directly or indirectly within the contents of her directive.
4. A participant tries to propose an offer to bribe the evaluators of her directive directly or indirectly either within the contents of her Directive or out in the Internet.
5. A participant publishes malevolent content (as defined by CEP) within the content of her directive that is not directly related to the associated exchange.
In addition, algorithms will be employed to detect malevolent "swarming” on the part of unscrupulous bands of individuals. Sanctions to those participants involved will then be determined by ideologi incorporated.
IMPORTANT: In the case of deployments that are independent of ideologi foundation (as is their right via the GNU GPL licensing of ideologi software, initiators are able to use or create a CEP of their own choosing.
Any participant of a particular exchange can volunteer to help enforce its CEP, provided that the initiator chooses to make this available to their participants. If so, then volunteers (called monitors) will be randomly assigned to investigations regarding ethical violations detected by participants within an exchange.
If three or more participants of an exchange perceive that a directive violates the CEP established by the initiator or operator of the site, then that directive will be ejected from the exchange at the end of the phase in which the third violation claim was made. The participant of the directive in question will then be notified that their directive is being evaluated by a randomly selected group of volunteers equal to 3 times the number of violation reports. None of the original notifiers of the violation are allowed to be monitors of the same directive.
The monitors will then review the directive in question, as well as the exchange's objective, to determine whether it can be identified as a violation. They are then given 100 Ions to be distributed among only two judgment directives as they see fit: "Positive" for violation & "Negative" for no violation. The length of the monitoring phase is the same length of time as the phases of the exchange.
If a majority of the Observation Ions accumulated are Positive, then the directive in question has been determined to be in violation of the CEP. The participant who submitted the violating directive will loose all Ions accumulated in the same exchange. the violating user shall be given one violation on their record, loose all ions (and profit) accumulated from the exchange thus far, and placed on notice for the period stated within the CEP. In the case of the CEP proposed for exchanges operated by Ideologi Foundation, if participants collect three violations on their ideologi record within a single month, they shall be banned immediately for participating in all exchanges for one month. The only exception to this rule is for participation in exchanges that are already in their evaluation process at the moment of ruling for the third offense.
If most of the Observation Ions accumulated are Negative, then the directive in question has been determined to not be in violation of the CEP. Therefore, the participants that originally marked the directive as “Violation” will each forfeit half of the Ions they've accumulate from the phase in which they falsely accused of violating the CEP. This is to designed to reduce the chances of accusations with low-confidence and to protect against cases of false accusation.
If the number of Positive and Negative Observation Ions accumulated during observation are equal (a statistical rarity), then the participants who marked the directive in question as a violation of CEP go about their business and the user that submitted the Directive is allowed to keep any points scored in the exchange through that discovery phase.
![[frdm] Support SFLC](http://www.softwarefreedom.org/img/support-sflc.png "Software Freedom Law Center"){kind=small}
