This is an attempt to lay-out what is involved in the agreement process. There are meta agreements, which are about controlling the agreement process. Arguably there are meta, meta, agreements, which are required for any form of (meta) agreement process.
Agreements are between individuals, who are assumed to have continuity and consistency of identity which is sufficient to treat them as an individual. Individuals can be groups, which have a sufficiently defined identity and rules for behaviour (such as how members of the group vote on group behaviour, or enter, or leave, the group). Individuals are typically humans, but can be machines.
The agreements are typically between humans and machines, or machines and machines; agreements between humans constitute 'civilization'.
Care is needed to ensure power relationships do not interfere with the forming of agreements, or result in coerced 'agreements'.
"We will attempt to cooperate and communicate. If possible directly. We will attempt to ensure that no one involved in the agreement process comes to harm as a result of the agreement process."
If this top-level (meta meta?) agreement isn't in place then any form of agreement is very difficult to negotiate. You could say "any form of civilized agreement", but civilization can be a bit of a tricky thing to define...
"Any agreement must be understood by those involved in it. If the involved are a group, understanding must apply to 95% or more of group members."
If an agreement isn't understood, it isn't valid. But, there will always be circumstances where some cannot understand, for one reason or other. If this is the case then those (rare) individuals (5% or less of those involved in the agreement) must have effective proxies who do understand the agreement, and will work in their best interests.
The need for understanding means agreements must be kept as simple as possible, and short. If an agreement looks like it is getting too complex then it must be broken-down into simpler, shorter, agreements. Agreements which are simple and straight-forward changes to, or build on, already understood agreements can be a good idea.
There is no reason why an agreement cannot have accompanying (clear, simple) examples, and a statement of aims and purposes. This can include (simple) references to previous agreements that this agreement replaces, and why there was a need for replacement.
Note that agreements, as defined here, are direct, as opposed to delegated, such as having technical, political, or legal representatives.
"Agreements are only subject to change as the result of a technically-informed reasoned discussion. Where agreements involve humans the discussion must include ethically-sound reasoning."
'Reason' means evidence-based, as opposed to conviction-based, discussion. Appeals to authority (proxies?) are no substitute for understanding. Personal attacks ditto, along with the other techniques of 'political debate', such as abuse of statistics, and not comparing like with like.
'Technically informed' means at least all the immediately underlying agreements to this agreement must be understood by those involved in the discussion. Some agreements require technical knowledge and expertise, sometimes at quite a deep level, to understand the consequences of making them.
'Ethically sound' means that human issues cannot be disregarded while making agreements. Agreements which do this at least wont work well, and may not work at all, even if they appear otherwise technically sound. In addition to basic human bio-mechanics, there are issues like pacing, privacy and security, and more or less complex social requirements.
Agreements might appear to be a simple matter on the surface, but when you look closely they affect many different layers of interacting social and technical systems. Unless the individual agreements are clearly structured, simple, and understandable there will be failures in the use of any system.
Understanding and controlling agreements requires meta agreements, agreements which handle how agreements are created and changed. There is also a need to ensure that agreements are explicit, as implicit agreements may not be equally understood by all those involved in them.
The human tendency to turn even highly-complex agreements into the mental equivalent of reflexes, and 'common sense', road traffic regulations might be a good example, can mean that digging-out the explicit agreements so they are properly documented is quite hard work.
(c) ROMsys Ltd, April 2013, permission given to use for non-profit making purposes
Back to 'Notes & Papers' index