Dispatches From The Internets

I love this project. It ticks so many boxes for me as it’s all about doing more with less. I love the ways they’ve found to reduce consumption on the server side and throughput to their users as well. So much to unpack!

Amy Leak offers a straightforward comparison of alt text and captions, including some excellent examples of when and how to use each. There’s so much great advice in this article, I won’t take up any more of your time. Go read it!

More great thoughts on privacy and permissions from Jo Franchetti. I love her list of suggested improvements at the end. Especially “Encourage/standardise clearer wording and UI to explain reasons for permissions and how long permissions will be granted.”

I’ve been thinking quite a bit about privacy and permissions both generally and in web browsers and our product that rely on them. I very much appreciated this perspective from Sally Lait.

[M]y personal preference is generally to continue as normal, and then to use whatever’s being requested as an enhancement, as a natural part of whatever task I am hoping to do, at a time that makes sense to me rather than having it pushed on me out of context or at a time that doesn’t make sense.

This is true for me as well. I hate going to a new site only to be immediately bombarded with requests to see my location or send me notifications. It makes your site appear desperate, socially-awkward, and a bit sociopathic.

I definitely believe there’s room for improvement in terms of how browsers relay requests for permission. Personally, I’d love to see permission requests require an accompanying link to the section of that site’s privacy policy covering how the information being requested will be used. Sure, most users probably won’t click on it, but having to provide something might make some developers think twice about it (and more directly tie these requests to the site’s governing entity from a legal standpoint).

This is a super-detailed post about building a textarea with feedback about how much you’ve written in it, relative to the defined maxlength. There are some great recommendations here, especially:

• Start with sensible helper text;
• Use JavaScript to smartly enhance the experience; and
• Always provide a non-JavaScript fallback.

I gave similar advice in a piece I wrote for Web Standards Sherpa a few years back.

One thing I’ve always felt was missing from the W3C process was input from real web designers and developers. Sure, a handful of us have been tapped to join specific workign groups as “invited experts,” but they have been few and far between. And few designers and developers have the piles of cash laying about that are required to join the W3C.

Over the years, I’d hoped to see an organization like the World Organization of Webmasters or the Web Standards Project—both of which I helped steer in some capacity­ at varying times—would step up an fill this suprising gap, but alas that never happened. And so I am so thankful to see the Fronteers folks (a web design community in the Netherlands) considering formally joining the W3C to fill this role. And I’m even more excited that Rachel Andrew is their first choice to act on our behalf.

I’m hopeful the Fronteers community will vote in favor of this so we can get a few of our own advocates on key committees.

I’m always learning new things about ARIA and in this article I learned about some ways to ensure that radio controls marked up in legend and fieldset are actually exposed as a radio group. Now to figure out a way to get automatic role assignment…

Related: The state of fieldset interoperability

Accessibility isn’t a checklist and testing for it can’t be left to machines. Here’s a deep dive from Eric Bailey on the limitations of automated testing and how to test better.

You can always count on Tim Kadlec to deep dive into the implications of new performance features and Chrome’s decision to remove block JavaScript in certain scenarios is no exception.

Long story short, the NOSCRIPT intervention looks like a really great feature for users. More often than not it provides significant reduction in data usage, not to mention the reduction in CPU time—no small thing for the many, many people running affordable, low-powered devices.