Re: [w3c/manifest] Icon shapes and masking (#555)

@dominickng, @kenchris and I had a half-hour VC about this and came to a rough consensus for Option B. Meeting notes follow:

- Ken/Matt: Worried about small players (e.g., new operating systems in the future) unable to enter the market to convince devs to add their platform-specific icons (under Option C). End up using iOS or Android icons, essentially ending up with apple-touch-icons all over again (or "vendor prefixing" used outside of the original vendor).
- Dom: Already have a default icon.
- Ken: Default icon is not good enough as a generic thing. Everybody wants a maskable icon of some kind. In the future we could add iOS specific icon in the future for top-tier polish.
- Dom: Same argument would apply then. If we're going to add platform-specific icons in the future, why not just do it now?
- Dom: Exposing the capability to map your icon to platform-specific form, but requiring transform.
- Dom: If only developers who really care are providing the maskable icon, then this is a fine-tuning/tweaking capability, not for widespread use.
- Matt: This isn't just for top 1% of developers; ideally this is accessible to all developers.
- Ken: Developer tools (or Lighthouse, etc) can show you how your maskable icon will look on various major platforms. [Good idea: we should talk to Chrome devtools and/or Lighthouse.]
- Dom: We should therefore push developers to maskable icons.
- Matt: We should give equal weight to both the traditional and maskable icons (both are equally valid, depends on the brand; e.g. Gmail is a top-tier app that just doesn't make sense to have a maskable icon).
- Dom: Option B takes control away from developers.
- Matt: Giving control also means maintenance burden. If the underlying platform changes, we want developers protected from underlying platform changes in the future, not having to update their site every time Android changes, for example.
- Dom: If we go with Option B, we would want non-normative text explaining the nuances, e.g., the border we would be injecting into the icon.
- Vague resolution: Proceed with Ken's #657 (Option B). Matt and Dom to look over and suggest non-normative text that highlights some of the nuances of this discussion.

Thanks everybody!

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3c/manifest/issues/555#issuecomment-407596320