On: An experimental software development environment

I would not recommend this objective (at least not for a few years). The reason I actually think it is important to comment on this is that I've seen the pattern a lot where IDE or language devs strive too early to build their own product in their product. While in theory this is useful, in practice I think it's better to wait 5 to 10 years before doing that. Otherwise it's too much of a hard problem and leads to many premature optimizations and resource allocation while taking the focus off of solving other people's problems.

This tool seems to me like it would be fantastic for building and managing strongly typed, 100% code coverage, reliable bug-free microservices. Lots of other tools help you build microservices, but they help you build BUGGY microservices. I think solving the problem of building reliable microservices could be huge.

But this product itself is not a microservice, instead it's an advanced IDE and backend with a lot of different components. So while it would be cool to use isomorf to build parts of isomorf, I'd vote you delay going for this goal for a long while.

Interesting. Thanks for the feedback. I think you are probably right that it sounds cool and appealing as a dev, but it's likely not a great use of resources in the short term.

Or maybe another way to put it: lots of successful languages and IDEs are implemented in themselves, but lots of unsuccessful ones are too. I think I've seen some folks believe achieving that beautiful recursive milestone will be a silver bullet, but rarely (if ever) does it turn out that way. It's a nice thing to have and a great way to dogfood your product but not necessarily that important.