`<tiles>`?

[Malvoz]

I saw a <tiles> element in some of the test or demo pages, it doesn't exist in the spec, should it be added? Also, maybe it could be named more distinctively from <tile>, e.g. <tilegroup>?

[prushforth]

Yes, I think this was not necessary. This element was added to the shadow DOM, and although it is or was likely used in some way by my code, I don't think it desirable to propose to add it to the HTML language, at least not at this time. As an action, we can leave this here to see if we want to try to remove use of this element by the polyfill, at some point in the future. Even in a native browser implementation, they will likely find it necessary to implement (internal) abstractions that allow rendering of spatial information appropriately. I ascribe this element to this category, that of internal abstraction that we don't want to have leak out.

[ahmadayubi]

It was originally just using a <div> but along the way it was suggested that it should use <tiles> instead, it should be able to revert back to <div> pretty straightforwardly, if <div> is the appropriate container to use in this case.

[prushforth]

I guess it would be fine to revert to div, but again because it's an "implementation detail", I don't think it's high priority to do so, and indeed it may add value / clarity for ourselves as implementers while debugging, although at the cost of some confusion as to what's in scope of the proposal vs what's out of scope. I could be off base.

[Malvoz]

Fixed in #515.