- From: Patrick H. Lauke <redux@splintered.co.uk>
- Date: Thu, 2 Jun 2016 09:56:30 +0100
- To: w3c-wai-gl@w3.org
On 01/06/2016 18:53, David MacDonald wrote: > Our original language for testable was 8 out of 10 experts would agree. > This would be an inter reliability ratting of 0.8 out of 1.0. It may be a minor point, but: I'd prefer language that's a little less...specific. Giving an actual figure of "8 out of 10" gives it a whiff of "it can be proven with hard numbers", sure, but really: if there's ever a disagreement, do we really expect somebody to gather 10 experts, get their opinions, and then make go for the option that had 8 votes? What if it's 5 out of 10...a draw (which is probably why you'd want 9 experts to be able to determine at least majority, barring abstentions). In short, I'd be in favor of more generic, common sense, plain language (though I can understand how some may feel that "vague" language like "most auditors will agree" or similar may not fly for a document used for legislation, but then I guess that's where precedents / case law would come into play). P -- Patrick H. Lauke www.splintered.co.uk | https://github.com/patrickhlauke http://flickr.com/photos/redux/ | http://redux.deviantart.com twitter: @patrick_h_lauke | skype: patrick_h_lauke
Received on Thursday, 2 June 2016 08:56:55 UTC