- From: John Foliot <john@foliot.ca>
- Date: Thu, 29 Apr 2021 08:57:24 -0400
- To: Wilco Fiers <wilco.fiers@deque.com>
- Cc: Alastair Campbell <acampbell@nomensa.com>, "w3c-wai-gl@w3.org" <w3c-wai-gl@w3.org>
- Message-ID: <CAFmg2sVavDvny6WMjWo_ve15m_LWYeYnXKahyhwycZLCLSh4sg@mail.gmail.com>
Wilco writes: > One of the most common time sucks I deal with in my job is people reporting that when they did two tests for which they expected to see the same results, get slightly different numbers. a HUGE +1 JF On Thu, Apr 29, 2021 at 6:53 AM Wilco Fiers <wilco.fiers@deque.com> wrote: > > Done well, the effect of *accounting* for subjectivity (that will always > be present) is that it smooths out the results, and you get a closer > agreement between testers overall. > > That has not been my experience. People having a different understanding > of what a success criterion means is a far more common cause of variations > in test results than them disagreeing about if some alt is good enough a > description for an image, or other subjective things like that. > > One of the most common time sucks I deal with in my job is people > reporting that when they did two tests for which they expected to see the > same results, get slightly different numbers. I'm definitely not looking > forward to consultants spending 20% of their time explaining why that 3.52 > average on page X is a 3.48 in the new test, even though page X didn't > change. > > > > On Thu, Apr 29, 2021 at 11:40 AM Alastair Campbell <acampbell@nomensa.com> > wrote: > >> JonA wrote: >> > > Folks seem to associate non-binary score with subjectivity. There is >> > > already subjectivity in WCAG today >> >> Patrick wrote: >> > ... heresy! *grins* >> >> It's a good point to bear in mind. >> >> Even in (seemingly) simple cases like alt text, an image can have alt >> text that may or may not have " equivalent purpose". >> >> Currently that would mean different testers have to choose between pass >> and fail, which makes the end result look wildly different. >> >> With categories / rating / percentages within guidelines, the difference >> between category 2 & 3 out of 4 (or between 50% and 60%) does not look as >> different. >> >> Done well, the effect of *accounting* for subjectivity (that will always >> be present) is that it smooths out the results, and you get a closer >> agreement between testers overall. >> >> Kind regards, >> >> -Alastair >> > > > -- > *Wilco Fiers* > Axe-core product owner - Facilitator ACT Task Force - Co-chair ACT-Rules > > > -- *John Foliot* | Senior Industry Specialist, Digital Accessibility "I made this so long because I did not have time to make it shorter." - Pascal "links go places, buttons do things"
Received on Thursday, 29 April 2021 12:57:54 UTC