week11y issue 95
Your weekly frequent11y newsletter, brought to you by @ChrisBAshton:
- This got shared around the work chat recently. A blog post by Ălvaro Montoro that busts 10 myths about accessibility:
- Accessibility is difficult – “Accessibility is not difficult. Do you know what’s difficult? Running at an Olympic level. A Web Developer can learn at least the basics of Web Accessibility within hours”.
- Accessibility is expensive – “Something that they could have avoided if they had implemented accessibility at the start”.
- Accessible websites are ugly – “Nothing could be further from the truth”.
- Accessibility is for blind people/screen readers – “The people who require Web Accessibility are not a homogeneous group”.
- Accessibility is for people with disabilities – “There are invisible and situation disabilities that impair people and limit what they can do temporarily (or even permanently)”.
- Automatic tests are enough for accessibility – “automatic tests only detect ~30% of the issues.”.
- Accessibility overlays are enough to ensure Web Accessibility – “The consensus in the Accessibility community is almost unanimous: overlays don’t work”.
- HTML is accessible by default – “elements like
<video>
or<audio>
have controls that are not fully keyboard-accessible and that differ considerably from browser to browser and cause frustration. ‘Mostly accessible’ looks much better.” - No ARIA > Bad ARIA – the equation should be “Good ARIA > No ARIA > Bad ARIA”. Otherwise we “perpetuate a false dichotomy. We can learn ARIA, practice ARIA, improve ARIA”.
- Prefers reduced motion means no motion – “We don’t need to cancel every motion on a website. Instead, we need to think through them, check what’s appropriate and not”.
Have Single-Page Apps Ruined the Web? (video, 19m)
- Talk by Rich Harris, creator of Rollup and Svelte, for Jamstack Conf.
- He explains what a Single Page App (SPA) is, then starts off critical. “SPAs are terrible because”:
- Bloated JS framework, complex tooling, poor performance.
- It will be buggy. Expectations of the web are broken, e.g. CMD + Click should open links in new tabs, but often won’t work in SPAs.
- Accessibility issues: focus management, scroll management, navigation announcements, etc.
- Less resilience. Rich shares a webpage he often points people to when they assume that everyone has JavaScript. It is a long flowchart of all of the reasons why JS may not load for someone. Rich also points out that these users are underrepresented in analytics, because analytics typically require JS to run.
- But then he highlights the benefits:
- SPAs allow you to do things you can’t do ‘natively’, e.g. transitions between pages.
- Only need to load third party scripts once, whereas multi-page apps (i.e. the traditional server-rendered sites) have to at least evaluate the JS on each page load, even if the scripts themselves are already cached.
- Single codebase of JS (rather than 1 for client, and 1 for server, which may be, for example, PHP).
- Rich proposes a new direction for the web, and coins the term “transitional apps” (from “transitional design” in the interior design industry), which are apps that take the best of both worlds. There is a lot happening in this space, including:
- Hotwire
- React Server Components
- Marko partial hydration
- qwik aggressive lazy loading
- astro islands architecture
- …and SvelteKit, which Rich gives an impressive demo on.
Stark, a hub for accessible software design, launches a Mac app in beta
- Stark is a suite of tools for the world’s most popular design apps. “Companies can upload their design files, which then identifies accessibility issues and suggests changes.”
- It “offers checks and suggestions to make sure that visual materials meet accessibility standards for visually impaired people”. More than 500,000 people have used Stark’s integrated plug-ins for apps like Adobe XD, Figma, Sketch and Google Chrome.
- Stark have now released Stark for Mac, a standalone app that works with Sketch and will soon work with Figma and Adobe XD. It allows you to detect accessibility issues at scale, rather than on a per-file basis as is the case with the plugins.
- Stark has a limited free plan, after which it costs $60 per year for the current suite of tools. The private beta version of Stark for Mac is currently free, with pricing yet to be announced.
Did you know that you can subscribe to dai11y, week11y, fortnight11y or month11y updates! Every newsletter gets the same content; it is your choice to have short, regular emails or longer, less frequent ones. Curated with â„ by developer @ChrisBAshton.