- From: Daniel Murphy <notifications@github.com>
- Date: Fri, 24 Jul 2020 11:39:30 -0700
- To: w3c/manifest <manifest@noreply.github.com>
- Cc: Subscribed <subscribed@noreply.github.com>
- Message-ID: <w3c/manifest/pull/834/c663677886@github.com>
If I understand this right, we will have 3 notions/options of scopebased on invalidations: 1. The scope in the json 1. The first implied scope, from the start_url 1. The second implied scopes, from a start_url that is changed by the document_url and I believe the proposal is to base the share targets/etc validation on the 2nd of these? I have a crazy idea here. Looking towards the goal of having the manifest being the source of truth for a web app (where we should be able to install a web app with just a manifest) what if we introduce a new noun into the spec - an 'isolated' or 'isolatable' manifest: An 'isolated' manifest is a manifest that has a start_url that includes an origin. 'isolated' manifests are applicable to installing/applying to a document_url IFF the document_url is the same origin as the start_url. This gets rid of the possibility of 3. above, and also doesn't seem to break anything currently (as per mgiuca@'s analysis above). This also starts reenforcing the idea of a manifest being 'all that you need' to install a web app, and starts that definition as something that has a start_url including an origin. It also doesn't invalidate manifest that, say, only have a theme_color, as that is fine and valid. This manifest is not 'isolated', and going towards the goal here, that manifest wouldn't be used to fully describe a webapp to be installed. -- 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/pull/834#issuecomment-663677886
Received on Friday, 24 July 2020 18:39:42 UTC