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)

@tomayac wrote:
> This [Solitaire](https://www.watsonbrosgames.com/solitaire/) game is a perfect example where installation without a manifest would be desirable.

OK, that's an example of a "genuine web app which doesn't provide a manifest" as I described above. In this case the page title makes for a bit of a long app name and the icon is an emoji inside an SVG, but they just about work as fallbacks to generate a shortcut icon. Now how does the user agent go about automatically choosing an appropriate navigation scope and display mode?

I'm assuming that's why if I try to "install" that app from Chrome on Android today it just says "This app cannot be installed", and Chrome on macOS does not offer to install it. That is why I say "any website is an installable web application" is an aspiration rather than a statement of fact.

Most mobile browsers will let you "add to home" to add a shortcut to that specific web page that opens in the browser, but is that really "installing" the website?

> The game doesn't have a manifest, but it works perfectly as an installed app.

But how does the user agent determine that? And what about the _billions_ of other web pages for which that is not true?

I would call that Solitaire game a "web app", but not an "installable web app", because it doesn't provide sufficient metadata to be installable.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3c/manifest/issues/1168#issuecomment-6190583392
You are receiving this because you are subscribed to this thread.

Message ID: <w3c/manifest/issues/1168/6190583392@github.com>