On Thursday, February 13, 2014 at 1:21 AM, Jonas Sicking wrote:

> I still think that leaving out name and icons from a manifest about
> bookmarks is a big mistake. I just made my case here
> http://lists.w3.org/Archives/Public/www-tag/2014Feb/0039.html
I'll reply separately.  
> Basically I think we need to make the manifest more self sufficient. I
> think that we're getting Ruby's postulate the wrong way around by
> making the file that describes the bookmark not contain all the data
> about the bookmark. Instead the two most important pieces about the
> bookmark, name and icons, will live in a completely separate HTML
> file, often with no way to find yourself from the manifest to that
> separate HTML file.

I still think that icons and name are outside the scope of V1 - but I've added 
them to V2. The whole manifest and icon updating mechanism you describe in your 
email to the TAG adds a bunch of complexity (yes, we need to deal with it 
eventually as it's an extremely valid use case - but we can defer it to HTML at 
this moment and for a few months... even if UAs don't do updating of icons and 
name from HTML). I still hold that we should get the most critical and least 
controversial functionality (display mode, default-orientation, and start-url) 
standardized before we do the other stuff. It also gives a chance for UA's to 
catch up and implement HTML's "application-name" and <link rel=icon> properly.  

UAs are going to need to support those HTML features to work with apps that 
don't make use of manifests. And apps the use manifests will work just fine 
till we add proper support for name and icons into the manifest - all web apps 
will need to include "application-name" and "link rel=icon" (as well as a bunch 
of proprietary stuff!) to target todays and yesterdays UAs regardless. So, 
IMHO, there is not much to be won by putting name and icons into V1 for 
implementers or for developers at this moment.

I would go as far as to say that it's initially harmful to have name and icon 
in V1 because it discourages UAs from fixing their support for application-name 
and <link rel=icon>. Having the fallback behavior explicitly tested in V1 of 
the manifest may help improve support for those features of HTML. 

So, I'm not saying let's never do name and icon - I'm saying let just do the 
easy stuff we have some agreement on first. 

Reply via email to