Providing a fallback image inside <x-picture> results in fallback always downloading

[uniqname]

Not sure there is a solution to this. It illustrates one of several things that a non-native solution can't do or can't do well.

[marcoscaceres]

can we wrap it in noscript to make it inert?

[uniqname]

We could (and should - or shall), but that only addresses SEO -- which, in theory is already addressed by the alt attr on the picture tag -- and the already minuscule and ( anecdotally) increasingly diminishing numbers of users who don't have javascript or have it turned off¹

It doesn't address a failure in x-picture or polymer².

But more important to my over-excited sense of principle, it means it won't mirror the proposal exactly (<x-picture> vs <picture> withstanding)³

In the end, I don't have a solution I am excited about, but <noscript> at least begins to address some issues, and it's far better than downloading an extra image that will never be seen. I wonder if it's better than not including an image fallback at all?

¹Numbers on JS disabled are incredibly hard to come by, with the afore citation from 2010 the most recent -- nearly irrelevant at this point.
²This may be acceptable. Both x-picture and Polymer should handle errors well internally.
³This, of course is me being impractical.

[marcoscaceres]

The guts of the implementation (in the shadow DOM) should not matter too much, IMHO. We gotta use whatever tools we have at our disposal to make it work.

[uniqname]

@marcoscaceres I'm second guessing how I understood your original comment, which makes me think I might have been too vague in the issue's description. The current problem is the fallback in the markup. The implementation in the shadow DOM is solid as far as when things download. If we include <noscript> around the fallback <img />, that solves for the example usage, but it is something developers using x-picture would need to remember to do as well, or their users will suffer the a double download too. Thoughts?

[marcoscaceres]

Sorry, I've been sitting on this for a while because of the src-n proposal. It will likely have significant implications on what we do here (specially if src-n kills the picture element).

On Monday, September 16, 2013 at 11:22 PM, Cory wrote:

@marcoscaceres (https://github.com/marcoscaceres) I'm second guessing how I understood your original comment, which makes me think I might have been too vague in the issue's description. The current problem is the fallback in the markup. The implementation in the shadow DOM is solid as far as when things download. If we include around the fallback , that solves for the example usage, but it is something developers using x-picture would need to remember to do as well, or their users will suffer the a double download too. Thoughts?

—
Reply to this email directly or view it on GitHub (#1 (comment)).

[uniqname]

Yes. I've been watching that closely. I've been meaning to find some time to prollyfill that as well to understand how well it would work.

On Oct 7, 2013, at 8:27 AM, Marcos Caceres notifications@github.com wrote:

Sorry, I've been sitting on this for a while because of the src-n proposal. It will likely have significant implications on what we do here (specially if src-n kills the picture element).

On Monday, September 16, 2013 at 11:22 PM, Cory wrote:

@marcoscaceres (https://github.com/marcoscaceres) I'm second guessing how I understood your original comment, which makes me think I might have been too vague in the issue's description. The current problem is the fallback in the markup. The implementation in the shadow DOM is solid as far as when things download. If we include around the fallback , that solves for the example usage, but it is something developers using x-picture would need to remember to do as well, or their users will suffer the a double download too. Thoughts?

—
Reply to this email directly or view it on GitHub (#1 (comment)).

—
Reply to this email directly or view it on GitHub.

[marcoscaceres]

Yeah, that was my plan also :) I've made a start here (still using the old syntax):
https://github.com/marcoscaceres/srcn.js

Will try to update it next week. Happy to collaborate.