Valve has also been called out by former employees for not being actually structureless. They don't have an official org chart, but there's an unofficial one at work anyways.
Any sufficient large group of people will self-organize into hierarchies. A refusal to write it down doesn't prevent this from happening, it just means people are even less accountable for the position they occupy.
"We believe in flatness" usually is code for "we believe in the appearance of flatness".
It's worth pointing out that the author was a Stalinist, and so thought the solution to the problem of hierarchies forming implicitly was to make those hierarchies explicit and enshrine them.
The article describes a real problem; it speaks to my experience and the experiences of my comrades; but it offers no real solution and is essentially a defense of the status quo. It doesn't preclude a better solution.
Agree with your points. I find that posting the essay is still quite valuable, though - as an industry we are still enamored with the mythology of structurelessness.
So yeah, her solutions may be pretty questionable, but IMO as an industry we're not even at the point of exploring solutions, considering that so much of our industry doesn't even acknowledge the problem exist.
Until then it's useful to keep flogging the essay until people accept that yeah, this problem actually exists.
Personally, I think that structurelessness is much more possible in "our industry" because it's so culturally, racially, sexually, and class homogeneous. It's worth noting that this article is about feminist activist groups, and not software companies. Software companies have both much more clearly defined goals and much better metrics of measuring their progress towards those goals than feminist activist groups do. That in itself is a gamebreaking difference.
> "Personally, I think that structurelessness is much more possible in 'our industry' because it's so culturally, racially, sexually, and class homogeneous ..."
> "Until then it's useful to keep flogging the essay until people accept that yeah, this problem actually exists."
I fully agree, but I'm pleased to note that this is the first such 'management is dead' story where multiple commenters have given salient counterpoints (small company, no HR/PR disasters to date, who can fire you, etc), rather than jumping on the bandwaggon.
> Any sufficient large group of people will self-organize into hierarchies. A refusal to write it down doesn't prevent this from happening, it just means people are even less accountable for the position they occupy.
My takeaway for Valve was they wanted self-organizing heirachies, but wanted to avoid people who's ambition was to "manage."
That is, people will naturally take leadership roles for a project, and at the projects conclusion, go back to the fold.
By requiring leaders to lead without authority, it demands a different kind of leader.
> "Ok, so this debate seems to be framed in terms of authoritarian bureaucracy vs. anarchy."
It isn't really. The observations surrounding structurelessness suggest that true anarchy is impossible. In the absence of some type of hierarchy one will emerge anyways. The notion of a group of true equals with no authority over one another is a pipe dream.
Not to mention the notion of defined hierarchies makes no statement on what the hierarchy looks like, nor the system that determines who goes where.
Corporate structures are frequently democratic. The Board of a company is very explicitly such a structure. Not to mention that even without explicit democratic structures managers still answer to those they have authority over due to market forces (nothing says "you suck as a manager" like your team quitting). In this industry in particular, it's very, very hard to build an authoritarian bureaucracy.
I'm not sure where you're getting the authoritarian/anarchy divide from, that wasn't in any way implied by my post, nor do I believe either extreme is possible in our industry.
You're talking about "participatory management". It was a fad in the 50s that comes up again from time to time. It sounds like a good idea, but it actually ends up increasing bureaucracy. There's a good article that discusses exactly what you're talking about here:
That essay is hilarious, it explains all these failures of companies doing participatory management half-assed, provides a few successful counterexamples of companies/divisions/plants that did it full-assed, and then concludes that the failures failed because they didn't take the changes far enough.
The Circles approach they describe in this article feels pretty structured. It includes clear responsibilities and authorities. It's less clear who defines those responsibilities and authorities, but I'll assume it's "collaborative" – and specifically it's supposed to be well documented and internally public, so there is some implicit peer review.
Any sufficient large group of people will self-organize into hierarchies. A refusal to write it down doesn't prevent this from happening, it just means people are even less accountable for the position they occupy.
"We believe in flatness" usually is code for "we believe in the appearance of flatness".