Closed pes10k closed 2 years ago
In #59 , I've included a text to clarify the relationship.
Pasting the content for an easier review:
This specification defines the set of metadata to describe MiniApps, extending the [[[APPMANIFEST]]] and the [[[MANIFEST-APP-INFO]]] by providing additional mechanisms to describe and set up MiniApps. The MiniApp Manifest directly reuses essential elements from these specifications (i.e.,
name
,short_name
,description
, andicons
), adding supplementary members that specifically affect MiniApps (e.g.,app_id
,version
,platform_version
,device_type
,pages
,req_permissions
, andwidgets
) and their look and feel (e.g.,color_scheme
andwindow
).
Hi @espinr , i think those are good changes. If this spec is designed to augment / suplement AppManifest, then it'd be ideal to only define those suplements / additions in your spec, instead of effectively inlining the entire AppManifest spec. Woudl that be possible?
Yes, you are right @pes10k. It makes sense. Historically, this manifest has included all the members, and we've kept them as a Web App manifest profile. We need to mention those members reused from the Web App Manifest because this document introduces some constraints to the Web App Manifest members (some are mandatory in this spec). But, you are right that we don't need to re-define all of them in the specification if we just reuse them. I suppose we can just refer to the original definition.
In the meantime, I've added that this spec introduces constraints to the Web App Manifest.
This issue is being filed as part of of PING's horizontal review
The substance of the proposal and its intended use cases are similar to the Web Application Manifest proposal (describing a packaged web or web-like application). It would be ideal to combine these in some way, specify one as an extension to the other, or otherwise remove the complexity of two different nearly-but-not-identical proposals.
If thats not possible, it would be good to at least have some non-normative text in this proposal explaining the relationship of the two specs and (possibly) why they need to be mostly-identical-but-fully-independent specs
This is not a blocking privacy issue, but I'm mentioning it since it came up during the review