Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
The Tyranny of Structurelessness (jofreeman.com)
36 points by orph on May 18, 2014 | hide | past | favorite | 15 comments


Pretty similar to this is Robert Michels' Iron Law of Oligarchy [1], which I understand to be a reaction to the trend of Marx-inspired leftist groups supposedly devoted to egalitarianism developing rigid hierarchies and power dynamics.

[1] https://en.wikipedia.org/wiki/Iron_law_of_oligarchy



Thanks. Burying as dupe.


At least at small scale, lack of formal structure may work very well. And often better way is no to use democracy, but do-ocracy. Usually some people have higher of lower position, but it is a feature, not - bug. Formalizing power, while gives information to outsiders, also (or: primarily) entrenches it.

See: http://www.communitywiki.org/cw/DoOcracy, https://www.noisebridge.net/wiki/Do-ocracy


Very good article, and pretty true. It also shows that you don't need to invent global conspiracies of world domination: the rich and the powerful naturally mingle, and will often do what is in each other's interest, not necessarily out of malice. Of course, the end result will not be necessarily different for those negatively affected by the policies created in this context.


Needs a [1970] (or perhaps a [1973]) in the title.

This is an incredibly valuable essay on just how hard it is to eliminate hierarchical power relationships within groups of people, and how it pops back up in hidden (and thus unaccountable) ways. I've often wondered whether similar issues occur in flatter open-allocation style organizations.


"I've often wondered whether similar issues occur in flatter open-allocation style organizations."

You can stop wondering: Yes.

Hierarchies emerge in almost any non-trivial human endeavor. It's not hard to understand why, a simple model can demonstrate it. Imagine a group of 50 people in a club or something. Imagine each of the 50 people have a probability distribution of how likely they are to prefer to listen to somebody uniformly randomly initialized. Every time a person listens to somebody, and the outcome is good, they're slightly more likely to listen in the future, imagine that if it goes badly, they're slightly less. Iterate on the outcome of decisions; in short order, a hierarchy will emerge, in only slightly more time, it will become very strong. (Probably even if the outcome of decisions is random, but certainly if the outcome is not, and in practice while there is always a random component there is certainly a difference in people's abilities to make quality decisions. Oh, and BTW, the exact definition of "quality" is use hardly matters... "decisions with outcomes that superficially look good to the rest of the club but is actually to their detriment while making me look good" is a popular pathological variant, for instance. Think math here... regardless of the decision quality metric chosen, that metric will produce some hierarchy.)

Claiming you're going to eliminate informal hierarchies is essentially equivalent to claiming you're going to eliminate communication, or the eliminate consideration of the consequences of decisions made... in either case you're creating an organization that is pathologically disconnected from reality from the get-go.

You're far better off harnessing this process, guiding it, and understanding where it can go wrong and how it can go wrong, than trying to simply deny it. The solution to weeds is tending the garden, not declaring this whole plant idea fundamentally flawed and salting the earth.


I learned about the essay, and read it, because of one of the previous times it came up here at HN. (A search find 7 other matches; the one with the most comments is https://news.ycombinator.com/item?id=7555013 .)

It's definitely influence my way of thinking about "open-allocation style organizations", if I understand your meaning correctly. Eg, one group I'm loosely associated with strongly asserts that it's unstructured, and declares that anyone can do anything. As a result, it doesn't seem to actually do anything except act as a boasting point for one of the original founders.

While another, which I used to be closely associated with, has an organizational structure, yearly meetings, GSOC participation, etc.


This is what I meant: http://en.m.wikipedia.org/wiki/Open_allocation

BTW, are you Andrew Dalke?


Thanks for the link. And yes, I am.


Which one did you prefer or like better?


The one with structure and specific goals, public organization meetings, elections to specific positions, and a hand-off of leadership over time.


"[do] similar issues occur in flatter open-allocation style organizations?"

YES, they do. The hive-mind of a group (even a flat leaderless one) many times leads the old members to reject new incoming members. The reasons being: To protect the group and its traditions and culture, and to prevent the group from undergoing unforseen negative changes due to an influx of new members, and to conserve hard-earned resources and rewards for themselves.

I've seen this at companies, online communities, and immigration politics. Here's the thing, it's not necessarily a bad quality. There are many communities (and even countries) that failed to preserve themselves and lost the very thing that made them great. With absolutely no new members the group's ideas become stagnant. With too many new members the group is overwhelmed, distrust grows high, and the original traditions and culture are destroyed. So, all things in moderation I guess.


> " I've often wondered whether similar issues occur in flatter open-allocation style organizations."

What makes you think they might not occur? The article is premised on the idea that humans in groups always form structures and then examines what happens when there is only an informal structure. Open-allocation seems like it would lead to an environment with only informal structures.


Well, because open allocation isn't a lack of all structure, just a lack of a multi-level hierarchical management structure.

It is different enough that the question seemed worth asking.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: