KDE is Chaos?

Thanks to Nate Graham for his follow up post to his ‘KDE is Anarchy’ blog.
I think this was probably needed and it was good to see some detail on the various types of organisation within the KDE cadre.
What exactly is anarchy?
The term ‘Anarchy’ scares some people who conflate it with a more modern misconception of ‘chaos’. It is not.
Anarchy itself simply means ‘no rulers’. It does not mean no rules. Nor does it deny leadership, organisation and structure.
But wait, Nate wrote there is the “benevolent dictator for life” model! It is not anarchy then, you may say! Well, not really. This is the world of Open Source. If you do not like the direction of a project you can always reject it or fork it. A real/political dictator would deny you the right to disassociate let alone fork or compete.
Being a “dictator” and controlling your own project is completely different to controlling others without their consent.
This is the key factor. In anarchist systems all things are voluntary as there is no ruler (or rulers) to command and control otherwise.
It also means there is no restriction on how people can organise.
Some may claim that anarchy also means there can be no hierarchy, but not even this is true. Any structure that does not involve involuntary control is compatible with anarchy. It could be spontaneous order, ‘the invisible hand’, a collectivist structure, or hierarchy - as long as it is by voluntary association. Only rulers can deny people on how they choose to interact.
Anarchy as Jeffrey Tucker explains is actually all around us!

“Anarchy is all around us. Without it, our world would fall apart. All progress is due to it. All order extends from it. All blessed things that rise above the state of nature are owned to it. The human race thrives only because of the lack of control, not because of it. I’m saying that we need ever more absence of control to make the world a more beautiful place. It is a paradox that we must forever explain.”

Inside KDE: leadership and long-term planning

how anarchy sometimes works

Anarchy is not a good way to run a development project because it means there is no collective goal and no vision. KDE’s biggest issue is that they do things at random on a whim and because every voice has the same power as everyone else it takes years to accomplish things just because some topics have more people who are passionate about something even if they completely miss the point of the discussion.

KDE has had Single Click by default for 20 years. It has been a problem since day one but every time it comes up there are people who shut it down because they dont like double regardless of the fact that it is expected by 99% of everyone. They only argue about superiority of the topic rather than the reason why it should be changed being “burden to beginners”. So for years and years people have attempted to get KDE to change it to the universal standard of 99% of computing but it doesnt happen because KDE is ran through anarchy meaning that if enough jump in the way progress even seemingly obvious progress is stalled.

Anarchy has some places where it can work but Anarchy does not scale and when a project reaches a large enough size Anarchy becomes a hindrance to progress.

There are many models not just Anarchy vs BDFL that can be deployed by an Open Source project and there are many ways it can be done to keep the philosophy in tact and ethics intact without jumping straight to Anarchy.

You say “no restriction on how people can organize” this means there won’t be organization. This is proven by KDE because they have a Bugzilla, GitHub, GitLab, Phabricator and a Forum for the same similar purpose of organizing.

Anarchy can be a great model in certain aspects but it can also easily get in the way. In KDE’s case, it is in the way more so than it isnt.

2 Likes

Hi Michael,

e: there is no collective goal and no vision
hmm, I am curious on why you say this. There is the KDE Vision and Manifesto, but the whole point of Nate’s recent blog was to explain that there is governance and direction. Current goals are Consistency, Wayland, and Apps.
I linked the article below so perhaps you missed it? If not how do you dismiss these factors??

re: every voice has the same power as everyone - there are aspects of this: “We democratize long-term planning by allowing the community itself to vote every two years” says Nate, yet he also points out that the “KDE Gardening Team, which is essentially our version of upper and middle management.”
Not to mention the ““benevolent dictator for life” model” in some projects.

re: it takes years to accomplish things.
Freedom can be messy but it is certainly preferable to authoritarianism. There is also a lot of arguments that show overall there is much more efficiency in decentralised systems. People can miss the point (or disagree) regardless of the organisational structure! It is not a case of every idiot having a say in areas of that they really no nothing about with the result forced upon others- which is more a description of state democracy.

“KDE has had Single Click by default”. Yes, you have said this many, many times. I agree it is a bad default, but why are you bringing this up? KDE gives you the option to change this as a user. I believe you even convinced Kubuntu to change it! This is a win! File under: Anarchy. If Apple for example hard coded it, then you would have no recourse at all. It may not be optimal but it is at the very least a ‘glass half full’ situation.
What is it that you would prefer? Surely not an autocrat to be given total power to set what they see as ‘sane’ defaults. How would they be chosen? What about a democratic process? How does this affect those who worked hard at the code with their own vision of what their work should be?
What is your solution? Any option other than forking sounds like it would create systemic discontent.

re: Anarchy does not scale
The opposite is actually true! Leonard Read wrote a book called ‘ipencil’ It shows how no one person can create something as seemingly simple as a pencil. It shows it is only through spontaneous order that a complex world can happen. (Efficient & ethical) scaling would be impossible otherwise.
https://youtu.be/IYO3tOqDISE

re: There are many models not just Anarchy
There are indeed other models than BDFL -yes. This is actually one of my points!! ‘Anarchy’ is not a model. It is not prescriptive. It is a negative term. An-archy means no-rulers. Not being a model it can incorporate any structure or model (that does not involve coercion - involuntary rulers)
It should be noted also that anarchy is not a description of a utopia. All it really states is that coercion should be abandoned.

That said, one type of order that is exclusive to a state of non-coercion is 'Spontaneous Order;.
Wiki:

*Spontaneous order, also named self-organization is the spontaneous emergence of order out of seeming chaos. “spontaneous order” is typically used to describe the emergence of various kinds of social orders from a combination of self-interested individuals who are not intentionally trying to create order through planning. The evolution of life on Earth, language, crystal structure, the Internet and a free market economy have all been proposed as examples of systems which evolved through spontaneous order.[1]

Spontaneous orders are to be distinguished from organizations. Spontaneous orders are distinguished by being scale-free networks, while organizations are hierarchical networks. Further, organizations can be and often are a part of spontaneous social orders, but the reverse is not true. Further, while organizations are created and controlled by humans, spontaneous orders are created, controlled, and controllable by no one.[citation needed] In economics and the social sciences, spontaneous order is defined as “the result of human actions, not of human design”*

This is what I believe Nate meant originally when he said KDE is “an anarchic society” That is, KDE is a Spontaneous Order. He also clarifies in the new post that there is leadership and planning.

“organizations can be and often are a part of spontaneous social orders” I think this actually fits the description of KDE. It is a minimal organisation utilising the power of spontaneous order. This is beautiful!
re: there won’t be organization
KDE is 22 years old and has grown. I attribute this to strong decentralisation and the wonderful phenomenon of anarchic Spontaneous Order!
Long may KDE embrace this successful attribute!

Honestly, I don’t see a point in discussing issues with KDE (or any aother topic) through added layer of anarchy/chaos instead of simply focusing on the issus themself.

Looking for root source of those problem is one thing, but I feel like we waste time by trying to fit one thing into unrelated model, just because there are some similarities, while ignoring that other aspects of it don’t fit. This leads to false comparisons.

2 Likes

Hi Kikuchiyo,

Thanks for your input but I find your statement here full of contradictions.
If you honestly ‘don’t see the point in discussing this issue’ …then why exactly are you commenting? “We waste time” - you are merely stating YOUR own time preference as something else.
By stating “some similarities, while ignoring that other aspects of it don’t fit.” you are not actually giving any argument. You are not providing me with any information that would persuade me to change my time preference.

You say ‘added layer’. I reject this preposition -It is not an ‘added’ thing. It is fundamental. You then flip and mention ‘root source’. This is closer to the mark!
My argument really asks the question; what ‘root source’ within the structure of KDE is there that has made it so successful, to endure, prosper and even expand for 22 years?
It is not an issue of ‘looking for’ a problem. It is not looking for the ‘one thing’ but attempting to identify the fundamental thing - of success.

This is comparable to the intention of Eric S. Raymond with his game changing book ‘The Cathedral and the Bazaar: Musings on Linux and Open Source by an Accidental Revolutionary’ It examines the struggle between top-down and bottom-up design.
As such it too was an investigation of structure within the FLOSS domain - and its benefits.
This seminal work rather than being ‘a waste of time’ influenced many and is arguably what transformed Libre software into a visible viable alternative to proprietary offerings.
Without it the beloved by many Firefox browser would not exist.
nb. Eric Raymond also later progressed to the principle of anarchy.

All this is to say the anarchic structure of sponataneous order is a key aspect to why KDE is a successful ‘organisation’. This is a valuable lesson that should be heeded if people wish Libre Open Source principles to thrive in practice.

I don’t know KDE’s structure well but if we’re strictly addressing it’s anarcho elements… the power is in hands of the developer to the degree other developers are unwilling, unable or unwelcomed by the group to take over their development.

Below i’m using the terms “minority” and “majority” to mean the collective value of individual(s) to the project as under anarchy value > personhood.

This has tremendous advantages in evading the vulnerabilities of dictatorial (minority) rule.

  • It’s easier for negative outside forces to compel a minority vs a majority.
  • The minority tends to lack insight into the individual experience of completing and maintaining each devs respective contributions.
  • There’s an exploitable gap between “good management” and “enough people willing to leave/fork” that’s very easy to exploit in a complicated project like KDE.
  • The project will be vulnerable to petty internal and external disputes by the minority which the majority may find hard to quash.
  • Consumer relationships are often placed above Inter-personal relationships.
  • Individuals who have no clue about things like kernel hooks end up with power over those who do project-wide.
  • The collective goal is more often placed above the needs of the individuals.

This has tremendous disadvantages in evading the vulnerabilities of anarcho rule.

  • It’s harder for positive outside forces to compel a majority vs a minority.
  • The majority tends to lack insight into the group experience of completing and maintaining a whole project.
  • There’s an exploitable gap between “good work” and “someone willing/able to take your job because you suck” that’s very easy to exploit in a complicated project like KDE.
  • The project will be vulnerable to petty internal and external disputes between individuals which may be easier to quash but will be more frequent.
  • Inter-personal relationships are often placed above consumer relationships.
  • Individuals who have no clue about things like UX end up with power over those decisions if they fall inside their area of the codebase.
  • The needs of the individuals are more often placed above the collective goal.

That isn’t to say the answer is in the middle, just that there’s a balancing act depending on objectives which is heavily informed by size. If my ability to point to things is any example, it’s that as things grow they need to lean more to the former than the later as much as the reverse is true.

If you honestly ‘don’t see the point in discussing this issue’ …then why exactly are you commenting?

You would have a point if commenting in this thread would be the same as “discussing an issue through added layer”. Clearly, it’s not, as making statements on the point of “adding layers” doesn’t require me to “add those layers” myself.

“We waste time” - you are merely stating YOUR own time preference as something else.

Yes. Kind of. It is my personal opinion that tacking discussion from this angle is counter productive. It doesn’t have much to do with “time preference”, but rather effectiveness of one approach to the discussion over the others.

You say ‘added layer’. I reject this preposition -It is not an ‘added’ thing. It is fundamental. You then flip and mention ‘root source’. This is closer to the mark!
My argument really asks the question; what ‘root source’ within the structure of KDE is there that has made it so successful, to endure, prosper and even expand for 22 years?
It is not an issue of ‘looking for’ a problem. It is not looking for the ‘one thing’ but attempting to identify the fundamental thing - of success.

Where I don’t agree with that approach is that I don’t believe chaos/anarchy truly represent this foundation. Only some of its aspects. While this itself isn’t bad. Comparisons with ideas that are “close enough” can help to understand them better. In this case, I don’t see much point. It would be much more productive to instead see what exactly makes parts of a project work or not. To see when it works and why. Declaring single idea as a core of success can lead to forcing with it in all cases instead of more nuanced approach.