Decision Making

36 views
Skip to first unread message

Michael Kubler

unread,
Jun 14, 2011, 6:43:06 AM6/14/11
to zm-global...@googlegroups.com, Zeitgeist Australia ( David Zwolski ), Benjamin Ellis

Hi all,

I apologise in advance for the longish email, but I've tried to make it structured and readable.

Wiki
Firstly, if you want more people using the wiki I suggest having a page on how to use the wiki, especially if there is a a quick video on how the dev team is using the wiki and how to actually use a wiki. I say this because my experiences with using a wiki were that it was a pain. You had to learn a language that wasn't quite markdown nor BBcode, it had it's own syntax.

Governance Structures
Regarding the original point of this thread. I can't directly point you to any well tested decision making models although I've listed some below. I can however suggest some great, well tested governance models. The three main ones being Hierarchical, Constellation Model, or (Starfish) Decentralized.

I've got a post about them here : http://zeitgeist-info.com/article/2011/04/governance-structures-decentralised-or-constellation-model/


Decision Making

There's certainly a few ways you can have a decision making system. Here's some I've seen or have just come up with...

Polling

Simply asking a lot of people a question rarely ends up with them giving an informed decision, it usually used more for opinions.
When looking at the results you have to wonder about who was answering the question. Was it only answered by a certain sub-group, like only the core ZM members, only the religiously minded, only those with the spare time, or maybe not many people were active during the time it was up.
Are they fully informed about the problem and ramifications of the proposed solutions?
How does the answer affect them and what are their motivations? e.g They could be making money out of the current problem.
What about the irrational human behaviour aspect? Maybe you wrote the poll in a certain way or it was placed next to something which affected it (a poll asking if we should abolish current meat production being shown below a McDonalds advert would likely get a different response than one below an advert for the doco Earthlings).

Given all the issues, polls seem far better at getting opinion than arriving at decisions.

Weighted polling
Like normal polling but some people's votes count more. E.g someone who's been an active ZM member for 3 years and has a better understanding of the tenants may get 3 votes, and a master in their field (like Linus Torvolds), might be worth 10 votes.

This still has a lot of issues, as a poll on a website or even in an email probably isn't a good way to make a big, long term decision which dramatically affects the course of the group or even all of humanity. e.g selecting which of the many transition plans to start implementing.

Multi-Voting (Votebox)
Different from weighted polling because everyone gets for example 10 votes a year to spend on the items they think are important. This is used for example when many possible projects are listed and allows the next one to start to rise to the top. E.g I could put 9 votes into one project and 1 vote into another or 1 vote into 10 projects. Dropbox's Votebox is a good example of this (although you need to have a login to access it).

This isn't useful so much for selecting if we should be using SVN or GIT, but what features or projects to work on next. Should we be worrying about running our own OAuth2 and OpenID server so we can start getting SSO (Single Sign-On) working amongst all the chapters (which would require an OAuth/OpenID plugin for Joomla, Drupal, Wordpress, Redmine and whichever other platforms and software people are using).
Or should we be focusing on helping develop the Zeitgeist Resources site to expand the database of peoples skills, knowledge and resources?


Committee Default
People who are thought leaders or masters in their field are likely to become aware of the symptoms, identify the problem(s) and come up with and analyse the various solutions. If there is a solution that is far more suitable than the others, but the effects are wider ranging than just something they themselves have to worry about (e.g changing platforms, programming languages, changing, adding or removing user facing functions) then they present the information to the rest of the group. Unless an even more suitable alternative is recommended then they proceed to implement the most effective solution. That is to say that they make a recommendation which, after an appropriate public review period, gets implemented by default. This would hopefully prevent some of the stagnation that I have seen amongst some ZM chapters and teams, especially when it comes to the administration side of running a chapter when most other people care about the activism or education side.

Note that when presenting the options to the group you can't just say that you are going to implement option c. You have to explain why that is the most effective solution, if there are other better solutions you would have chosen if you had more resources, and option c compares and is more suitable than option a or b.
This process ensures that you've checked the options and provided the right context for others to make suggestions.
For example you might try both SVN and GIT and find that Git's Windows clients suck, however it might turn out that a shared Dropbox folder is a great alternative. If however there are almost no developers who use Windows then maybe Git would be better. Or there could be certain project requirements regarding us running our own secure repository which means GitHub, Google Code, Dropbox and the like can't be used (but we can still run our own GIT or SVN servers).

Overall this approach is more of a discussion based method and requires the people who actually care to be both knowledgeable but also flexible. It's unfortunately still likely to be affected by personal opinion (e.g one very loud person pushing for CVS might sway people more towards something like SVN than Git, or maybe even Mercurial).


OTHER

You might want to check out the Citizens Handbook as they have a lot of valuable information.

Of course you can also talk in terms of democracy (direct voting) and representative democracy (voting on someone to decide for you), representative democracy (vote by actually doing), dictatorship (you don't vote, one person makes all the decisions) and various other forms of governance.

There's bound to be other forms of decision making. A dictatorship might actually be effective for a (likely short) period of time if the dictator can come up with better decisions and can inspire their implementation better than a committee.
However we know that absolute power corrupts absolutely and most governance structures and decision making systems are designed to prevent any single or small group of corrupt people from unduly affecting the system. It is also meant to smooth out the issues in our irrational behaviour and allow multiple perspectives and forms of information to converge on the one issue, although in doing so we have to ensure we understand and counter-act any of the predictably irrational behaviour.


Predictably Irrational Behaviour

For those who haven't seen the TED talks by behvaiour economics professor Dan Ariely's or read his books, Predictably Irrational behaviour is when people don't just make wrong decisions, under certain circumstances they will consistently make them.

As a side note, this is one of the issues with economics. The 'Invisible hand of the market place' is in part meant to stem from humans apparently infinite capacity for reasoning (as the view of the day was that we were created in God's image, which if you look at the environment and how stuffed up it is, says more to do with God than humans). If humans were just irrational, then the chaotic nature of it would cancel itself out. Someone purchased too much of something somewhere, someone else sold stuff at the wrong price, etc.. However because we are predictably irrational this doesn't happen. E.g 80% of people facing a certain decision will make the wrong choice, causing a far from optimal outcome.

Here is an example. Most people think that if you increase the incentive (money) then you increase the performance in a linear way. Hence why so many American CEOs get paid vast sums of money. But money is both a motivator and stressor, so for mental (rather than physical tasks), if you get paid a lot more then whilst your desire to do well may increase, your actual performance will decrease. [Paraphrased from http://danariely.com/2011/06/05/upside-of-irrationality-chapter-1/]


Being the Development team I would assume that our main question would be 'How could technology fix or prevent this?'

Sometimes the answer to that question will be that it can't and that it requires humans to change, but usually technology can at least help with such a change.

--
Michael Kubler

Skype : kublermdk
Email : mic...@zeitgeistsa.com
Mobile : +61 433 483 008

Challenge your perception of the present to design a better future for all humanity, support the Zeitgeist Movement

Want to keep up to date with South Australian chapter events? You can subscribe to the South Australia ZM discussion mailing list. Also check out the Zeitgeist Australia site and the Facebook group.

Want to understand advanced Zeitgeist concepts? Check out zeitgeist-info.com

Want to know the latest in science news and information? Check out zeitnews.org

If anyone has any RBE transition proposals please contact the newly forming Transition Ideas Team. Email trans...@zeitgeist-info.com

"It is better to have understandings than beliefs"


Nick Polimeni

unread,
Jun 14, 2011, 10:01:15 PM6/14/11
to zm-global...@googlegroups.com, Zeitgeist Australia ( David Zwolski ), Benjamin Ellis
Mediawiki was picked because it was already installed, and it has great flexibility, and most developers take onto it like ducks to water... you're right, I should write up a brief get started sheet... 

Meanwhile to participate it's easy; just click on the topic you like, and click on the "Discuss here" link for various topics. If you don't find your topic, write to me, and if I haven't yet written up the starting  page... I'll guide you.

If you don't get your technical suggestions and recommendation in the wiki, there's no guarantee that it will ever be seen by the coordinators and developers working on the Phoenix Project.

The other topics you raise, are marginally connected, when it comes to deciding what software or platform to use. But as you say, this depends a lot of well experienced folks, and the rest of us learning as we move forward. Most software problems discovered after the fact can be corrected. It's an ever improving system.

I don't know where the part about Polls came in. They are essentially a way to find emotional buttons and then exploit them in advertising.  So I don't see their use here.

Dr Zin

unread,
Dec 5, 2011, 6:09:16 PM12/5/11
to zm-global...@googlegroups.com, Zeitgeist Australia ( David Zwolski ), Benjamin Ellis
http://en.wikipedia.org/wiki/Consensus_decision-making

Consensus decision-making is a group decision making process that seeks the consent, not necessarily the agreement, of participants and the resolution of objections. Consensus is defined by Merriam-Webster as, first, general agreement, and second, group solidarity of belief or sentiment. It has its origin in a Latinword meaning literally feel together.[1] It is used to describe both the decision and the process of reaching a decision. Consensus decision-making is thus concerned with the process of reaching a consensus decision, and the social and political effects of using this process.

Consensus should not be confused with unanimity[2][3] or solidarity.

Reply all
Reply to author
Forward
0 new messages