"I have done some original research outside of academia (mostly in the form of source code, some semi-published)" I'm not sure what you're saying here. Source

"I have done some original research outside of academia (mostly in the form of source code, some semi-published)" I'm not sure what you're saying here. Source isn't something that gets published, source code is at best a demonstration of the concept you publish about. Either you have some papers published (that means, peer-reviewed from a real journal, not uploaded to xyz 'open access' journal) or you don't. To 'work the system', you need to get real papers published, several, for a number of years, and work with people in positions that can get you into a university. That way you can sidestep the several years postdoc route. Or become an adjunct. As an adjunct professor you are 'affiliated' with a university, giving you access to some resources (usually) and a name/email address to put on your papers. So yes you need a source of income besides that. " Since I am in software I don't really need facilities...." Yes you do, maybe not in the sense that you need hardware, but you need an environment to reflect your ideas on, and to give you social proof that you're not a quack working from his basement. Yes it's theoretically possible to get published without an academic affiliation, and it sometimes happens, but you'll never be taken serious to the same extent your competition is.