| I commented on a 2 day old post on reddit (https://www.reddit.com/r/programming/comments/845s2c/how_graphql_replaces_redux/dvrcqs9/?context=3) but I think it might be more of a HN question someone said >> I'm.telling you guys : fuck frameworks. Fuck graphql.
Yep. This is why I prefer my custom languages that cover my specific needs. and I said If you're not being sarcastic, can I ask what you use to create them? I've been looking at MPS, but I also need to consider whether it will be possible to get people who have MPS competencies, -- seems there are not many MPS developers, so it seems like an infeasible business strategy to depend on it. I would think if an organization uses its own special language it might be difficult to find developers willing to work with it -- and so a custom language could be an infeasible business / development strategy -- better to build a framework in a hireable language. https://i.imgur.com/Z3TQTwj.jpg Or are the advantages of custom languages so great that it doesn't matter? I guess what I'm saying is, -- yeah sure don't get hung up in frameworks, but if you're an organization that wants to be able to hire people in the future to deal with your code base, maybe you don't want to construct some arcane language. Maybe a custom language is not really feasible as a business strategy if you aim to be scalable and to scale, which is critical for investment. Maybe the better strategy is to be able to attract developers. Developers want to evolve skills relevant to their future hirability, or to future projects. If you're a lone dev on some indie project that's never gonna be worked on by anyone else -- sure -- do what you like. -- but is it a decent long term strategy? And so ultimately, is it a good investment? |