- From: Daniel Dardailler <danield@w3.org>
- Date: Tue, 22 Jun 1999 17:36:47 +0200
- To: Al Gilman <asgilman@iamdigex.net>
- cc: w3c-wai-gl@w3.org
> Drawing a line in the sand to divide what is OK from what is not OK may not > be the most appropriate response to all needs. Perhaps publishing > reading-level information in page metadata is a way we can offer relief to > the non-reader without banning reading from the Web. Several questions come to mind. First, can we come up with an objective rating system, vs a subjective one, in the area of readability ? An objective system is one that can be used reliabily by different persons and still give the same answer: e.g. I look at a page, I looked at the definition of reading level 0 to 4, and I say, this one is a 3, and so would anyone with the understanding of the definitions. In the field of metadata for protection of children (i.e. level of sex, nudity, violence, etc rating), the example would go like: a subjective rating system: 0 level of violence ok for toddler 1 level of violence ok for pre-teen 2 level of violence ok for teen 4 level of violence ok for adult an objective rating system: 0 No aggressive violence; No natural or accidental violence 1 Creatures injured or killed; damage to realistic objects 2 Humans or creatures injured or killed. 3 Killing with blood and gore 4 Wanton and gratuitous violence Whereas I wouldn't be able to come up with the same answer as my neighbour with the first system (what I find OK for my kids is my business), I would probably arrive at the same conclusion with the second (either I see gore - as defined out-of-band, either I don't). Anyway, that's the first issue, and the effort required to come up with a standard rating system in a given field is proportional to the difficulty of attaining objectiveness in the ratings. Conversely, without an objective rating system, we won't go very far there. The second point is deployment: say we have an objective rating system, how to we go about having people using it on their pages ? That means a syntax to encode it (PICS is possible, RDF too), promotion, tools to produce it, tools to filter on it, etc. I don't want to paint a dark picture here, but I've been on this path before with deployment of PICS and it's all but an easy path.
Received on Tuesday, 22 June 1999 12:20:53 UTC