Community protection by Democracy

This thread wants to suggest ideas on how to protect the Rocky Linux project from hostile and subversive take-over.

Democratic Community

  • There should be a Board of Community Managers and Technical Experts who are leading the Rocky Linux project.
  • There should be organized votes, in which Board members are elected by the Community.
  • Election votes should be organized in fixed, periodic time intervals. (e.g. yearly)
  • Board decisions need to be made public within the Community. (e.g. Board decision blog)
  • Community members should have the right to oppose Board decisions with a Referendum vote.
  • Results of Referendum votes can overrule Board decisions.
  • If a Referendum vote confirms the Board decision, it helps the Board by adding additional confidence.

Why Democracy?

  • It should protect the Community from becoming hijacked by hostile/subversive take-over.
  • It should protect the Community from wrong or inactive people being in the Board.
  • As votes are likely to be based on meritocracy, it will guarantee that the best people will be in the Board.
  • Decisions backed by a majority will be more robust and “here to stay” than decisions made by a minority.

Why not Democracy?

  • A democratic process is indeed more lame and time consuming than a quick shot made by one person.
  • Requires additional structures for safe voting.

Is “Democratic Community” a good idea to protect the Community?
Are there better ideas?
Please share your thoughts!

Toni Feric


It is a double edged sword so there should be someone as s project manager with protocols in place that can make some autonomous decisions. That said yes independent input with the right team would be a good thing. There may be a place for special interest groups that control certain things within a given set of protocols. I specifically work with VoIP as a former staff developer for FreePBX working heavily with asterisk. There is a large community behind open source communications.

1 Like

This is a good start. As @James notes above, there ultimately always ends up being someone with ultimate authority anyway. In not-for-profits, that ends up being the board (whatever we may call them). So it’s very important to safeguard the board from undue influence, and a lot of that comes down to culture and who we elect. Going a little bit into my comment in another thread - I am intrigued by the idea of a referendum process that could override a board decision. I’ve never seen that done in a legal entity before, because the board of a legal organization is the ultimate decision making authority. That isn’t to say that can’t work; I’m just intrigued by it and would love to see how that might work. However, democracies are also subject to subversion, so that is definitely a double edged sword as @James says.

Some questions that would immediately stand out to me is – who is qualified to vote in a referendum? Who has standing to demand/request one? How would we guard against a situation where a referendum is stacked against the best interests of the community?

1 Like

The best protection against subversion is short and hard rules, potentially with consequences. Democracy is not immune to hostility (for a modern example, see bitcoin).

So simple and immutable statements like

  • Rocky(version) will always be a build from source of RHEL(version). Any proposal that runs counter to this statement will be rejected, and the proposer will lose any and all administrative power, direct, or by proxy (voting), within the Rocky [project,foundation,etc].
  • Any and all who propose a Code of Conduct or other means by which active and contributing members of the community might be censured or ejected for reasons outside of their contributions, shall be banned from participation in the community.
  • Official proposals to alter or remove these rules will result in the permanent ejection of and ban from contribution by the proposing community or board member(s).

This type of thing would have to be at the top of a very short list of organizational bylaws to prevent someone from changing the purpose of the Rocky project.


There are a lot of good ideas here, as well as valid concerns that the community is beginning to unearth.

My interpretation of much of this discussion is that the principal concern is the potential subversion of Rocky to some kind of group/company/organization that does not have the best interest of the community at heart in one form or another. This is a valid concern, and it is one that I share as well.

At this early time, it’s important to maintain perspective. We haven’t even existed for half a week! Our immediate priorities are:

  1. Preparing the website, GitHub, and other associated resources with appropriate wording, copy, and branding. I’m thrilled at how well this is progressing!

  2. Laying the necessary infrastructure to support the growth and development of the community. We’re making progress here, too.

  3. Organization of ourselves to begin to prepare for the delivery of the Rocky OS.

The third point is critical, and we need to ensure we don’t lose sight of what brought us all here. Ultimately, we need to deliver a solid successor to CentOS. There are a lot of companies and organizations that are justifiably concerned over the LTS of their deployments, and we really don’t have a lot of time to deliver on that.

The Rocky community has come together rapidly, and it’s tremendously exciting to see! But we’re at a critical stage right now. We’re like a star that has the potential to go super nova, but if we don’t do things right, if we get in the way of ourselves, we run the risk of collapsing before we even had the chance to really get off the ground.

All this to say that while these kinds of discussions about how Rocky should best be governed and directed are important and necessary, for now they remain more of a longer term goal until we can ensure that many of the most fundamental items have been addressed, and momentum is not lost at this critical juncture of our early development.

Here’s what Greg had to say on the subject in Slack recently:

This is a great thread, and apologies for my latency. At the moment, and until the landscape is better understood, I will maintain the lead of Rocky Linux. This may change in the future, but for now I believe this is best for the project and the community.

My commitment is to the project and keeping the project open and free from commercial interest (even from my own company). I’m not saying that organizations shouldn’t monetize on Rocky Linux, quite the opposite, but the decisions for the project should always be in the best interest of the project and community of users. To the best of my ability, I will always ensure that happens or I will step down.

I agree with this, and it appears that many of you do as well. I want everyone to be clear that while we may focus on ensuring we don’t collapse in on ourselves at this early stage, that doesn’t mean your voices aren’t being heard, or that your concerns are being dismissed. They’re not. I just think the wise approach for now is to ensure we have some basics in place (and I mean really basic—I haven’t even had an email account for more than a couple of days!).

The last thing I’ll say on it is that I hope no one considers my thoughts authoritative. The above is how Greg sees things, how I see things, and how it looks like many of you see things as well, based on conversations I’ve seen in Slack. I always want to do my best to make sure that, where possible, I’m listening to everyone and helping to enable the community and our project, and I’m submissive to the will of the community. I’m grateful to all of you who will help keep me honest on that front!


Fully agree. A project manager would be like a CEO, who receives strategic decisions from a Board and executes them by creating smaller task to achieve the strategic goal.

In so far, Democracy (or voting on decisions) does not need to be applied on every tasklet, but only on major decisions.
In other words: use Democracy where it makes sense, and don’t use it, where it does not make sense.

The tricky part is to define that precisely in written form as some sort of “Community Constitution”.

The idea is borrowed from Swiss Direct Democracy. In Switzerland, there is no instance of power above “the people” (=citizens with a right to vote). The people can change the constitution and stop parliamentary decisions.

In the context of Referendum, it should be understood as some form of veto, that is accept/reject on parliamentary decisions. If a parliamentary decision is rejected by the people, the parliament has to make a new proposal which will eventually be accepted. Notice, not every parliamentary decision needs to pass a Referendum vote. Only those who are controversial.

It would be interesting to see if Direct Democracy, which works in Switzerland really well, would also work in OSS Communities.

I think it is deliberating to admit that there is no perfect solution that will guarantee to prevent failure. If Democracy can enhance resilience - that is reduce the probability of failure - it should be used.

These are all excelent questions.
And in the context of FOSS Communities, we appear to be in Terra Nova.

Maybe we need to figure out the implementation details ourselves. Maybe we could try to borrow ideas from the Swiss implementation and convert it to the software world?
Admittedly, I don’t know. Yet.

If we think Democracy, “short hard rules” would be somewhat the adequate of a Constitution, wouldn’t it?
Like a “Community Constitution”?

That’s correct. But Democrycy is supposed to provide more resilience against hostility. That is it should reduce the probability of failure.

I don’t think that just a list of hard rules/consequences is sufficient.
Who is going to enforce this kind of law, when the rule breaking happens at the top of the hierarchy? For that reason, the highest instance of power should be the Community (i.e. the sum of all members) and not the Board.

Hi Leigh
I think it is safe to assume that nobody in this thread expects these changes to be implemented anytime soon. At least I fully agree on the priorities you outlined.

As far as I understand, there is no such thing yet as Direct Democracy in OSS Communities, so we seem to be exploring Terra Nova. That exploraration may be a time-consuming process, and if it takes months, then why not start now, and keep discussing it?

Is your main concern, that people like Greg would love to participate in this discussion, but don’t have the time to do so?

While I understand your points, they are quite dismissive of the core problem with democracy, in that, democracy works best with homogeneous communities who are accountable to each other.

An OSS online community is accountable to nobody, and members who vote to destroy the community-driven product (like Rocky Linux) may leave the community anonymously after they ruin it, with no consequence.

I am hoping that Greg does maintain point and again sets up some very strict and immutable (according to mission statement, constitution, or other foundational document).

I don’t think democracy is even more resilient to hostility, since it is more likely that there are infinite hostile actors compared to a small or tight knit and purpose-driven community. Add this to the fact that most people are idiots and easily swayed by rhetorical arguments, and you’re in for a pretty bumpy ride if anybody who speaks or writes well ever has it out for your organization and can convince those ‘democratically’ voting on issues to force the community to vote against itself.

For now, given that the founder of the organization, community, and product is on board with staying the course, that is enough, but I still think that consequences for attempting to cause the organization to abandon its original purpose (as CentOS just did) should be met with exclusion from the community, whether it means simply proposing to do something counter to the purpose (building LTS RH clones from the open source), or proposing to eject contributing members of the community for something unrelated to their contributions.

In another organization that I was a board member of, we had an annual election and one of the questions on it was a place where regular members could affirm the actions of the board in specific areas. This seemed strange when I originally asked about it, but as the board changed direction (and I eventually left the organization on moral grounds, not entirely unlike the CentOS story), it now makes sense to me.

I like what you are all discussing above. Along with the pro / con of democracy, I will also mention explicitly that moderation is key. Knee jerk reactions in either direction lead to ignorance of consequences. I like that you are starting to think it through this early in the program, rather than immediately jumping equal and opposite reaction to what united us all here.


[…]we had an annual election and one of the questions on it was a place where the members could affirm the actions of the board in specific areas[…]

Ahh! Capital ‘M’ members? It took me awhile and a few re-readings to catch that. Depending on how the foundation is being setup (that process is underway now) we could certainly have Members who act as a counter-balance to the Board. For those unacquainted with this idea: it is sort of like the not for profit version of shareholders. Members can be given certain voting powers regarding organizational decisions.

Knee jerk reactions in either direction lead to ignorance of consequences.

This is a good point. At the end of the day I don’t see a way forward for a project that runs exclusively on votes cast by “any” interested party. A structure where we have many with voting power who aren’t exclusively on the board could be the best of both worlds and would certainly be a stronger protection against what the CentOS board allowed while still having decision-makers who can cut through never-ending discussion.

It was the National Capital FreeNet board, and in order to use the services you would have to become a “Member”. “Member” had voting rights in the annual election, which included 3-year rotation of seats on the board (“Board Members”) and a set of important questions. Some of the important questions were “housecleaning” like the one that affirms the actions of the board for the last year. For example, one was about remuneration of staff, since not-for-profit has to care a great deal about where money goes. Some were more like referendum questions about what should happen next.

It was a great cause in the '90s and early '00s. Free Internet access (dial-up modem since 1991 or 1992, which was before most people even knew what the Internet was!). But, once broadband Internet came about, and FreeNet became basically a reseller of DSL services (something that ironically, I helped broker :slight_smile: ), I was concerned that we weren’t helping the community anymore, but actually holding it back. Every project has its time in the sun. Some just don’t know when to sunset.


Thank you for proposing to talk about this particular subject, divisions about responsibility and distribution of decision power and types of government are indeed important and better discussed soon.

While it is true that early decisions on this subject may seem superfluous, I believe they are important.

Currently, I would assume that we are somewhat of a technocracy, decisions are/should be taken as a result of a scientific debate, leadership is/should be elected by their degrees of technical capabilities. I would say that ATM it may not be a bad thing, especially considering the fast pace that we all want to maintain. However, I think it would be a very good idea to have everyone voice their opinions for a particular matter (or against).

Also, I believe what would be more important than any of that is to set the philosophy and imutable objectives of the initiative. Because more than having clear leadership, understanding clearly where we are heading and the culture that we would like to spread is mandatory for the success of collaboration.

About the “overthrow” issue, it is a genuine concern ofc, althouth there exists no gov system that can avoid that, I believe a corporate leadership is not a good idea (board of directors). Indirect democracy, e.g. active members, seems like a good idea, even in that case though, so believe the voice of the users should be heard and matter in certain decisions, I don’t think this would be an efficient protection against overthrows however.

Technically speaking, does anybody know an open-source platform for Direct Democracy?
It has been a while since I studied the matter, like making sure that everybody eligible can vote, the vote is secret, and as well can verify that the vote got effectively counted.

What we should avoid is a ‘no structure’. Especially when the project is picking up steam and size as Rocky Linux is doing atm. For smaller projects it can even be a loose structureless community. Some people will stand up and do what they are good at. As soon the project grows, lack of structure will bite in the back. A long time ago Jo Freeman wrote a very interesting piece on this. She was active in the Female Rights movement back in the late 60’s, early 70’s. It is quite a long read, but to me it makes sense: