Irakli Gozalishvili

> I’m not sure what this accomplishes. > > Is this meant to convert between the new and old CID implementations? > > If so, this isn’t going to work...

@mikeal I think you were reading this in context of new CIDs work happening in IPLD, which is likely due to me mentioning comment about `toString` method changes. I think...

> As proposed it looks like asCID is doing a bunch of the checks that the CID class does in its constructor. It does same exact check as `isCID` as...

> Having to use an asCID() would mean we're creating a ton of null assignments for most object properties and doing some amount of internal property inspection for a checkCIDComponents()....

> Throwing when it encounters invalid data is exactly what the constructor [does right now](https://github.com/multiformats/js-cid/blob/master/src/index.js#L114). Ok but here is the prominent pattern I see today within PL code base and...

Had a conversation with @mikeal today about this & I would like to provide a sketch as what I think we agreed was a worthy avenue to explore. ```js class...

Me and @mikeal also have talked about the idea of representing `CID` as data view over the underlying buffer similar to how `Uint8Array`. If we were to go about it...

> Why is it nice to have to break JSON.stringify() if your data contains CIDs? Please correct me @mikeal if I got it wrong, but I think it was something...

I see now that `JSON.stringify` was also discussed at https://github.com/multiformats/js-multiformats/pull/18#issuecomment-645068878 and was decided in favor of not breaking `JSON.stringify`, I've updated sections to reflect that decision.

Me @achingbrain and @hugomrdias had a discussion about this on call other day, and I will try to summarize it here. - There is a strong preference towards having either...