What can we conclude from the State of HTML 2023 survey results?

The State of HTML 2023 cyberpunk logo. The word 'HTML' is the center. In the background, it has 6 diagonal blades with a blueish gradient with a wrap-around ribbon with the words 'state of' on one side and '2023' on the other side.

The State of HTML survey results have come in! The results highlight the desire for HTML to evolve. It points out that some of us would benefit from knowing HTML better and education is probably a factor!

Developers would like to see more interactive HTML elements, all the top missing elements are interactive. Form elements were the elements most frequently recreated – styling form elements is problematic. This concurs with the general sentiments I expressed last October.

The most common usage of HTML is to build Web apps. People are interested in building native-like web apps but are being held back by unresolved PWA/platform issues. It is well documented that Apple and Google have hard stances on PWAs that make it difficult for PWAs to become a viable alternative to native apps. The EU has been pushing them on this issue lately but Apple pushed back. At least Apple reversed their decision to block PWAs on iOS in the EU in March.

We have seen a few HTML improvements recently with the <dialog> element and the Popover API being widely adopted by browsers. The biggest bottleneck is with the standards bodies, the new elements that people are craving need to be standardized. Augmenting some existing elements is possible too, but there are limitations because we can’t change elements in a way that breaks the web. One element recently augmented was <details>, which can be used as an exclusive accordion through the name attribute. The priorties and dynamics behind moving standards forward is hard to understand from the outside.

The Web Components suite of APIs (custom elements, HTML templates, shadow DOM) had the potential to relief the bottleneck with the standard bodies, to enable developers to make their own custom elements. However, there were setbacks and false dawns. It looks like Web Components are in a better place now and are being used more widely. Although Web Components do appear to still be falling short of developer needs with a range of pain points listed for consuming them and a range of pain points for producing web components. They topped the charts of worst experiences across the survey. There is more work to be done.

There are still a lot of pain points around interactivity. There are some proposals like the InvokeEvent API (Invokers) that would solve some of these issues. I would love to see them come to fruition. Declarative interactivity is a major weakness of the web platform still. Frameworks have taught us that developers love declarative powers.

Thanks to all the folks who contributed to the survey to make the issues more visible! Let’s hope it is a motivating force for change.