Open bmeck opened 6 years ago
That's a good idea to have. Only thing is that this could technically be an entirely separate module format. So for example just being a different module type could be the signal to V8/Eval/Interpretor that whatever is inside of the tag is an ESM module.
If that's not strong enough we could certainly talk about having that.
If it uses a script tag, we should definitely require type=module
. I think I am mentioning this 3 times in my Polymer presentation.
That is if it is HTML!
If you aren't writing html please use non-conflicting tag names
Other languages share syntax in one way or another. The goal is to have likeness to respective techniques, that can be parsed accordingly, meanwhile augmenting the behavior.
Honestly I don't see a problem with script type module. But I also don't see it necessary also.
Other languages share syntax in one way or another. The goal is to have likeness to respective techniques, that can be parsed accordingly, meanwhile augmenting the behavior.
I agree, as long as we don't ever call this stuff HTML that seems fine.
It seems odd that it looks so much like HTML on that point. Is there a reason to look similar enough to be mistakable but not conform? Doesn't that just add overhead vs making a different syntax?
Honestly I don't see a problem with script type module.
Because it invalidates programmer intuition that script tags are Scripts without type=module
. They now need domain specific knowledge around how their toolchain works w/ pseudo-HTML.
But I also don't see it necessary also.
I tend to think that things should not be done solely when necessary, very few things are truly necessary. I argue that it is necessary to include type=module
for consistency if this language which is not HTML is attempting to use HTML semantics without adding to developer fatigue.
Required for export syntax in DOM.