I made something like this as well. A system for handling load order with a cool syntax, cycle detection and small footprint. There's something cool about writing dependency tools. I like the feeling of my own code bootstrapping my own code. No foreign includes in the system. But this is only on JS client side web, where ES imports haven't landed yet and I don't like the idea of using require or amd. On server side I'm perfectly happy to use require and npm and loads of other people's code.
I wish i know how to make code like this popular. I guess the problem is js developers already already have workable solutions that are basically monopolies. Still it's an interesting challenge to take their market.
It seems very hard though because if you think about any Improvement you could make to your code over theirs the existing solutions could make that Improvement too and their Improvement would have so much more utility than your own because it would be able to instantly reach all their existing customers. So it's quite the challenge to take market share from require and Amd and so on.
I guess the trick then is the old trick try to pick a niche that the Monopoly doesn't really serve. Amd and require are for everywhere .... But solutions like yours and mine might work better for small to medium sized projects in the browser. Or an even smaller niche maybe programmers of games or Demos in the browser. Size and complexity conscious.
I'm not trying to take anyone's market with this, more that I couldn't find any de facto solution that solved this problem when I went looking for why my scripts were waiting 400ms to load, and started thinking about async scripts.
So I went and wrote something simple, and hopefully versatile!
2 comments
[ 3.0 ms ] story [ 17.3 ms ] threadI wish i know how to make code like this popular. I guess the problem is js developers already already have workable solutions that are basically monopolies. Still it's an interesting challenge to take their market.
It seems very hard though because if you think about any Improvement you could make to your code over theirs the existing solutions could make that Improvement too and their Improvement would have so much more utility than your own because it would be able to instantly reach all their existing customers. So it's quite the challenge to take market share from require and Amd and so on.
I guess the trick then is the old trick try to pick a niche that the Monopoly doesn't really serve. Amd and require are for everywhere .... But solutions like yours and mine might work better for small to medium sized projects in the browser. Or an even smaller niche maybe programmers of games or Demos in the browser. Size and complexity conscious.
So I went and wrote something simple, and hopefully versatile!