- From: Adrian Hope-Bailie <notifications@github.com>
- Date: Mon, 12 Sep 2016 07:11:00 -0700
- To: w3c/manifest <manifest@noreply.github.com>
- Message-ID: <w3c/manifest/issues/161/246359248@github.com>
> Unfortunately that's not how the web's built today. Take github pages for example - one origin, many sites. I don't follow this. Github uses subdomains so each Github site is a unique origin. > So adding a payment app will require a permission prompt? Yes, that's certainly my assumption. > I'm trying to get to the root of why using manifest is so desirable here. Re-use of the manifest data model. An "extensible Web" approach to Web architecture. The issue is that the concept of an application manifest is broad and so is the definition of a Web application. The idea of providing meta-data about your application has a many use cases. This spec has taken the idea of a standard way of defining app meta data and put it in a spec that is very difficult to use unless your definition of an app, including how it is installed is the same as the editors'. This is a website manifest, not an app manifest. This data model, a nice generic one that SHOULD be reusable and extensible, is tightly coupled to an installation process that is not. It's just one of many examples where there are major inconsistencies between W3C specs on simple architectural concepts like what defines an app boundary. Pity the poor fool who wants to become a Web developer 😢 -- 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/161#issuecomment-246359248
Received on Monday, 12 September 2016 14:11:33 UTC