Pretty cool how you're able to come out in the open with all this stuff like this!
Having checked Zestful's page out, I got a suggestion for you. How about building a client lib ready for developers for at least a few mainstream languages? This should be rather easily achievable if you type in the Swagger specs to your API. That allows you to generate (Swagger codegen) those libs with possibly just minimal tweaking left on your part.
>Having checked Zestful's page out, I got a suggestion for you. How about building a client lib ready for developers for at least a few mainstream languages? This should be rather easily achievable if you type in the Swagger specs to your API. That allows you to generate (Swagger codegen) those libs with possibly just minimal tweaking left on your part.
This is something I've thought about. I'd really like to have a PyPI package that you can just pip install and it parses ingredients using the free tier, then prompts you to enter a license key when you reach the limit.
Zestful's hard for me because I could spend a year pursuing all of the ideas I have for it, but historically, feature ideas I come up with myself haven't attracted new customers. I set a rule for myself to not implement any new features until a paying customer requests it, and customers haven't mentioned wanting language-specific libraries. I talked a bit more about that in my August 2019 retrospective:
As someone who has done quite a bit of shopping for various services, having a language-specific library ready to go is a major factor. I doubt that someone already paying you would still need one, though, so it should probably be considered a strictly "customer acquisition" feature.
I understand your rule, but I don't think it should be that strict.
It could also be his competitive advantage over others.
I mean wasn't Stripe known to become so successful exactly because of easiness of use, in other words a good user experience for developers with simple ready-made libraries. Making a new payment gateway is hardly innovative by itself after all.
Right, I don't know if it makes sense business-wise or not. Probably not if no one is voicing out requests about it.
However, writing Swagger API specs for your future projects while you develop them might make you more productive, plus you do get that code generation for free on top of it. Depending on what language you work with, it may come essentially for free or with non-marginal effort.
Having checked Zestful's page out, I got a suggestion for you. How about building a client lib ready for developers for at least a few mainstream languages? This should be rather easily achievable if you type in the Swagger specs to your API. That allows you to generate (Swagger codegen) those libs with possibly just minimal tweaking left on your part.