The phrase "open source" (still) matters September 16, 2022 on Drew DeVault's blog

In 1988, “Resin Identification Codes” were introduced by the plastic industry. These look exactly like the recycling symbol ♺, which is not trademarked or regulated, except that a number is enclosed within the triangle. These symbols simply identify what kind of plastic was used. The vast majority of plastic is non-recyclable, but has one of these symbols on it to suggest otherwise. This is a deceptive business practice which exploits the consumer’s understanding of the recycling symbol to trick them into buying more plastic products.

The meaning of the term “open source” is broadly understood to be defined by the Open Source Initiative’s Open Source Definition, the “OSD”. Under this model, open source has enjoyed a tremendous amount of success, such that virtually all software written today incorporates open source components.

The main advantage of open source, to which much of this success can be attributed, is that it is a product of many hands. In addition to the work of its original authors, open source projects generally accept code contributions from anyone who would offer them. They also enjoy numerous indirect benefits, through the large community of Linux distros which package and ship the software, or people who write docs or books or blog posts about it, or the many open source dependencies it is likely built on top of.

Under this model, the success of an open source project is not entirely attributable to its publisher, but to both the publisher and the community which exists around the software. The software does not belong to its publisher, but to its community. I mean this not only in a moral sense, but also in a legal sense: every contributor to an open source project retains their copyright and the project’s ownership is held collectively between its community of contributors.1

The OSD takes this into account when laying out the conditions for commercialization of the software. An argument for exclusive commercialization of software by its publishers can be made when the software is the result of investments from that publisher alone, but this is not so for open source. Because it is the product of its community as a whole, the community enjoys the right to commercialize it, without limitation. This is a fundamental, non-negotiable part of the open source definition.

However, we often see the odd company or organization trying to forward an unorthodox definition of the “open source”. Generally, their argument goes something like this: “open” is just an adjective, and “source” comes from “source code”, so “open source” just means source code you can read, right?

This argument is wrong,2 but it usually conceals the speaker’s real motivations: they want a commercial monopoly over their project.3 Their real reason is “I should be able to make money from open source, but you shouldn’t”. An argument for an unorthodox definition of “open source” from this perspective is a form of motivated reasoning.

Those making this argument have good reason to believe that they will enjoy more business success if they get away with it. The open source brand is incredibly strong — one of the most successful brands in the entire software industry. Leveraging that brand will drive interest to their project, especially if, on the surface, it looks like it fits the bill (generally by being source available).

When you get down to it, this behavior is dishonest and anti-social. It leverages the brand of open source, whose success has been dependent on the OSD and whose brand value is associated with the user’s understanding of open source, but does not provide the same rights. The deception is motivated by selfish reasons: to withhold those rights from the user for their own exclusive use. This is wrong.

You can publish software under any terms that you wish, with or without commercial rights, with or without source code, whatever — it’s your right. However, if it’s not open source, it’s wrong to call it open source. There are better terms — “source available”, “fair code”, etc. If you describe your project appropriately, whatever the license may be, then I wish you nothing but success.


  1. Except when a CLA is involved. A CLA is an explicit promise that the steward of an open source project will pull the rug out later and make the project proprietary. Never sign a CLA. Don’t ask contributors to sign one, either: consider the DCO instead. ↩︎

  2. This footnote used to explain why this argument is incorrect, but after five paragraphs I decided to save it for another time, like when the peanut gallery on Hacker News makes some form of this argument in the comments on this article. ↩︎

  3. Sometimes these arguments have more to do with the non-discrimination clause of the OSD. I have a different set of arguments for this situation. ↩︎

Articles from blogs I read Generated by openring

Status update, April 2024

Hi! The X.Org Foundation results are in, and I’m now officially part of the Board of Directors. I hope I can be of use to the community on more organizational issues! Speaking of which, I’ve spent quite a bit of time dealing with Code of Conduct matters latel…

via emersion April 16, 2024

M2dir: treating mails as files without going crazy

Sometime recently in the past I complained about Maildir. You can go read the post, but the executive summary is that I think Maildir uses an actively user-hostile directory structure and extremely convoluted filenames that do not convey any meaning at all. …

via blogfehler! April 15, 2024

Go Developer Survey 2024 H1 Results

What we learned from our 2024 H1 developer survey

via The Go Blog April 9, 2024