Re: [w3ctag/design-reviews] Web Neural Network API (#570)

Discussed this briefly with @kenchris who kindly volunteered to share his high-level review comments for the explainer:
- Consider making [use cases](https://webmachinelearning.github.io/webnn/#usecases-application) more prominent in the explainer, perhaps note the use cases (as bullets?) in the beginning (currently use cases are noted in the explainer [key scenarios](https://github.com/webmachinelearning/webnn/blob/master/explainer.md#key-scenarios) and linked from the [spec header](https://webmachinelearning.github.io/webnn/))
- Note in the explainer (or in the TAG review request?) there is a [spec-compliant polyfill](https://github.com/webmachinelearning/webnn-polyfill) that [passes the test suite](https://webmachinelearning.github.io/webnn-polyfill/test/) as well as [samples](https://github.com/webmachinelearning/webnn-samples) that implement selected use cases using this polyfill (maybe the review request template could include new fields for polyfill and samples?)
- Note this effort has rather [diverse participation](https://www.w3.org/community/webmachinelearning/participants), including major browser vendors, key ML JS frameworks, interested hardware vendors, web developers
- Note the design process of this API started by identifying [key use cases](https://webmachinelearning.github.io/webnn/#usecases-application), working down the levels of abstraction [decomposing the key use cases into requirements](https://github.com/webmachinelearning/webnn/blob/master/op_compatibility/first_wave_models.md), aligned with the guidance to [put user needs first](https://w3ctag.github.io/design-principles/#priority-of-constituencies)

I probably missed some of @kenchris insights, so please fill me in.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/design-reviews/issues/570#issuecomment-768151707