The Cathedral and the Bazaar

[First draft]

In 1996 Eric S Raymund held a presentation at the Linux Congress called “The Cathedral and the Bazaar”. The main premise of that talk was the comparison between software development from a strictly controlled and monolithic approach (the Cathedral) or a loosely organized get-together in which people can walk in, contribute (or set up a shop) and walk out again as they please (the Bazaar).

What this is about

In 2014 I will be working on something new and scary (for me personally), that will include the input and possible collaboration of many different parties.

The first milestone: 30 000 people

The first milestone will be a reader-audience of 30 000 people in the Netherlands. 


I think the most effective approach in this is a collaborative model. Most important aspects are:

  1. Respect of individuality — Each person and group active now has their own reasons, motivations and goals to do what they are doing. The first factor leading to failure is to try and force each of these people into the same direction and force them to work from only one set of values.
  2. Respect for the past — Whatever has happened before has lead to the current state of things, which are not that bad at all. We have 5 publishing houses in the Netherlands publishing good looking books within the genres of SF, Fantasy and Horror. We have four annual short story contests. There are over 250 active Sf, Fantasy and Horror writers.
  3. Finding the common ground — Instead of “creating” one, it is much more effective to find out what drives each of the parties and (instead of pushing things or asking for even more investments than already done) find out which activities touch a common ground and can be boosted to new heights without creating more strain.
  4. Keeping it simple — Complex and/or ambitious projects are usually easier prone to failure than simple ones. If you need to go from A to B, taking the train and the bus might be an easier solution than building your own car.

Possible directions to go

In writing and getting published, but also in creating an environment in which writing can flourish and reach an audience, one can go several directions. I mention four:

  1. Closed/for sale —  Use a (closed) business model based on the sales of individual works to an audience that is as large as possible.
  2. Open/for free — Get the work out there (for free), available for as many people as possible, with the least amount of barriers possible.
  3. Centralized management — Direct all actions from one centralized place
  4. Distributed contribution — Distribute as many actions as possible to as many people as possible willing to help and contribute


  1. Reaching a bigger audience — Right now, most writers (and illustrators) roughly reach 100 to 500 people, unless they are lucky enough to be selected and published by a bigger publishing house.
  2. A steady production (of good quality) — I believe it is important to have a steady stream of work coming from a diverse group of writers and illustrators. Once you reach an audience, it is nice to keep that audience. The work these writers (and artists) might produce? Work related to (but not limited to): Hard SF, Politically engaged SF, High Fantasy, Urban Fantasy and Young Adult.
  3. A rise in quality — There is a certain bias of Dutch readers against Dutch writers: “Dutch writing is (per definition) inferior”. This is mainly related to American writers (and illustrators). The only way to really break through that prejudice is to get to a point where we produce a steady stream of attractive work of high quality. Not easy, but very doable in a period of 3 to 5 years.
  4. Competing against several detractors — The audience that could be reached (probably 100,000 to 200,000 people, see this post) is furthermore spoiled by a saturated market of games, movies and award winning stories in which almost everything of their needs can be found. If you can’t at least match the high part of the average standard (the better stories, the better art work) you will remain in a small market of supporters.

Why I believe collaboration (with some competitive elements) is the best answer

Competition has a very limited value. Sure it helps to put things on the edge and motivate people to make that extra step and to strive further than they currently do. Without collaborative models underneath, competition has only one single outcome (once you have beaten all your competition): solitude. Collaborative models (with some competitive elements) strive for a common growth. In the case of the talks I will have in 2014, that common goal is to break through to a larger audience. (from 500 to 30,000 people). You do not reach this by one single winner within a game of elimination. You reach this by a group of awesome people in a game of inclusion and communal growth.

Awesome people

“Awesome” here is relative. “Awesome” can be:

  1. Awesome organizers — People who get things done. Who can get people together and combine individual elements into something bigger
  2. Awesome supporters — People who help in any way possible to get specific tasks done, from making sandwiches to preparing all the elements needed to make things work.
  3. Awesome writers — People who write stories that engage other people. From total adventure-pulp to incredibly complex literary works and everything else that is possible.
  4. Awesome artists — People who can create audiovisual experiences, illustrations, art
  5. Awesome promoters — People who know how to involve other people in new stories and endeavors.
  6. Awesome publishers — People who can package and sell stories.
  7. Awesome feedback — People who read the stories, look at the results of organizers and others involved and are able to tell you what can be better in such ways that you can clearly see the benefits and are happy to do so.

The possible challenges

In each environment, there are at least 1 or 2 challenges to be met. Below I define five.

  1. Underdeveloped market — In many places, including the Netherlands, SF and Fantasy is an underdeveloped market. You write. You create art. You put it online or wherever out there, maybe find a publisher. And then what? Who is able to (help you) polish and improve your work? Who has the reach
  2. Underdeveloped reach — How many people (of your potential audience) are aware of what happens on their own home-ground? How many people really know what is good and what is mediocre? How many people know you (and your peers) actually exist to begin with? How many of those will actually be able to find and read/consume your work?
  3. Closed / isolated / fragmented approach — In most cases, initiatives created are isolated and fragmented (instead of connected to other initiatives) and many of these initiatives follow the classic centralized/closed approach where most responsibilities will be centered on one person. This includes websites, publishing houses, magazines, contests and writing stories / producing art.
  4. Underdeveloped eco-system — Each and all parties present (magazines, publishers, writers, artists, websites) are part of a bigger eco-system. This eco-system has many interdependencies. Publishers like to reach an audience. Magazines usually already reach that audience. Publishers need writers and writers need both magazines and publishers to reach an audience. When there is not enough funding, only a limited audience aware of what is available, the eco-system itself reaches a halting-state at some point: below what it could become.
  5. A polluted / toxic eco-system — There are many factors (including strife, greed, anger, old pain, resentment) hindering (several forms of) collaboration to actually happen. These (usually unintended) factors (plus the frustration of things not happening) can lead to a pollution of the eco-system. If this goes on for too long, that eco-system can become so toxic and/or polluted that it is easier to drop it as a whole and start from scratch.

A look at the open / closed models

Learning from Open Source Software projects

Open Source Software development is one of the most clear and most current models that has shown what you can achieve using an Open model. Open Source Software development initiatives fail more often than they succeed. To avoid too much wordage, let me summarize the success-factors:

  1. A shared problem — There has to be a shared problem that requires a solution
  2. An already existing willingness (and/or need) to invest — People have to be already willing to invest in an Open Source project even before the project itself exists.
  3. Acceptance / embrace of differences — The reason you join are (most likely) different than the reasons I join. The time I invest will be (most likely) different from yours. The quality of my work will be (most likely) different from the work you produce.
  4. The willingness to listen and share — Open Source Development is a shared and communal effort. Everyone is important. Everyone plays a role. If you are incapable of sharing your space (successes, results, your contributions) with others, you should not join an Open Source effort. Listenign to others (and their opinions) is a very important part of this as well.
  5. A clear understanding of what you can contribute — You are not an employee in an Open Source model. There is nobody telling you what you should do or when that should be finished (at the most: requesting you to deliver a certain contribution, but that’s it). Understanding what you can contribute makes your own efforts and work much more centered and focused. In short: you do what you like most and what you are best at.

Centralized or decentralized?

I believe a mix is the most effective approach. You need some centralized activities (or attractors) to bring things together. Where will you be going? What is the purpose? Each part of the total system, however works best when it is independent. Why try do everything from one single place? Apart from the organisatoric aspects (the amount of people you need, the amount of work involved in managing these people, the amount of money needed to reward them) there is the aspect of vulnerability. One single organisation can’t do everything in the best way possible. Nor can it please all it would like to reach. When something goes wrong within that single organisation, it is possible all will feel the consequences. Several independent organisations working loosely together can cater a much wider audience, specialize on specific aspects and specific audiences and deliver a very specific quality for that audience. When one fails, the damage is usually limited to that single entity. As for the organization: as each entity is smaller in size, the overhead is usually smaller as well. Less people are involved. Communication lines are shorter. Decisions can be made faster.

Why things usually don’t work

Here are five recurring reasons I have encountered in several projects in the past:

  1. Similar goals, similar investments, different purpose — While it might seem that people from different groups are doing similar things, their work/effort is not combineable because they have different purposes in mind. For instance: “earning money by selling the end product” versus “using the end result as a stepping stone for something else”
  2. Uncombinable personalities — In some cases, people simply do not fit on personal levels. Even when and if their work is very similar and even if they are willing to work together with someone, their personalities simply clash.
  3. Differences in personal relevance — Where the end result is crucial for one person, it can be a side issue for another. Where on person will give everything possible to reach the end goal, the other might give up the moment things become something resembling an effort.
  4. Old pain — Something has happened in the past. Between people who might be able to work together, or between non-related people in the past. The old pain that still exists hinders any real form of collaboration.
  5. A toxic environment — Sometimes the environment in which things could happen has become so toxic (strife, hatred, disdain, exclusion of certain people and certain groups) that nothing new can really grow or flourish.


I close with several considerations related to open, closed, centralized and decentralized models.

Benefits of a centralized model

The closed and centralized approach allows for maximum control. You decide what will happen when and where and how. Any decisions made first go through a process directed by you, so mistakes (and possible losses to reputation and sales) are minimized.

Disadvantages of a centralized model

The problem with a centralized model is that the reach you have (in the amount of people contributing) are limited to the size of your resources (in most cases: money). Scaling up (more people to create a bigger reach) will introduce more cost and more risk as the money you invest in a product might never be paid back by your sales. While you have more control over the products you deliver and the way they are distributed, you also need a lot of money to get things really going.

Benefits of an open model

An open model usually kicks off by invitation. You invite others like yourself, with similar needs and wants: to contribute in your model or idea. In most cases as you are all doing the same (or similar) thing and joining forces will be much more efficient than all doing the same things in slightly different ways. Instead of doing all yourself, you work with others to help create that thing you feel you all can benefit from. Since all involved would have invested in this thing already on their own, the Return on Investment (when the joint effort is successful) is incredibly more high than what you would have been able to achieve on your own. In short: 2 + 2 can become 50.

Disadvantages of  an open model

One of the biggest “disadvantages” of an open model is the loss of control and the loss of ownership. You are no longer the only one deciding what should happen and what has the highest priority. You no longer “own” the end result. Meaning that anyone involved can use the end result as he/she likes, for any purpose they have in mind and all money that comes out of sales of the end result is not just “your” money but that of everyone involved.

Making money out of open models

If and when (making) money is an issue (for instance, because the end result is an important part of your business), not the end result, but what you can do with that end result is where your possible earnings will be.

Open source

Open Source code, including the source code of Linux, is widely available online. For free. You can download compiled libraries or the raw code itself. You can change the content of the code and make your own versions and variations as you like. You can take out specific bits of that code and create something different with it. Whatever it is, whatever you want, you are free to do so.


If and when the Open Source code is under a specific licence (basically a summary of the intention under which the code is released as Open Source, plus a set of rules you are invited to implement when you use that Open source code, or parts of it).

Closed Source

Closed Source software is software that only comes in complied form. To understand how it works and to see the lines of code you need to decompile that software. The software that does this takes the binary code of the compiled Closed Source, read it step by step, figures out what is related to what in which way and then presents the best it could produce in human-readable programming code.

Commerce / business model

The business model behind Open Source is completely different from that of Closed Source software. Since Open Source software is freely available and open, most business models based on Open Source software are based on a “spin off effect” of that Open Source. For instance:

  1. Building tailor made solutions with that Open Source software
  2. Maintaining environments running on that Open Source software
  3. Offering infrastructure to run that Open Source software

With Closed Source software, the software itself is the product on which part of the business model is based. Microsoft Office is one example of such Closed Source software.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s