Garrett talks through some handy HTML attributes: spellcheck, autofocus, autocapitalize, autocomplete, and autocorrect:
While they feel like small details, when we set these attributes on inputs, we streamline things for visitors while also guiding the browser on when it should just get out of the way.
Here’s a nice HTML web component that uses structured data in the markup to populate a Leaflet map.
Personally I’d probably use microformats rather than microdata, but the princple is the same: progressive enhancement from plain old HTML to an interactive map.
This looks like a handy collection of HTML web components for common interface patterns.
drab does not use the shadow DOM, so you can style content within these elements as usual with CSS.
Scott has written a perfect description of HTML web components:
They are custom elements that
- are not empty, and instead contain functional HTML from the start,
- receive some amount of progressive enhancement using the Web Components JavaScript lifecycle, and
- do not rely on that JavaScript to run for their basic content or functionality.
The main reason I’m so hot on Light DOM is that I find the styling story of Web Components using Shadow DOM annoying.
Here’s a good walkthrough of adding microformats to your site, starting with h-card and moving on to h-entry.
This is an interesting idea from Scott—a templating language that doesn’t just replace variables with values, but keeps the original variable names in there too.
Not sure how I feel about using data- attributes for this though; as far as I know, they’re intended to be site-specific, not for cross-site solutions like this.
Huh! I did not know this. Good to know!
This is a really interesting proposal, and I have thoughts.
Remember when I wrote about sizes="auto"? Well, it’s coming to Chrome! Hallelujah!
This is an excellent step-by-step walkthrough by Tess of creating a web component, with real thought given to what should be in the HTML (which will act as a fallback) and what’s better generated in the Shadow DOM (like buttons for interactivity).
This perfectly mirrors something Chris was saying in a recent episode of the Shop Talk Show:
I think of the image comparison one. That’s a classic example in Web component. What’s inside is just two IMG tags. That’s it. When it fails, you don’t want a weird div with little arrows on it being rendered on the page. That’s not doing anything because it has failed to load the JavaScript.
You just take some normal HTML markup, wrap it with a custom element, and then write some JS to add capabilities which you can then style with regular CSS! Everything’s of the Light Side of the Web. No need to pierce the Vale of Shadows or whatever.
I think Eric’s approach here should be the default for most web components: you probably don’t need to mess around with the shadow DOM, and you should definitely be wrapping your web component around existing HTML instead of witing opening and closing tags with nothing in between.
Augment, don’t replace.
Vasilis gives the gist of his excellent talk at the border:none event that just wrapped up in Nuremberg. The rant at the end chimed very much with my feelings on this topic:
I showed a little interaction experiment that one of my students made, with incredible attention to detail. Absolutely brilliant in so many ways. You would expect that all design agencies would be fighting to get someone like that into their design team. But to my amazement she now works as a react native developer.
I have more of these very talented, very creative designers who know how to code, who really understand how the web works, who can actually design things for the web, with the web as a medium, who understand the invisible details, who know about the UX of HTML, who know what’s possible with modern HTML and CSS. Yet when they start working they have to choose: you either join our design team and are forced to use a tool that doesn’t get it, or you join the development team and are forced to use a ridiculous framework and make crap.
The slides from Hidde’s presentation at Paris Web—a great overview of using and misusing ARIA.
Now, this is how you design a web component. It’s a progressive enhancement.
Wrap your existing table element inside table-saw and it will behave responsively. If anything goes wrong with the JavaScript, the fallback is the regular table that’s already in your markup.
I just wish the installation didn’t assume that you’re using npm …it’s not really “zero dependency” if it depends on that.
Bruce raises an interesting question with media playing in popovers—shouldn’t the media pause when the popover is closed? I agree with Bruce that this is a common use case that should be covered declaratively.
Gosh! And I thought I had strong opinions about markup!
A great reminder of just how much you can do with modern markup and styles when it comes to form validation. The :user-invalid and :user-valid pseudo-classes are particularly handy!
I enjoyed this self-documenting journey of exploration.
This is a handy guideline to remember, even if there exceptions:
When a keyboard user follows a link, their focus should be taken to the new place; when a keyboard user presses a button, focus should remain on that button.