I do like the idea of people habitually paying for FOSS, but I've wished for a while that there were more of an ecosystem of consulting companies around it. I k

I do like the idea of people habitually paying for FOSS, but I've wished for a while that there were more of an ecosystem of consulting companies around it. I know of companies that support specific software, but I'm thinking of a full-stack consulting shop: they go to a customer, figure out what they want, and then make whatever fixes they need to make that happen, from the kernel on up. Maybe I'm just not seeing them, but other than Red Hat, I don't know of companies like that. I think they could make money for two reasons: first, people (especially businesses) are willing to pay for convenience, and second, they would be making money for collecting, configuring and modifying software, which means doing actual work. They could pretty easily make the case that if their customers didn't pay them, they would have to do the same work themselves. The companies would have an advantage because, unlike a small IT department at a mid-sized company, they could afford to hire enough developers to have someone who really knew all of the software in a standard GNU/Linux system. And they could afford to hire those people because they would have lots of different contracts. They would have an advantage over proprietary software companies because they would have a huge amount of flexibility to modify the software they use. Again, I think Red Hat does something like this, but we need a few different companies that compete with each other to really grow an ecosystem. I've been hoping that Canonical will turn into something like this (and I think they might hope that too). Having multiple vendors makes customers feel safer, because they can't be locked-in. I think a lot of developers would want to work at a place like that, as long as they had any sort of reasonable division of labor between working on specific customer projects and maintaining software. Maybe you'd spend half of your time working directly on a project, and the other half refactoring the coreutils (or whatever). But that's really a business necessity, because if you didn't take the time to refactor, you'd never be able to help other customers in the future because the code would become a mess.