W3C home > Mailing lists > Public > public-silver@w3.org > May 2020

RE: Scoring and Dashboards

From: Alastair Campbell <acampbell@nomensa.com>
Date: Fri, 22 May 2020 22:17:04 +0000
To: John Foliot <john.foliot@deque.com>
CC: Silver TF <public-silver@w3.org>
Message-ID: <AM7PR09MB4167745D7199C556A0BCB76BB9B40@AM7PR09MB4167.eurprd09.prod.outlook.com>
Hi John,

> My point however is that we've publicly decided to use a "Score" based  measurement system, and that scoring mechanism needs to work for both short-term (development time) and longer-term reporting metrics. This is why (for example) I keep asking about depreciation of 'scores' over time.

People create scores out of current WCAG issues already, possibly with numbers of instances as well (I assume). That’s a score, and we don’t currently provide a decay rate.

I’m still not seeing why a decay rate should be built into the standard. If you’re using car analogies, in the UK we have to get a car tested once per year (after a 3 year grace period when it’s new). You get it tested and a soon as you drive away, that MOT is not valid any more. I.e. it was a point in time test.

In the context of websites, there are so many variables to consider I don’t see how it is right or possible for a group to define those for everyone. There may be time triggers (e.g. yearly), but if I were creating tooling for this, I’d also want to build in triggers for when content or features are updated (to be configured by the customer).

I think it should be up to regulators (e.g. the EU Public Sector Bodies regs define yearly, with some flexibility for how much testing is done yearly), and toolmakers to provide suitable options for different situations.


> with WCAG 3.x all of those tools will need to be able to use the same WCAG 3.x scoring mechanism, so that in practice each tool would return the same basic 'score' of any given piece of content.

Sure, but how is that different from now? The score should be more granular, but I don’t see a decay rate as a logical conclusion from that.


> So to be crystal clear, what I am suggesting is that we, as a group, need to also define that time-based metric (i.e. the impact of time on test results) as part of any "Final Score", which is one of the main deliverables of this TF

I disagree, and it isn’t something I can see in the requirements. It could be added, with agreement, but it wasn’t part of the plan as far as I can tell.

Kind regards,

-Alastair
Received on Friday, 22 May 2020 22:17:19 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 24 March 2022 20:31:48 UTC