RE: WOFF and extended metadata

On Thursday, May 20, 2010 4:19 AM John Daggett wrote:
> 
> Christopher Slye wrote:
> 
> > These points are understood and not disputed:
> >
> > - The WOFF extended metadata block is optional.
> > - A user agent is not required to do anything with metadata if/when
> >   it's present.
> 
> Good.  Requiring metadata doesn't make much sense to me, it's really up
> to the person packaging a font whether metadata is necessary or not.
> And as Sergey pointed out there may be user agents that would never 
> access the metadata information, analogous to user agents that lack 
> "View Source" functionality.
> 

I think we need to be clear about two very different things - requiring metadata to be added to a WOFF file vs. requiring UA to access it. In my opinion, considering metadata as a special purpose API is a good analogy. When an API is supported by all User Agents - a "developer" is not obligated to use it, but it can be used whenever it makes sense. For developer - any API is optional meaning that it is only used when necessary.

However, this becomes very much different issue when the API support itself is optional. From a developer perspective - if API is not universally supported on all platforms it loses its value for a developer. A feature that is known to work in some UAs and not work in others can only be useful if the percentage of UAs supporting it is *significant*.

Going back to the question about extended metadata support:
- I think we are all in agreement that requiring metadata to always be present makes no sense, it should be present only on "as needed" basis.
- Requiring user agents to always parse and validate the metadata makes no sense - UA should do it only when an end user wants to see it using some form of UI for displaying metadata. 

The big question remains - what can we do to ensure that the significant percentage of UAs offer support for some form of UI displaying extended metadata to enable font vendors, web authors and end users "reap the full benefits, whatever they may be".

Regards,
Vladimir

Received on Thursday, 20 May 2010 09:37:36 UTC