There is another factor, too: we as coders are principally biased to work on projects which are fun, and that can supersaturate this market. So I just spent a couple days working on a project to embed a couple of lisp-like structures in JavaScript for building up properly-indented HTML templates. The idea came from seeing people who were successful at using the C preprocessor to create templating schemes. I might port the code to a language that I'm less familiar with, like Clojure, just to get a feel for that language, but I wanted the core structure to exist in a context that I was already comfortable with and could debug. (Really that's half the reason why I use Node.js -- back in the IE versus Netscape days, JavaScript was my second real programming language and it always has the benefits of comfort, even as I learn more Lisp and Ruby.)
Anyway, this happens to be useful for me, because it lets me define macros which I can use to build up XML trees. But now if we come to the question: could I sell it for a profit? -- then the answer is "no, because it was too fun to build." Don't get me wrong, it will save me a bit of time (and it might save others a bit of time) to be able to separate my template-development language from HTML -- but I fully expect that a programmer at the startup of your choice who needs it will also say, "this looks like a fun two-day project, let me see what I can do."
Anyway, this happens to be useful for me, because it lets me define macros which I can use to build up XML trees. But now if we come to the question: could I sell it for a profit? -- then the answer is "no, because it was too fun to build." Don't get me wrong, it will save me a bit of time (and it might save others a bit of time) to be able to separate my template-development language from HTML -- but I fully expect that a programmer at the startup of your choice who needs it will also say, "this looks like a fun two-day project, let me see what I can do."