Comment by atombender
8 months ago
I think NATS is definitely popular. What is your standard for "taken off", though?
One thing I've learned in my career is that the field is broken into many smaller "bubbles". There is lots of software that flies under the radar if you're not inside the right bubble. Blogs and books aren't necessarily an accurate measurement of how healthy an open source project is.
Sure, there may be less noise around NATS than, say, Kafka. But there could be other explanations for that. From my perspective NATS is one of those well-engineered workhorse technologies that get the job without fuss or hype or (until now) drama. Some tools are just quiet because people are getting stuff done with it. I'm willing to bet it's got decent adoption in Fortune 500 companies.
I don't have any hard numbers. But if you wander into the NATS community Slack, for example (which isn't easy to find, admittedly!), you'll find that it's quite active compared to many other open source projects. Looking at Docker Hub, NATS has very healthy stats, beating prominent projects like Cassandra in terms of absolute pulls, and not far from Kafka.
One thing is certain, Synadia is not going to make NATS more popular by moving to a proprietary license.
On the other hand, if they're pouring money into a project that doesn't make them enough money to make it worthwhile, that does seem unsustainable. Maybe NATS should get less popular but become an option in EKS/AKS/GKE that Synadia runs.
The sustainability of open source doesn't seem to come down to whether the product is open source or not.
As an example, Elastic was a $10B company in 2021 — the year they went dual license — with revenue of around $500-600m. They showed that you can build a huge business on something that's given away for free. I don't know anything about Synadia's financials. It's possible they're successful, but simply want more. It's possible they've not been able to build a sustainable business on NATS. Their commercial offering is basically support plus a closed-source control plane, which isn't exactly a big carrot when the alternative is $0.
I also wonder if there's any VC pressure happening here that could explain the sudden shift. Synadia raised $25m in 2024, and it may be that, one year later, the investors just aren't seeing the progress they were expecting.
Maybe - I wasn't making a point about open source particularly. But you can also have a few unusual examples in times of zero interest rates and they don't necessarily make useful reference points. Only the people in charge will know the whole story and how it relates to their company.