RE: WOFF and extended metadata

John Hudson said:

> Recommendations for font use that relate to typography, i.e. to best use of 
> the font, seem to me font-level metadata, rather than container-level metadata, 
> and I would expect to find them in the font data, e.g. in your 'epar' table, 
> rather than in the WOFF metadata.

> Since WOFF is designed so that individual font data tables can be separately unpacked, 
> I'm not sure whether there would be a benefit to including such information in the WOFF 
> metadata in addition to or instead of in the font data. And when I say I'm not sure, 
> I mean that I don't know, not that I'm dismissing the possibility.

I agree. I do not quite understand why people are so eager to place lots of additional formatting-rich information into _container_ format. If there is a place for it, it at least should be font itself and not wrapper around it. I'll be happy with simple string content that can be shown to browser's end user, but very basic relevant to copyright and simple licensing terms. For everything else, there is a URL to vendor's site where you can place whatever you want. I would not want to see story of font creation or you photo with Obama with font in background stored in metadata and make UA display it to every user who wants to see what this font is about. This is not the purpose of metadata as I see it.

Vendor-specific metadata is important, but it is definitely not for user's eyes. We can reserve defined place in metatadta which will be known as not touched and displayed by UA. For strings that are not part of predefined fields, but vendor really wants user to see, there may be another clearly designated place, but only restricted to plain strings. It may allow localization similar to defined fields, but not more than that. Add URL to the site where you can get more information, and I'll be happy with the result.

Thanks,
Sergey

P.S. On the other hand, allowing vendor-specific metadata may efficiently kill private data block for any purposes except digital signature.

Received on Friday, 28 May 2010 16:19:53 UTC