It’s hard for me to see what the future is for this “commercially supported open source” business model, particularly for infrastructure components like this.
I start with the assumption that most companies want to use managed services for their infrastructure. This makes sense, because the marginal cost for (eg) AWS to provide the managed service is basically nothing over the cost of the underlying infrastructure (because at that scale you have to automate the management and can’t have a human in a loop for anything routine). But for the company using that service, the cost is non-zero since they likely don’t have that automation in place and have to have expensive humans do that work. So in a reasonably efficient market, managed infrastructure should be the correct financial decision (and I tend to think that the popularity of it implies that things are generally working that way).
So if you want to use a managed service, it’s fine if you’re dealing with something that is (eg) MIT licensed because you can expect a healthy competitive market for running it as a managed service. But now you look at something like elasticsearch, and at this point what benefit do I have from using it over Amazon’s OpenSearch fork? I’m just as locked in if I use elastic since their license precludes anyone else from providing it as a managed service which is a fundamental requirement of my usage of the tool.
Idk, I could be wrong! But I really don’t know what the path forward is for this model.
I start with the assumption that most companies want to use managed services for their infrastructure. This makes sense, because the marginal cost for (eg) AWS to provide the managed service is basically nothing over the cost of the underlying infrastructure (because at that scale you have to automate the management and can’t have a human in a loop for anything routine). But for the company using that service, the cost is non-zero since they likely don’t have that automation in place and have to have expensive humans do that work. So in a reasonably efficient market, managed infrastructure should be the correct financial decision (and I tend to think that the popularity of it implies that things are generally working that way).
So if you want to use a managed service, it’s fine if you’re dealing with something that is (eg) MIT licensed because you can expect a healthy competitive market for running it as a managed service. But now you look at something like elasticsearch, and at this point what benefit do I have from using it over Amazon’s OpenSearch fork? I’m just as locked in if I use elastic since their license precludes anyone else from providing it as a managed service which is a fundamental requirement of my usage of the tool.
Idk, I could be wrong! But I really don’t know what the path forward is for this model.