Does open source still make business sense? How can the open source movement be sustained and funded in the future? Is the open source business model broken or does the magic cauldron still have power?

Programming is interesting because once a problem is solved, it is solved forever. We need not reinvent the wheel; we can share information freely with one another through open source. Code is a formula for automation, like a mathematical equation we can use to evaluate a chess position.

If code is like a math equation, then software is like statistics. I hate statistics, but alas, it can be used for things like predicting the weather, so I suppose it has some utility. As the climate changes, so do our formulas for predicting the weather. It is the same with software. In this way, software engineering makes sense from a business sense. Open-source software can be profitable in this way.

ESR claims two things that I would like to talk about:

  1. The closed-source approach makes independent peer review impossible.
  2. The open-source approach prevents profits from data collection.

Generally, these two principles are true, but the claims overlook some key points. While closed-source software is naturally opaque and therefore untrustworthy, there are ways to provide services without requiring users to trust the closed-source software. Cryptographic techniques exist to make information private. In this way, no peer review would be necessary.

The other statement is also not necessarily true. Open-source software can and often do collect data. Most of the time, it is just crash diagnostics and usage analytics, but it can be more. All services are vulnerable to cybersecurity threats, and breaches can and do happen. Depending on the amount of information stored, data can be leaked. So while they may not profit from data collection, users will always have something to lose by sharing personal data.

Nowadays, I think the open-source business model is sustainable, but not as lucrative as the open-core business model. The open-core business model is to the MIT license as the open-source business model is to the GPL license. Open-core is more permissive with what you can profit from, whereas open-source restricts you to open software only.

With that said though, I actually think there could be a push for more open open-core services. Apple’s private cloud compute service comes to mind. While it’s not fully open-source, they did make some components available on GitHub, albeit with a very restrictive license. Obviously, this doesn’t necessarily prove anything; there’s no way to prove that their closed-source software is the same or comparable to their open-source components. But, I think it shows that it is profitable to be open-source, even if only partially. Of course, that all depends on how much you trust Apple.