written consensus procedures

process

written consensus procedures

Version: 5/04/2003, 3:29PM

Given that, we may end up with incompatible ones specialized for each medium, or flexible ones that work in any of the written media, and need not be unlearned or relearned when one is shifting media. Clearly some tradeoffs of access, ethics, habits and culture must be made.

Serious attempts to unify semantics across media tend to stress several features:

1. reliance on normal recognizable natural language keywords or semantic tags, rather than abbreviations that must be learned.
2. keeping questions and answers separate and marked clearly, and having a collection of "frequently asked questions" (FAQ) that is available all the time at a reliable location.
3. keeping action items (proposals, bids, votes, offers and requests) marked separately, and organizing agendas around these, in time to discuss the action they imply.
4. opening and closing issues, stating positions and arguments and evidence, stating concerns or limitations or constraints, and calling for consensus, must not be confused with action items and are strongly encouraged to not be in the same item with facts, action items, or statements of actual consensus
5. keep attributions (who said what) absolutely clear, and if conventions are used (">" quoting, "name:" quoting, attachment, indentation) then these are used consistently and identities are never confused - altering quoted text is never acceptable unless it is simply to abbreviate (using the ">..." or similar conventions).
6. keep statements of identity, causality and morality ("is", "was", "since", "therefore", "must", "should") to a bare minimum - using these terms permits anyone else to restate them into two or more separate messages, and delete the original message, or move it into a 'story' or 'source' file that others can then access.
7. using E Prime (verb "to be" is banned, replaced by "becomes", "remains" and "equals") to prevent dogmatic statements re: "what is", and to encourage careful differentiation of future-to-present, through-present, and future-to-past thinking.
8. adjusting subject lines, using "re:" effectively and accurately, and including explicit links to prior threads on the same issue/s or action/s.
9. avoiding Blind CC's, as they make it unclear who is being talked to - and for the same reasons never posting messages that were sent to anything other than an absolutely public group (e.g. removing all messages sent to a semi-private group before one can make its archives visible to others not known to the group).
10. accomodating messages about the conversation itself ("meta")
11. avoiding 'votes', 'surveys', 'polls' as a way to decide actions on the spur of the moment (as not everyone concerned is consulted) - instead splitting the thread or conversation for those who take a position, and those who take the counter-position, so that both factions (or 'tendencies') can develop their position separately.
12. making participation possible across wider spans of time (via the FAQ method, archives, stored messages) and making it clear how to continue to add material to archived material.
13. having some way to finally end a matter (or bring it to temporary or limited closure) while making it possible for new people joining a group to review how it was settled - or to raise new evidence, which has the potential to re-activate a dormant issue, i.e. to move beyond the existing 'consensus' with little or no risk of offense to those who worked hard to create it.
14. having reliable ways to close a thread or sub-thread so that it is not visible to the rest of the group if it's not a widespread concern.
15. marking dissenting positions as such so that they too can be annotated in future, checked against actual results, and etc., and so that there is accountability for the outcomes of the majority decision - perhaps a dissensus can be used to provide a basis for comparison with the consensus?.
16. in threaded systems, avoiding deep threads with multiple interwoven concerns.
17. clearly identifying different versions of a proposal or policy resolution or offer - keeping time frames clear.

Another current method is that of the crit software, which allows any web page to be annotated with threaded commentary. This makes it easy to rendezvous with other groups using crit to comment on that same page, regardless of any formal organization affiliations. This is to be preferred when an external entity, e.g. a corporation or organization whose behavior is at issue, is the focus of the action. If it's a larger concept, this works less well and is likely to fragment discussion. For such comment on larger concepts and their perception, one might use for instance crit on the wikipedia page for that concept, e.g. the crit argument regarding the wikipedia entry on persuasion technology. Such conventions are technically quite demanding, and may require very good computers and high bandwidth to give satisfactory performance for everybody.

This suggests a few 'don'ts' that apply to written consensus:

1. don't use technology that not everyone can learn quickly and use with a high level of proficiency.
2. don't use technology that not everyone can buy, e.g. requiring 128MB of RAM, latest processor, etc., e.g. MSN Messenger (unless of course you can get access to the same capabilities using other software, e.g. trillian.cc or jabber.org
3. don't make essential email or web URLs invisible on a printed page, i.e. don't assume the page is clickable on the web, it may have been printed and handed to someone.
4. don't use threaded discussions to debate things that require, by their nature, constant assessment of 'what we now agree to', e.g. any matter of great controversy or difficulty to track
5. don't let anyone use sysop or moderator powers to set the agenda
6. don't make decisions until everyone has seen the debate and had enough time to consider the various issues and positions
7. don't make the decision just because it's 'on the table' - very important to have time horizons, deadlines, known clearly in advance.
8. don't move conversations across media, e.g. from a wiki to an invisible mailing list, with different participants allowed, unless this is a deliberate and clear policy used for all such escalation.
9. don't let the older or founding members refer to old policy decisions or experiences that led to them without forcing them to provide the links to URLs that explain the context - if this is not followed strictly, the group will ossify and die as the decisions of these members become impossible in practice to challenge
10. don't let new members raise the same old issues without instantly referring them to the debate on that issue, and giving them a way to comment on it without bothering the group.
11. don't let FAQs or other compiled statements of undisputed positions fall out of date, or you're wasting the time of those who consult them, and expressing disrespect for those who do not spend all their time watching and debating the issues.

etiquette

• don't use ad hominem argument unless you are absolutely sure that it is a certain person's agenda and beliefs, those that strictly contrast with those of the group, that lead them to make observations or proposals. Ideally, don't use it at all, even then.
• don't use profane language or accuse anyone of motivations or intentions (you can't read their mind, you can't intimidate them to silence)
• don't confuse statements of ethical judgement for calls to action, and don't confuse your own morality ("should") with what "must" happen.
• don't exploit superior access to, or trust of authorities with, technology, in order to curtail or change the nature of debate, e.g. do not use IP bans or removal from lists to enforce your views - a moderator or sysop should in general NOT BE ALLOWED TO SPEAK ON THE ISSUES AT ALL, except to ask questions and compile the answers (failures to do this honestly should result in removal).
• don't go running to authority when you get a virtual nose bleed - if you want to settle tough issues you're going to have to be very tough.
• you MUST LEAVE A FORUM the SAME WEEK you see sysop authority abused and this abuse tolerated - this is the ONLY ABSOLUTE RULE of written consensus - those who take it dead seriously also should state that they do not report physical violence or intimidation directed towards such a sysop or moderator who abuses those priveleges. Those who cut off an open debate, suffer the inevitable consequences of that: decision made by intimidation not debate. Failing to let this feedback loop operate properly is very unwise, and leads to the instant rise of groupthink based on who controls the sysop's powers to censor commentary.
• don't take sysop power if you're not willing to be beaten up over your decisions, and maybe killed. Yes, it should be exactly that serious, if the decisions you are making are of any import. Keep this in mind when you decide whether to limit access by an individual yourself, or whether you take it through some process the group itself has chosen (the only way to avoid feedback falling on you personally).

If you are not willing to discuss difficult matters to the point of anger or abuse, and you are not willing to take responsibility for moderating such discussion and taking heat off of moderators or sysops, then you probably should avoid written consensus itself.

Foolish decisions are the only possible result of sticking to the etiquette at the expense of the agenda, favoring the group's current members over the results the group is expecting to achieve for others (customers, citizens, refugees, animals or ecologies, whatever). That said, one must know how to exclude, and the recommended way to do that is the same on reality game shows: disapproval voting.

Properly handled, a disapproval voting system can reduce the powers of a sysop or moderator to nil, and make everyone equally responsible for the ejection of an intolerable participant. If all participants bear equally this responsibility, then no one can be targetted ad hominem (to whatever level of slander, libel, or violence), and the group can expect to maintain its solidarity.

This is critically important for any group that regularly undergoes pressure of a physical kind, e.g. strike actions, blockades, demonstrative protests, outdoor expeditions, combat.

Failure to make etiquette mirror body risk, and failure to keep to etiquette norms 'under fire', and in particular falling prey to the idea of 'virtual community' (in which all participants regardless of bodily risk levels get an 'equal say'), are fatal to most groups. They cannot hope to behave the same way under fire as they do on the written communication medium, unless they have carefully aligned their etiquette to the real life situation, not the writing.

ethics

Ethics are so different across groups that nothing more than this single recommendation will be made here: create an ordered list of priorities for different 'rights', keep it on a single page, and do not explain it, but be very clear what 'right' you are talking about on each line. Call the completed page the "ZZZ Ethic" where "ZZZ" is the name of the organization or group, and make everyone either re-order the items, or agree to them in the order they are.

When complete, post the Ethic (singular) along with the archives, and ONLY PERMIT ETHICAL DISPUTES TO PROCEED WHERE SOMEONE HAS A CASE THAT THEY HAVE A HIGHER RIGHT THAT IS BEING OVER-RIDDEN BY A LOWER RIGHT on the list, e.g. if "the right for women to be heard" is higher on the group's list than "the right for the group's time not to be wasted", then women must have a way of objecting they are not heard, even if everyone thinks it is a waste of time. That simple. Such an Ethic is a constitution for the group, and pre-makes the kinds of decisions a Supreme Court might make balancing the rights of a nation's legal constitution.

But, a single Ethic per group is a lot simpler and easier to create.

logistics

Guess what? We still have paper ballots for a very good reason: everyone can use them without any ability other than recognizing a name (not even reading, really). Inclusion is job one in logistics for decision making systems.

Here are some suggestions for logistics, which must be altered if you find in your setting they don't work with your medium or rate of transmission of written communication, or make etiquette or ethics you require impossible in practice. These are just guidelines:

proposal titling and versioning

use normative language so that goals of the proposal are clear, e.g. "Safe Streets Act" can be challenged on the basis that it does not lead to Safe Streets, whereas "Omnibus Act 434343" doesn't give anyone a real chance to criticize it based on overall intent.

avoid normative language in proposal titles where trust is low, to avoid appearances of promoting a particular measure over others.

Use a numbering scheme that reflects the time action will be taken, and when the proposal will expire (all proposals must have a top end time limit)

writing style and format

comment flow: best practices

rules of procedure (when things get tight)

followup after decision

time/action/issue/position/argument/evidence/source/authority

avoid attachments