Re: Partial conformance

I think “substantially conforms” would be good to have to reflect implementation reality and reward those who work hard to get their stuff accessible but have to live with some issues they cannot fully bring in line, so thumbs up for this one. It is the inverse of the concept of “tolerances” which has been around for some time. 

For most SCs one can describe situations where implementation is less then perfect but no serious issues exist. Would it be too arbitrary to collect a compendium of such cases per SC as a kind of example-based benchmark (which might be regularly updated to reflect new techniques)? The problem of course in documenting such slack is that it might invite implementors to do things that they shouldn’t. Might still be helpful to build consensus in WG around assessments of ‘tolerance’ or ‘substantially conforms’.
Detlev

Sent from phone

> Am 09.04.2019 um 17:16 schrieb Jeanne Spellman <jspellman@spellmanconsulting.com>:
> 
> We want to do "substantially conforms" (partial conformance is a different concept and we want to keep them separate).

Received on Wednesday, 10 April 2019 05:47:54 UTC