Curious about why still at stage-1 after 2 years?

[jay-jlm]

Hi, Last week I ran across a podcast interview from Oct 2015 where this project was mentioned. As a big fan of RxJS, I'm just curious on why this is still at Stage 1 after almost 2 years?

[dinoboff]

AFAIK, There's no much interest from browser vendors to use them (e.g. as a returned value from addEventListener - or from an alternative method).

[benjamingr]

@jhusain ping

[dinoboff]

Also https://github.com/rwaldron/tc39-notes/blob/master/es8/2017-05/may-25.md#17iiia-observable-proposal-to-stage-2

[benjamingr]

Looks like this proposal is dead then. Shame, was a lot of effort.

[zenparsing]

@benjamingr I wouldn't quite put it like that. There are supporters on the committee and to be honest, this proposal is a lot more solid than some proposals which are currently sitting at stage 2.

But even if it remains at stage 1, we've successfully created a high-quality de-facto standard, useful both inside and outside of the "reactive programming" scene. I'm not sure TC39's blessing is all that important at this point.

There are a couple of things that make me sad:

• I still think that the current error-swallowing behavior is wrong (and I don't think it will get traction anyway).
• The placement of the @@observable symbol on Symbol is a bit of a pain, since it has to be polyfilled. It makes sense to put it there if Observable is a built-in, but I don't like that we have to polyfill. Maybe we need a "well-known symbols" registry on NPM where we can put symbols like that.

[keithamus]

The placement of the @@observable symbol on Symbol is a bit of a pain, since it has to be polyfilled. It makes sense to put it there if Observable is a built-in, but I don't like that we have to polyfill. Maybe we need a "well-known symbols" registry on NPM where we can put symbols like that.

Symbol.for with a well agreed upon name could be used for this; e.g. Symbol.for('--tc39proposal-observable')

[zenparsing]

@keithamus True, except some of us still have to support IE11 (for the time being) 🙄

[keithamus]

Yes; sadly 😒

[benjamingr]

Thanks for still being involved @zenparsing :) I'm not as optimistic as you, I'd definitely like native support at some point. I think we should be looking for happy DOM consumers if we can.

[jhusain]

@benjamingr that's really the next hurdle: to convince implementer that Observable is a compelling enough improvement over EventTarget to make progress. As @zenparsing points out, the work put into this proposal is still valuable regardless of what happens. At the very least we can use it put together an interoperability spec.