Re: [w3c/manifest] The new definition of "installable web application" is inconsistent with the normative text of the specification (Issue #1168)

benfrancis left a comment (w3c/manifest#1168)

@dmurph If there is a consensus between browser vendors that any website should be "installable" then I don't want to stand in the way of that (not that I could if I wanted to 😆), and am happy to contribute to figuring out how that fits with the Manifest specification. We actually tried that direction for Firefox OS 3.0 and it opens up interesting possibilities.

The only reason I keep pushing on this issue is that it appears to me that the Manifest specification says one thing (in the informative text) and then specifies another (in normative algorithms), and that has been the case for some time now. With the TAG quoting the phrase "any website is an installable web app" from this specification in a TAG Finding, it suddenly seemed more urgent to resolve that inconsistency, so that if the TAG is saying "[Installing Websites Must Be Easy](https://w3ctag.github.io/install-finding/)" there is actually a concrete definition of what that means. (Thank you for responding there as well.)

Assuming that browser vendors want to standardise the idea of "installing" any website I can think of a couple of options:
1. Specifying that inside this specification by defining a default manifest that can be applied when the website doesn't provide one, to turn a website into an installed app
2. Specifying "add to home/add to dock/create shortcut" for a website as a _separate_ mechanism to installing a web app, which re-uses some of the primitives like display modes and navigation scope, but defines a separate processing algorithm which doesn't require a manifest and manifest URL as input, to create an app from a website. ie. "create an app from a website" vs. "install an installable web app".

Since you've pointed out multiple browsers which already implement a feature like this, maybe it could be useful to document the use cases, and the defaults current implementations fall back on, in order to start to draw out the common behaviour for standardisation?

-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3c/manifest/issues/1168#issuecomment-6214926196
You are receiving this because you are subscribed to this thread.

Message ID: <w3c/manifest/issues/1168/6214926196@github.com>