Julie Grundy kicks off this, our fifteenth year, by diving headlong into the snowy issue of customising form inputs. Nothing makes a more special gift at Christmas that something you’ve designed and customised yourself. But can it be done while staying accessible to every user?

In my work as an accessibility consultant, there are some frequent problems I find on people’s websites. One that’s come up a lot recently is that people are making custom select inputs for their forms. I can tell that people are trying to make them accessible, because they’ve added ARIA attributes or visually-hidden instructions for screen reader users. Sometimes they use a plugin which claims to be accessible. And this is great, I love that folks want to do the right thing! But so far I’ve never come across a custom select input which actually meets all of the WCAG AA criteria.

Often I recommend to people that they use the native HTML select element instead. Yes, they’re super ugly, but as Scott Jehl shows us in his article Styling a Select Like It’s 2019 they are a lot easier to style than they used to be. They come with a lot of accessibility for free – they’re recognised and announced clearly by all screen reader software, they work reliably and predictably with keyboards and touch, and they look good in high contrast themes.

But sometimes, I can’t recommend the select input as a replacement. We want a way for someone to choose an item from a list of options, but it’s more complicated than just that. We want autocomplete options. We want to put images in there, not just text. The optgroup element is ugly, hard to style, and not announced by screen readers. The focus styles are low contrast. I had high hopes for the datalist element, but although it works well with screen readers, it’s no good for people with low vision who zoom or use high contrast themes.

Select inputs are limited in a lot of ways. They’re frustrating to work with when you have something which looks almost like what you want, but is too restricted to be useful. We know we can do better, so we make our own.

Let’s work out how to do that while keeping all the accessibility features of the original.

Semantic HTML

We’ll start with a solid, semantic HTML base. A select input is essentially a text input which restricts the possible answers, so let’s make a standard input.

<label for="custom-select">User Type</label>
<input type="text" id="custom-select">

Then we need to show everyone who can see that there are options available, so let’s add an image with an arrow, like the native element.

<label for="custom-select">User Type</label>
<input type="text" id="custom-select">
<img src="arrow-down.svg" alt="">

For this input, we’re going to use ARIA attributes to represent the information in the icon, so we’ll give it an empty alt attribute so screen readers don’t announce its filename.

Finally, we want a list of options. An unordered list element is a sensible choice here. It also lets screen reader software understand that these bits of text are related to each other as part of a group.

<ul class="custom-select-options">
  <li>User</li>
  <li>Author</li>
  <li>Editor</li>
  <li>Manager</li>
  <li>Administrator</li>
</ul>

You can dynamically add or remove options from this list whenever you need to. And, unlike our <option> element inside a <select>, we can add whatever we like inside the list item. So if you need images to distinguish between lots of very similar-named objects, or to add supplementary details, you can go right ahead. I’m going to add some extra text to mine, to help explain the differences between the choices.

This is a good base to begin with. But it looks nothing like a select input! We want to make sure our sighted users get something they’re familiar with and know how to use already.

Styling with CSS

I’ll add some basic styles similar to what’s in Scott Jehl’s article above.

We also need to make sure that people who customise their colours in high contrast modes can still tell what they’re looking at. After checking it in the default Windows high contrast theme, I’ve decided to add a left-hand border to the focus and hover styles, to make sure it’s clear which item is about to be chosen.

This would be a good time to add any dark-mode styles if that’s your jam. People who get migraines from bright screens will thank you!

JavaScript for behaviour

Of course, our custom select doesn’t actually do anything yet. We have a few tasks for it: to toggle the options list open and closed when we click the input, to filter the options when people type in the input, and for selecting an option to add it to the input and close the list. I’m going to tackle toggling first because it’s the easiest.

Toggling

Sometimes folks use opacity or height to hide content on screen, but that’s like using Harry Potter’s invisibility cloak. No-one can see what’s under there, but Harry doesn’t cease to exist and you can still poke him with a wand. In our case, screen reader and keyboard users can still reach an invisible list.

Instead of making the content see-through or smaller, I’m going to use display: none to hide the list. display: none removes the content from the accessibility tree, so it can’t be accessed by any user, not just people who can see. I always have a pair of utility classes for hiding things, as follows:

.hidden-all {
  display: none;
}

.hidden-visually {
    position: absolute;
    width: 1px;
    height: 1px;
    padding: 0;
    overflow: hidden;
    clip: rect(0,0,0,0);
    white-space: nowrap;
    -webkit-clip-path: inset(50%);
    clip-path: inset(50%);
    border: 0;
} 

So now I can just toggle the CSS class .hidden-all on my list whenever I like.

Browsing the options

Opening up our list works well for our mouse and touch-screen users. Our styles give a nice big tap target for touch, and mouse users can click wherever they like.

We need to make sure our keyboard users are taken care of though. Some of our sighted users will be relying on the keyboard if they have mobility or dexterity issues. Usually our screen reader users are in Browse mode, which lets them click the arrow keys to navigate through content. However, custom selects are usually inside form elements. which pushes screen reader software to Forms Mode. In Forms mode, the screen reader software can only reach focusable items when the user clicks the Tab key, unless we provide an alternative. Our list items are not focusable by default, so let’s work on that alternative.

To do this, I’m adding a tabindex of -1 to each list item. This way I can send focus to them with JavaScript, but they won’t be part of the normal keyboard focus path of the page.

csOptions.forEach(function(option) {
    option.setAttribute('tabindex, '-1')
}) 

Now I can move the focus using the Up and Down arrow keys, as well as with a mouse or tapping the screen. The activeElement property of the document is a way of finding where the keyboard focus is at the moment. I can use that to loop through the elements in the list and move the focus point forward or back, depending on which key is pressed.

function doKeyAction(whichKey) {
  const focusPoint = document.activeElement
  switch(whichKey) {
    case: 'ArrowDown':
      toggleList('Open')
      moveFocus(focusPoint, 'forward')
      break
    case: 'ArrowUp':
      toggleList('Open')
      moveFocus(focusPoint, 'back')
      break
  }
}

Selecting

The Enter key is traditional for activating an element, and we want to match the original select input.

We add another case to the keypress detector…

case 'Enter':
  makeChoice(focusPoint)
  toggleList('Shut')
  setState('closed')
  break 

… then make a function which grabs the currently focused item and puts it in our text input. Then we can close the list and move focus up to the input as well.

function makeChoice(whichOption) {
    const optionText = whichOption.documentQuerySelector('strong')
    csInput.value = optionText
}

Filtering

Standard select inputs have keyboard shortcuts – typing a letter will send focus to the first item in the option which begins with that letter. If you type the letter again, focus will move to the next option beginning with that letter.

This is useful, but there’s no clue to tell users how many options might be in this category, so they have to experiment to find out. We can make an improvement for our users by filtering to just the set of options which matches that letter or sequence of letters. Then sighted users can see exactly how many options they’ve got, and continue filtering by typing more if they like. (Our screen reader users can’t see the remaining options while they’re typing, but don’t worry – we’ll have a solution for them in the next section).

I’m going to use the .filter method to make a new array which only has the items which match the text value of the input. There are different ways you could do this part – my goal was to avoid having to use regex, but you should choose whatever method works best for your content.

function doFilter() {
  const terms = csInput.value
  const aFilteredOptions = aOptions.filter(option => {
    if (option.innerText.toUpperCase().startsWith(terms.toUpperCase())) {
    return true
    }
  })
  // hide all options
  csOptions.forEach(option => option.style.display = "none")
  // re-show the options which match our terms
  aFilteredOptions.forEach(function(option) {
    option.style.display = ""
  })
} 

Nice! This is now looking and behaving really well. We’ve got one more problem though – for a screen reader user, this is a jumble of information. What’s being reported to the browser’s accessibility API is that there’s an input followed by some clickable text. Are they related? Who knows! What happens if we start typing, or click one of the clicky text things? It’s a mystery when you can’t see what’s happening. But we can fix that.

ARIA

ARIA attributes don’t provide much in the way of additional features. Adding an aria-expanded='true' attribute doesn’t actually make anything expand. What ARIA does is provide information about what’s happening to the accessibility API, which can then pass it on to any assistive technology which asks for it.

The WCAG requirements tell us that when we’re making custom elements, we need to make sure that as a whole, the widget tells us its name, its role, and its current value. Both Chrome and Firefox reveal the accessibility tree in their dev tools, so you can check how any of your widgets will be reported.

We already have a name for our input – it comes from the label we associated to the text input right at the start. We don’t need to name every other part of the field, as that makes it seem like more than one input is present. We also don’t need to add a value, because when we select an item from the list, it’s added to the text input and therefore is exposed to the API.

But our screen readers are going to announce this custom select widget as a text entry field, with some images and a list nearby.

The ARIA Authoring Practices site has a pattern for comboboxes with listboxes attached. It tells you all the ARIA you need to make screen reader software give a useful description of our custom widget.

I’m going to add all this ARIA via JavaScript, instead of putting it in the HTML. If my JavaScript doesn’t work for any reason, the input can still be a plain text field, and we don’t want screen readers to announce it as anything fancier than that.

csSelector.setAttribute('role', 'combobox') 
csSelector.setAttribute('aria-haspopup', 'listbox')
csSelector.setAttribute('aria-owns', '#list') 
csInput.setAttribute('aria-autocomplete', 'both')
csInput.setAttribute('aria-controls', 'list')

The next thing to do is let blind users know if the list is opened or closed. For that task I’m going to add an aria-expanded attribute to the group, and update it from false to true whenever the list changes state in our toggling function.

The final touch is to add a secret status message to the widget. We can use it to update the number of options available after we’ve filtered them by typing into the input. When there are a lot of options to choose from, this helps people who can’t see the list reducing know if they’re on the right track or not.

To do that we first have to give the status message a home in our HTML.

<div id='custom-select-status' class='hidden-visually' aria-live='polite'></div>

I’m using our visually-hidden style so that only screen readers will find it. I’m using aria-live so that it will be announced as often as it updates, not just when a screen reader user navigates past it. Live regions need to be present at page load, but we won’t have anything to say about the custom select then so we can leave it empty for now.

Next we add one line to our filtering function, to find the length of our current list.

updateStatus(aFilteredOptions.length)

Then we send that to a function which will update our live region.

function updateStatus(howMany) {
    console.log('updating status')
    csStatus.textContent = howMany + " options available."
}

Conclusion

Let’s review what we’ve done to make an awesome custom select input:

• Used semantic HTML so that it’s easily interpreted by assistive technology while expanding the types of content we can include in it
• Added CSS styles which are robust enough to survive different visual environments while also fitting into our branding needs
• Used JavaScript to provide the basic functionality that the native element has
• Added more JavaScript to get useful functionality that the native element lacks
• Carefully added ARIA attributes to make sure that the purpose and results of using the element are available to assistive technology and are updated as the user interacts with it.

You can check out my custom select pattern on GitHub – I’ll be making additions as I test it on more assistive technology, and I welcome suggestions for improvements.

The ARIA pattern linked above has a variety of examples and customisations. I hope stepping through this example shows you why each of the requirements exists, and how you can make them fit your own needs.

I think the volume of custom select inputs out there shows the ways in which the native select input is insufficient for modern websites. You’ll be pleased to know that Greg Whitworth and Simon Pieters are working on improving several input types! You can let them know what features you’d like selects to have. But until that work pays off, let’s make our custom selects as accessible and robust as they can possibly be.

About the author

Julie Grundy is an accessibility expert who works for Intopia, a digital accessibility consultancy. She has over 15 years experience as a front-end web developer in the health and education sectors. She believes in the democratic web and aims to unlock digital worlds for as many people as possible. In her spare time, she knits very slowly and chases very quickly after her two whippets.

More articles by Julie

Amy Kapernick sings us through numerous ways of improving the robustness and reliability of our front end code with a comprehensive rundown of ideas, tools, and resources. The girls and boys won’t get any toys until all the tests are passing.

Anyone who’s spoken to me at some point in November may get the impression that I’m a bit of a grinch. But don’t get me wrong, I love Christmas, I love decorating my tree, singing carols, and doing Christmas cooking - in December. So for me to willingly be humming the 12 days of Christmas in October, it’s probably for something that I think is even more important than banning premature Christmas decorations, like front end testing.

On the 12th day of Christmas, my front end dev, she gave to me, 12 testing tools, 11 optimised images, 10 linting rules, 9 semantic headings, 8 types of colour blindness, 7(.0) contrast ratio, 6 front end tests, 5 browser types, 4 types of tests, 3 shaken trees, 2 image types, and a source controlled deployment pipeline.

Twelve Testing Tools

1. axe does automated accessibility testing. Run as part of your development build, it outputs warnings to your console to let you know what changes you need to make (referencing accessibility guides). You can also specify particular accessibility standard levels that you’d like to test against, eg. best-practice, wcag2a or wcag2aa, or you can pick and choose individual rules that you want to check for (full list of rules you can test with axe).

2. BackstopJS runs visual regression tests on your website. Run separately, or as part of your deployment/PR process, you can use it to make sure your code changes aren’t bleeding into other areas of the website. By default, BackstopJS will set you up with a bunch of configuration options by running backstop init in your project to help get you started.

3. Website Speed Test analyses the performance of your website specifically with respect to images, and the potential size savings if they were optimised.

4. Calibre runs several different types of tests by leveraging Lighthouse. You can run it over your live website through their web app or through the command line, it then monitors your website for performance and accessibility over time, providing metrics and notifications of any changes.

5. Cypress is for end-to-end testing of your website. When visual regression testing may be a bit much for you, Cypress can help you test and make sure elements are still on the page and visible (even if they’re not pixel for pixel where they were last time).
6. pa11y is for automated accessibility testing. Run as part of your build process or using their CLI or dashboard, it tests your website against various Web Content Accessibility Guidelines (WCAG) criteria (including visual tests like colour contrast). While axe is run as part of your dev build and gives you an output to the console, it can be combined with pa11y to automate any changes as part of your build process.
7. whocanuse was created by Corey Ginnivan, and it allows you to view colour combinations as those with colour blindness would (as well as testing other visual deficiencies, and situational vision events), and test the colour contrast ratio based on those colours.

8. Colour Blindness Emulation was created by Kyo Nagashima as an SVG filter to emulate the different types of colour blindness, or if you’re using Gatsby, you can use a plugin based off of gatsby-plugin-colorblind-filters.
9. Accessible Brand Colors tests all your branding colours against each other (this is great to show designers what combinations they can safely use).

10. Browser dev tools - Most of the modern browsers have been working hard on the features available in their dev tools:
    • Firefox: Accessibility Inspector, Contrast Ratio testing, Performance monitoring.
    • Chromium: (Chrome, Edge Beta, Brave, Vivaldi, Opera, etc) - Accessibility Inspector, Contrast Ratio testing, Performance Monitoring, Lighthouse Audits (testing performance, best practices, accessibility and more).
    • Edge: Accessibility Inspector, Performance monitoring.
    • Safari: Accessibility Inspector, Performance monitoring.

11. Linc is a continuous delivery platform that makes testing the front end easier by automatically deploying a version of your website for every commit on every branch. One of the biggest hurdles when testing the front end is needing a live version of the site to view and test against. Linc makes sure you always have one.
12. ESLint and Stylelint check your code for programmatic and stylistic errors, as well as helping keep formatting standard on projects with multiple developers. Adding a linter to your project not only helps you write better code, it can reduce simple errors that might be found during testing time. If you’re not writing JavaScript, there are plenty of alternatives for whatever language you’re writing in.

If you’re trying to run eslint in VS Code, make sure you don’t have the Beautify extension installed, as that will break things.

Eleven Optimised Images

When it comes to performance, images are where we take the biggest hit, with images accounting for over 50% of total transfer size for websites. Many websites are serving excessively large images “just in case”, but there’s actually a native HTML element that allows us to serve different image sizes based on the screen size or serve better image formats when the browser supports it (or both).

<!-- Serving different images based on the width of the screen -->
<picture>
    <source
        srcset="/img/banner_desktop.jpg"
        media="(min-width: 1200px)"
    />
    <source
        srcset="/img/banner_tablet.jpg"
        media="(min-width: 700px)"
    />
    <source
        srcset="/img/banner_mobile.jpg"
        media="(min-width: 300px)"
    />
    <img src="/img/banner_fallback.jpg">
</picture>

<!-- Serving different image formats based on browser compatibility -->
<picture>
    <source
        srcset="/banner.webp"
        type="image/webp"
    />
    <img src="/img/banner_fallback.jpg">
</picture>

Ten Linting Rules

A year ago, I didn’t use linting. It was mostly just me working on projects, and I can code properly right? But these days it’s one of the first things I add to a project as it saves me so much time (and has taught me a few things about JavaScript). Linting is a very personal choice, but there are plenty of customisations to make sure it’s doing what you want, and it’s available in a wide variety of languages (including linting for styling).

// .eslintrc
module.exports = {
    rules: {
        'no-var': 'error',
        'no-unused-vars': 1,
        'arrow-spacing': ['error', { before: true, after: true }],
        indent: ['error', 'tab'],
        'comma-dangle': ['error', 'always'],
        // standard plugin - options
        'standard/object-curly-even-spacing': ['error', 'either'],
        'standard/array-bracket-even-spacing': ['error', 'either'], },
}

// .stylelintrc
{
    "rules": {
        "color-no-invalid-hex": true,
        "indentation": [
            "tab",
            {
                "except": [
                    "value"
                ]
            }
        ],
        "max-empty-lines": 2,
    }
}

Nine Semantic Headings

No, I’m not saying you should use 9 levels of headings, but your webpage should have an appropriate number of semantic headings. When your users are accessing your webpage with a screen reader, they rely on landmarks like headings to tell them about the page. Similarly to how we would scan a page visually, screen readers give users a list of all headings on a page to allow them to scan through the sections and access the information faster.

When there aren’t any headings on a page (or headings are being used for their formatting rather than their semantic meaning), it makes it more difficult for anyone using a screen reader to understand and navigate the page. Make sure that you don’t skip heading levels on your page, and remember, you can always change the formatting on a p tag if you need to have something that looks like a heading but isn’t one.

<h1>Heading 1 - Page Title</h2>
<p>Traditionally you'll only see one h1 per page as it's the main page title</p>
<h2>Heading 2</h2>
<p>h2 helps to define other sections within the page. h2 must follow h1, but you can also have h2 following another h2.</p>
<h3>Heading 3</h3>
<p>h3 is a sub-section of h2 and follows similar rules to h2. You can have a h3 after h3, but you can't go from h1 to h3.</p>
<h4>Heading 4</h4>
<p>h4 is a sub-section of h3. You get the pattern?</p>

Eight Types of Colour Blindness

Testing colour contrast may not always be enough, as everyone perceives colour differently. Take the below colour combination (ignoring the fact that it doesn’t actually look nice). It has decent colour contrast and would meet the WCAG colour contrast requirements for AA standards – but what if one of your users was red-green colour blind? Would they be able to tell the difference?

Red-green colour blindness is the most common form of colour blindness, but there are 8 different types affecting different parts of the colour spectrum, all the way up to complete colour blindness.

Protanopia

Inability to see red end of the colour spectrum.

Protanomaly

Difficulty seeing some shades of red.

Deuteranopia

Inability to see the green portion of the colour spectrum.

Deuteranomaly

Difficulty seeing some shades of green.

Tritanopia

Inability to see blue end of the colour spectrum.

Tritanomaly

Difficulty seeing some shades of blue.

Achromatopsia

Inability to see all parts of the colour spectrum, only able to perceive black, white and shades of grey.

Achromatomaly

Difficulty seeing all parts of the colour spectrum.

Seven (.0) Contrast Ratio

Sufficient colour contrast is perhaps one of the best steps to take for accessibility, as it benefits everyone. Having adequate contrast doesn’t just make the experience better for those with vision impairments, but it also helps those with situational impairments. Have you ever been in the sun and tried to read something on your screen? Whether you can view something when there’s glare could be as easy as making sure there’s enough contrast between the text and its background colour.

The WCAG have defined a contrast ratio of at least 4.5:1 for normal text (18.5px) and 3:1 for large text (24px) to meet AA accessibility standards, but this should be an absolute minimum and isn’t always readable. All four below examples have sufficient contrast to pass AA standards, but you might be hard pressed to read them when there’s glare or you have a dodgy monitor (even more so considering most websites use below 18.5px for their base font size).

To meet the AAA standard you need to have a ratio of 7:1 for normal text and 4.5:1 for large text, which should be sufficient for those with 20/80 vision to read.

Six Front End Tests

1. Adding default axe-core testing to Gatsby:

   //gatsby-config.js
   {
       resolve: 'gatsby-plugin-react-axe',
       options: {},
   },

2. Running pa11y tests on homepage at various screen sizes:

   // tests/basic-a11y_home.js
   const pa11y = require('pa11y'),
       fs = require('file-system')
   
   runTest()
   
   async function runTest() {
       try {
           const results = await Promise.all([
               pa11y('http://localhost:8000', {
                   standard: 'WCAG2AA',
                   actions: [],
                   screenCapture: `${__dirname}/results/basic-a11y_home_mobile.png`,
                   viewport: {
                       width: 320,
                       height: 480,
                       deviceScaleFactor: 2,
                       isMobile: true,
                   },
               }),
               pa11y('http://localhost:8000', {
                   standard: 'WCAG2AA',
                   actions: [],
                   screenCapture: `${__dirname}/results/basic-a11y_home_desktop.png`,
                   viewport: {
                       width: 1280,
                       height: 1024,
                       deviceScaleFactor: 1,
                       isMobile: false,
                   },
               }),
           ])
   
           fs.writeFile('tests/results/basic-a11y_home.json', JSON.stringify(results), err => {
               console.log(err)
           })
       } catch (err) {
           console.error(err.message)
       }
   }

3. Running pa11y tests on a blog post template at various screen sizes:

   // tests/basic-a11y_post.js
   const pa11y = require('pa11y'),
       fs = require('file-system')
   
   runTest()
   
   async function runTest() {
       try {
           const results = await Promise.all([
               pa11y('http://localhost:8000/template', {
                   standard: 'WCAG2AA',
                   actions: [],
                   screenCapture: `${__dirname}/results/basic-a11y_post_mobile.png`,
                   viewport: {
                       width: 320,
                       height: 480,
                       deviceScaleFactor: 2,
                       isMobile: true,
                   },
               }),
               pa11y('http://localhost:8000/template', {
                   standard: 'WCAG2AA',
                   actions: [],
                   screenCapture: `${__dirname}/results/basic-a11y_post_desktop.png`,
                   viewport: {
                       width: 1280,
                       height: 1024,
                       deviceScaleFactor: 1,
                       isMobile: false,
                   },
               }),
           ])
   
           fs.writeFile('tests/results/basic-a11y_post.json', JSON.stringify(results), err => {
               console.log(err)
           })
       } catch (err) {
           console.error(err.message)
       }
   }

4. Running BackstopJS on a homepage and blog post template at various screen sizes:

   // backstop.json
   {
     "id": "backstop_default",
     "viewports": [
       {
         "label": "phone",
         "width": 320,
         "height": 480
       },
       {
         "label": "tablet",
         "width": 1024,
         "height": 768
       },
       {
         "label": "desktop",
         "width": 1280,
         "height": 1024
       }
     ],
     "onBeforeScript": "puppet/onBefore.js",
     "onReadyScript": "puppet/onReady.js",
     "scenarios": [
       {
         "label": "Blog Homepage",
         "url": "http://localhost:8000",
         "delay": 2000,
         "postInteractionWait": 0,
         "expect": 0,
         "misMatchThreshold": 1,
         "requireSameDimensions": true
       },
       {
         "label": "Blog Post",
         "url": "http://localhost:8000/template",
         "delay": 2000,
         "postInteractionWait": 0,
         "expect": 0,
         "misMatchThreshold": 1,
         "requireSameDimensions": true
       }
     ],
     "paths": {
       "bitmaps_reference": "backstop_data/bitmaps_reference",
       "bitmaps_test": "backstop_data/bitmaps_test",
       "engine_scripts": "backstop_data/engine_scripts",
       "html_report": "backstop_data/html_report",
       "ci_report": "backstop_data/ci_report"
     },
     "report": [
       "browser"
     ],
     "engine": "puppeteer",
     "engineOptions": {
       "args": [
         "--no-sandbox"
       ]
     },
     "asyncCaptureLimit": 5,
     "asyncCompareLimit": 50,
     "debug": false,
     "debugWindow": false
   }

5. Running Cypress tests on the homepage:

   // cypress/integration/basic-test_home.js
   describe('Blog Homepage', () => {
       beforeEach(() => {
           cy.visit('http://localhost:8000')
       })
       it('contains "Amy Goes to Perth" in the title', () => {
           cy.title().should('contain', 'Amy Goes to Perth')
       })
       it('contains posts in feed', () => {
           cy.get('.article-feed').find('article')
       })
       it('all posts contain title', () => {
           cy.get('.article-feed')
               .find('article')
               .get('h2')
       })
   })

6. Running Cypress tests on a blog post template at various screen sizes:

   // cypress/integration/basic-test_post.js
   
   describe('Blog Post Template', () => {
       beforeEach(() => {
           cy.visit('http://localhost:8000/template')
       })
       it('contains "Amy Goes to Perth" in the title', () => {
           cy.title().should('contain', 'Amy Goes to Perth')
       })
       it('has visible post title', () => {
           cy.get('h1').should('be.visible')
       })
       it('has share icons', () => {
           cy.get('.share-icons a').should('be.visible')
       })
       it('has working share icons', () => {
           cy.get('.share-icons a').click({ multiple: true })
       })
       it('has a visible author profile image', () => {
           cy.get('.author img').should('be.visible')
       })
   })
   
   describe('Mobile Blog Post Template', () => {
       beforeEach(() => {
           cy.viewport('samsung-s10')
           cy.visit('http://localhost:8000/template')
       })
       it('contains "Amy Goes to Perth" in the title', () => {
           cy.title().should('contain', 'Amy Goes to Perth')
       })
       it('has visible post title', () => {
           cy.get('h1').should('be.visible')
       })
       it('has share icons', () => {
           cy.get('.share-icons .share-link').should('be.visible')
       })
       it('has a visible author profile image', () => {
           cy.get('.author img').should('be.visible')
       })
   })

Five Browser Types

Browser testing may be the bane of our existence, but it’s gotten easier, especially when you know the secret:

Not every browser needs to look the same.

Now, this may differ depending on your circumstances, but your website doesn’t have to match pixel for pixel across all browsers. As long as it’s on-brand and is useable across all browsers (this is where a good solid HTML foundation is useful), it’s ok for your site to look a little different between browsers.

While the browsers you test in will differ depending on your user base, the main ones you want to be covering are:

• Chrome/Chromium
• Firefox
• Safari
• Internet Explorer
• Edge

Make sure you’re testing these browsers on both desktop and mobile/tablet as well, sometimes their level of support or rendering engine will differ between devices – for example, iOS Chrome uses the Safari rendering engine, so something that works on Android Chrome may not work on iOS Chrome.

Four Types of Test

When it comes to testing the front end, there are a few different areas that we can cover:

1. Accessibility Testing: doing accessibility testing properly usually involves getting an expert to run through your website, but there are several automated tests that you can run against various standard levels.
2. Performance Testing: performance testing does technically bleed into the back end as well, but there are plenty of things that can be done from a front end perspective. Making sure the images are optimised, our code is clean and minified, and even optimising fonts using features like the font-display property. No amount of optimising the server and back end will matter if it takes forever for the front end to appear in a browser.
3. Visual Regression Testing: we’ve all been in the position where changing one line of CSS somewhere has affected another section of the website. Visual regression testing helps prevent that. By using a tool that compares before and after screenshots against one another to flag up what’s changed, you can be sure that style changes won’t bleed into unintended areas of the site.
4. Browser/device testing: while we all want our users to be running the most recent version of Chrome or Firefox, they may still be using the inbuilt browser on their DVD player – so we need to test various browsers, platforms and devices to make sure that our website can be accessed on whatever device they use.

Three Shaken Trees

Including (and therefore requiring your users to download) things that you’re not using affects the performance of your application. Are you forcing them to download the entire lodash library when you’re only using 2 functions? While a couple of unused lines of code may not seem like a huge performance hit, it can greatly affect users with slower devices or internet connections, as well as cluttering up your code with unused functions and dependencies. This can be set up on your bundler – Webpack and Parcel both have guides for tree shaking, and Gatsby has a plugin to enable it.

Two Image Types

While there are several different types of images, most of the time they fall into one of two categories:

Informative

The image represents/conveys important information that isn’t conveyed by the content surrounding it.

Decorative

The image only adds visual decoration to a page.

From these two categories, we can then determine if we need to provide alternative text for an image. If an image is purely decorative, then we add alt="" to let screen readers know that it’s not important. But if an image is informative, then we need to be supplying a text alternative that describes the picture for anyone who’s using a screen reader or isn’t able to see the image (remember the days when a standard internet connection took a long time to load a page and you saw alt text before an image loaded).

<img src="./nice-picture.jpg" alt="" />
<img src="./important-graphic.png" alt="This is a picture of something important to help add meaning to the text around me" />

If you have a lot of images with missing alt text, look into services that can auto-generate alt text based on image recognition services.

One Source Controlled Deployment Pipeline

While front end tests are harder to automate, running them through a source control and deployment pipeline helps track changes and eliminates issues where “it works on my computer”. Whether you’re running tests as part of the PR process, or simply against every commit that comes through, running tests automatically as part of your process makes every developer’s life easier and helps keep code quality at a high standard.

We already knew that testing was important, and your project can’t be run unless all your unit and integration tests are written (and pass), but often we forget about testing the front end. There are so many different tests we need to be running on the front end, it’s hard to work out what your need to test for and where to start.

Hopefully this has given you a bit of insight to front end testing, and some Christmas cheer to take you into the holidays.

About the author

Amy wears many hats as a freelance developer, business owner and conference addict. She regularly shares her knowledge with her peers and the next generation of developers by mentoring, coaching, teaching and feeding into the tech community in many ways.

Amy can be found volunteering her time with Fenders, ACS, SheCodes (formerly Perth Web Girls) and MusesJS (formerly NodeGirls). She also works as an evangelist for YOW! Conferences, is a Twilio Champion and has been nominated for the WiTWA awards for the last 2 years.

In her spare time Amy shares her knowledge and experience on her blogs and speaking at conferences. She has previously given keynotes at multiple events as well as speaking at several international conferences in the US and Europe.

More articles by Amy

Michelle Barker appears as one of a heavenly host, coming forth with scroll in hand to pronounce an end to janky scrolljacking! Unto us a new specification is born, in the city of TimBL, and its name shall be called Scroll Snap.

Sponsor: Order any Standard paperback(s) and get a surprise gift card in the box for YOU. While supplies last, from your pals at A Book Apart!

One area where the web has traditionally lagged behind native platforms is the perceived “slickness” of the app experience. In part, this perception comes from the way the UI responds to user interactions – including the act of scrolling through content.

Faced with the limitations of the web platform, developers frequently reach for JavaScript libraries and frameworks to alter the experience of scrolling a web page – sometimes called “scroll-jacking” – not always a good thing if implemented without due consideration of the user experience. More libraries can also lead to page bloat, and drag down a site’s performance. But with the relatively new CSS Scroll Snap specification, we have the ability to control the scrolling behaviour of a web page (to a degree) using web standards – without resorting to heavy libraries. Let’s take a look at how.

Scroll Snap

A user can control the scroll position of a web page in a number of ways, such as using a mouse, touch gesture or arrow keys. In contrast to a linear scrolling experience, where the rate of scroll reflects the rate of the controller, the Scroll Snap specification enables a web page to snap to specific points as the user scrolls. For this, we need a fixed-height element to act as the scroll container, and the direct children of that element will determine the snap points. To demonstrate this, here is some example HTML, which consists of a <div> containing four <section> elements:

<div class="scroll-container">
  <section>
    <h2>Section 1</h2>
  </section>
  <section>
    <h2>Section 2</h2>
  </section>
  <section>
    <h2>Section 3</h2>
  </section>
  <section>
    <h2>Section 4</h2>
  </section>
</div>

Scroll snapping requires the presence of two main CSS properties: scroll-snap-type and scroll-snap-align. scroll-snap-type applies to the scroll container element, and takes two keyword values. It tells the browser:

• The direction to snap
• Whether snapping is mandatory

scroll-snap-align is applied to the child elements – in this case our <section>s.

We also need to set a fixed height on the scroll container, and set the relevant overflow property to scroll.

.scroll-container {
  height: 100vh;
  overflow-y: scroll;
  scroll-snap-type: y mandatory;
}

section {
  height: 100vh;
  scroll-snap-align: center;
}

In the above example, I’m setting the direction in the scroll-snap-type property to y to specify vertical snapping. The second value specifies that snapping is mandatory. This means that when the user stops scrolling their scroll position will always snap to the nearest snap point. The alternative value is proximity, which determines that the user’s scroll position will be snapped only if they stop scrolling in the proximity of a snap point. (It’s down to the browser to determine what it considers to be the proximity threshold.)

If you have content of indeterminate length, which might feasibly be larger than the height of the scroll container (in this case 100vh), then using a value of mandatory can cause some content to be hidden above or below the visible area, so is not recommended. But if you know that your content will always fit within the viewport, then mandatory can produce a more consistent user experience.

In this example I’m setting both the scroll container and each of the sections to a height of 100vh, which affects the scroll experience of the entire web page. But scroll snapping can also be implemented on smaller components too. Setting scroll snapping on the x-axis (or inline axis) can produce something like a carousel effect.

In this demo, you can scroll horizontally scroll through the sections:

The Intersection Observer API

By implementing the CSS above, our web page already has a more native-like feel to it. To improve upon this further we could add some scroll-based transitions and animations. We’ll need to employ a bit of Javascript for this, using the Intersection Observer API. This allows us to create an observer that watches for elements intersecting with the viewport, triggering a callback function when this occurs. It is more efficient than libraries that rely on continuously listening for scroll events.

We can create an observer that watches for each of our scroll sections coming in and out of view:

const sections = [...document.querySelectorAll('section')]

const options = {
  rootMargin: '0px',
  threshold: 0.25
}

const callback = (entries) => {
  entries.forEach((entry) => {
    if (entry.intersectionRatio >= 0.25) {
      target.classList.add("is-visible");
    } else {
      target.classList.remove("is-visible");
    }
  })
}

const observer = new IntersectionObserver(callback, options)

sections.forEach((section, index) => {
  observer.observe(section)
})

In this example, a callback function is triggered whenever one of our sections intersects the container by 25% (using the threshold option). The callback adds a class of is-visible to the section if it is at least 25% in view when the intersection occurs (which will take effect when the element is coming into view), and removes it otherwise (when the element is moving out of view). Then we can add some CSS to transition in the content for each of those sections:

section .content {
  opacity: 0:
}

section.is-visible .content {
  opacity: 1;
  transition: opacity 1000ms:
}

This demo shows it in action:

You could, of course, implement some much more fancy transition and animation effects in CSS or JS!

As an aside, it’s worth pointing out that, in practice, we shouldn’t be setting opacity: 0 as the default without considering the experience if JavaScript fails to load. In this case, the user would see no content at all! There are different ways to handle this: We could add a .no-js class to the body (which we remove on load with JS), and set default styles on it, or we could set the initial style (before transition) with JS instead of CSS.

Position: sticky

There’s one more CSS property that I think has the potential to aid the scroll experience, and that’s the position property. Unlike position: fixed, which locks the position of an element relative to the nearest relative ancestor and doesn’t change, position: sticky is more like a temporary lock. An element with a position value of sticky will become fixed only until it reaches the threshold of its parent, at which point it resumes relative positioning.

By “sticking” some elements within scroll sections we can give the impression of them being tied to the action of scrolling between sections. It’s pretty cool that we can instruct an element to respond to it’s position within a container with CSS alone!

Browser support and fallbacks

The scroll-snap-type and scroll-snap-align properties are fairly well-supported. The former requires a prefix for Edge and IE, and older versions of Safari do not support axis values. In newer versions of Safari it works quite well. Intersection Observer similarly has a good level of support, with the exception of IE.

By wrapping our scroll-related code in a feature query we can provide a regular scrolling experience as a fallback for users of older browsers, where accessing the content is most important. Browsers that do not support scroll-snap-type with an axis value would simply scroll as normal.

@supports (scroll-snap-type: y mandatory) {
  .scroll-container {
    height: 100vh;
    overflow-y: scroll;
    scroll-snap-type: y mandatory;
  }

  section {
    height: 100vh;
    scroll-snap-align: center;
  }
}

The above code would exclude MS Edge and IE, as they don’t support axis values. If you wanted to support them you could do so using a vendor prefix, and using @supports (scroll-snap-type: mandatory) instead.

Putting it all together

This demo combines all three of the effects discussed in this article.

Summary

Spending time on scroll-based styling might seem silly or frivolous to some. But I believe it’s an important part of positioning the web as a viable alternative to native applications, keeping it open and accessible. While these new CSS features don’t offer all of the control we might expect with a fully featured JS library, they have a major advantage: simplicity and reliability. By utilising web standards where possible, we can have the best of both worlds: Slick and eye-catching sites that satisfy clients’ expectations, with the added benefit of better performance for users.

About the author

Michelle is a Lead Front End Developer at Bristol web agency Atomic Smash, author of front-end blog CSS { In Real Life }, and a Mozilla Tech Speaker. She has written articles for CSS Tricks, Smashing Magazine, and Web Designer Magazine, to name a few. She enjoys experimenting with new CSS features and helping others learn about them.

More articles by Michelle

Stuart Robson rolls up his sleeves and begins to piece together the jigsaw puzzle that is design tokens and component based design. Starting with the corners, and working around the edges, Stu helps us to piece together a full picture of a modern design system.

If you stare at your twitter feed long enough, it can look like everyone is talking about Design Systems. In some cases you could be persuaded to think how shallow the term can go.

“Isn’t this what we called Style Guides?”, “Here’s my React Design System”, “I’ve just updated the Design System in Sketch”

To me, they are some and all of these things. Over the last 4 years of consulting with two clients on their Design System, my own view has changed a little.

If you dig a little deeper into Design Systems twitter you will probably see the term “Design Tokens” pop up at least once a day somewhere. Design Tokens came out of work that was being done at Salesforce with Jina and others who pioneered the creation of Design Tokens as we know them today – creating the first command line tool in Theo that had started the adoption of Design Tokens to the wider Design Systems Community.

A cool term but, what are they?

If you look at your client work, your companies site, the project you’re working on you should notice some parts of the page have a degree of consistency: the background colour of your form buttons is the same colour as your link text, or your text has the same margin, or your card elements have the same spacing as your media object.

These are design decisions, and they should be littered across the overall design of your project. These decisions might start off in a Sketch file and make their way into code from detailed investigation of a Sketch file, or you may find that the design evolves from your design application once it gets into code.

These design decisions can change, and to keep them synchronised across design and development in applications, as well as a larger documentation site in your Design System, is going to take some effort.

This is where Design Tokens come in, and I find the best way to succinctly reiterate what they are is via the two following quotes…

“Design Tokens are an abstraction for everything impacting the visual design of an app/platform.”
– Sönke Rohde

…and

“We use them in place of hard-coded values in order to maintain a scale-able and consistent visual system.”
– Jina

There are several global design decisions that we can abstract to create a top level design token – Sizing, Font Families, Font Styles, Font Weights, Font Sizes, Line Heights, Border Styles, Border Colours, Border Radius, Horizontal Rule Colours, Background Colours, Gradients, Background Gradients, Box Shadows, Filters, Text Colours, Text Shadow, Time, Media Queries, Z Index, Icons – these can all be abstracted as required.

So, spicy Sass variables?

We can look at Design Tokens as an abstraction of CSS, sort of like Sass variables, but spicier. Looking at them like this we can see that they are (in either .yaml or .json) a group of related key value pairs with more information that can be added as needed.

The great thing with abstracting design decisions outside of your CSS pre-processor is that you’re not tying those decisions to one platform or codebase.

As a crude example, we can see here that we are defining a name and a value that could then become our color, background-color, or border-color, and more.

# Colours
# -------
- name: color-red
  value: #FF0000
- name: color-green
  value: #00FF00
- name: color-blue
  value: #0000FF
- name: color-white
  value: #FFFFFF
- name: color-black
  value: #000000

These can then generate our Sass variables (as an example) for our projects.

$color-red: #FF0000 !default;
$color-green: #00FF00 !default;
$color-blue: #0000FF !default;
$color-white: #FFFFFF !default;
$color-black: #000000 !default;

Why are they so good

Ok, so we now know what Design Tokens are, but why do we need them? What makes them better than our existing solutions (css pre-processors) for defining these design decisions?

I think there are 5 really good reasons why we all should start abstracting these design decisions away from the CSS that they may live in. Some of these reasons are similar to reasons some developers use a pre-processor like Sass, but with added bonuses.

Consistency

Much like using a CSS pre-processor or using CSS custom properties, being able to define a background colour, breakpoint, or font-size in more than one place using the same key ensures that we are using the Sass values across the entire product suite we are developing for.

Using our Design Tokens in their generated formats, we can be sure to not end up with 261 shades of blue.

Maintainability

By using a pre-processor like Sass, or using native CSS custom properties, we can already have maintainable code in our projects. Design Tokens also do this at the abstracted level as well.

Scalability

“Design Tokens enable us to scale our Design across all the permutations.”
– Jina

At this point, we’re only talking about abstracting the design decisions for use in CSS. Having Design Tokens allows design to scale for multiple brands or multiple projects as needed.

The main benefit of Design Tokens in regards to scalability is the option that it gives us to offer the Design Tokens for other platforms and frameworks as needed. With some of the tools available, we can even have these Tokens shared between applications used by designers and developers.

Your marketing site and your iOS application can soon share the same design decisions codified, and you can move towards creating an Android app or web application as required.

Documentation

If we abstract the design decisions from one platform specific programming language it would be no good if it wasn’t made to be easily accessible.

The tools and applications available that are mentioned later in this article can now create their own documentation, or allow you to create your own. This documentation is either hosted within a web-based application or can be self-hosted with the rest of your Design Systems documentation.

Most of the command line tools go further and allow you do add more details that you wish to convey in the documentation, making it as unique as it is required for your project.

Empowerment

When you abstract your design decisions to Design Tokens, you can help empower other people on the project. With the tools available today, and the tools that are just around the corner, we can have these design decisions determined by anyone on the team.

No-one necessarily needs to understand how to set up the codebase to update the colour slightly. Some of the tools I mention later on allow you to update the Design Tokens in the browser.

Design Systems are already “bridging the gap” between design and development. With Design Tokens and the tooling available, we can create better team relationships by closing that gap instead.

Some of the benefits of creating and using Design Tokens are the same as using a pre-processor when it comes to authoring CSS. I feel the added bonuses of being able to empower other team members and document how you use them, as well as the fundamental reasoning in that they can be platform agnostic, are all great “selling points” to why you need to start using Design Tokens today.

Tools

There are several tools available to help you and your team to create the required files from your abstracted Design Tokens:

Command Line Tools

There are several tools available on the command line that can be used as part of, or separate to, your development process.

These tools allow you to define the Design Tokens in a .json or .yaml file format which can then be compiled into the formats you require.

Some have built in functions to turn the inputted values to something different when compiled – for example, turning hexadecimal code that is a Design Token into a RGB value in your .css file. These command line tools, written in JavaScript, allow you to create your own ways in which you want things transformed.

My current client has certain design decisions for typography in long form content (font size, weight, line height and margins) which need to be together to make sense. Being able to write JavaScript to compile these design decisions into an independent Sass map for each element allows us to develop with assurance that the long form content has the correct styling.

WYSIWYG Tools

WYSIWYG (What You See Is What You Get Tools) have been around for almost as long as we have been able to make websites. I can just about remember using Dreamweaver 2, before I knew what a <table> was.

When browsers started to employ vendor prefixes to new CSS for their browsers, a flurry of online WYSIWYG tools came with it built in. They’re still there, but the industry has moved on.

Design Tokens also have a few WYSIWYG tools available. From simpler online tools that allow you to generate the correct Sass variables needed for your design decisions to tools that store your decisions online and allow you to export them as npm packages.

These types of tools for creating Design Tokens can help empower the team as a whole, with some automatically creating documentation which can easily be shared with a url.

Retrofitting Tools

If you are starting from scratch on a new re-design or on a new project that requires a Design System and Tokens, the many of the tools mentioned above will help you along your way. But what if you’re in the middle of a project, or you have to maintain something and want to start to create the parts required for a Design System?

Luckily there are several tools and techniques to help you make a start.

One new tool that might be useful is Superposition. Currently in private beta with the public release set for Q1 of 2020 Superposition helps you “Extract design tokens from websites and use them in code and in your design tool.”

Entering your domain gives you a nice visual documentation of your sites styles as Design Tokens. These can then be exported as Sass Variables, CSS Custom Properties, JavaScript with the team working on exports to iOS and Android.

If you have an existing site, this could be a good first step before moving to one of the other tools mentioned above.

You could also make use of CSSStats or Project Wallace’s Analysis page that I mentioned earlier. This would give you an indication of what Design Tokens you would need to implement.

Component Based Design

So, we’ve created our Design Tokens by abstracting the design decisions of brand colours, typography, spacing and more. Is that as far as we can go?

Levels of Design Decisions

Once we have created our first set of Design Tokens for our project, we can take it that little bit deeper. With command line tools and some of the applications available, you can link these more global decisions to a more deeper level.

For example, you can take your chosen colours and make further design decisions on the themes of your project, such as what the primary, secondary, or tertiary colours are or what your general component and layout spacing will be.

With this, we can go one step further. We could also define some component specific design decisions that can then be compiled for the developer to use. Invest in time to check over the designs with a fine toothcomb and make sure you are using the correct Sass variable or CSS custom property for that component.

If you are going more than one or two levels of design decision making, you can compile each set of these Design Tokens to your Sass variables which can then be used as required. So you can provide: global, theme, component level Sass variables which can be used in the project. Or you could choose to only compile what you need for the components you are creating.

Variables, Maps, Custom Properties

Some of the tools available for creating and maintaining your Design Tokens allow you to compile to certain programming languages.

With my current client work, I am making use of Sass variables, Sass maps, and CSS custom properties. The Design Tokens are compiled into one or more of these options depending on how they will be used.

Colours are compiled as global Sass variables, inside of a couple of Sass maps and CSS custom properties. Our macro layout breakpoints are defined as a single Sass map.

If we know we are creating a component that has the ability to be themed, we can make use of CSS custom properties to reduce the amount of CSS we need to override, and also allow us to inline things that can be changed via a CMS as required. Which leaves us using Sass variables for parts of a component that won’t change. We are using Sass maps differently still. As mentioned, we generate a Sass map containing the design decisions for each element of text, and we can use long form text. This Sass map is then compiled into separate CSS declarations as needed using Sass mixins.

I find the beauty of being able to make use of the global, themed, and component level design decisions by compiling them into various formats (that essentially become CSS) and that gives us more power in authoring components.

Creating Consistent Utility Classes

As you have created your more global generic design decisions, you can create your own small set of utility classes.

Using a pre-processor like Sass you can define a set of mixins and functions that can take your Design Tokens that have been compiled down into variables and maps and generate separate classes for each design decision.

By making tokens available to all digital teams, we can enable them to create custom experiences that are aligned to current visual standards when a component does not (or will not) exist in the design system. Maya King

In creating utility classes with Design Tokens (using something like Sass) you have consistency with the overall Design System for times when you or a team need to create a one-off component for a project.

These exceptions tend to be something that won’t make it as part of the overall Design System, but it still needs that look and feel.

Having classes available that we can guarantee use the generic, global design decisions from the Design Tokens means these one-off components should be well on their way to have the overall look and feel of the project, and will get any updates with little to no additional overhead.

Wrapping Up

I think we are starting to see the potential of using Design Tokens as Design Systems become even more popular. I think that, from this overview, we can see how they can help us close the gap that still exists in places between the designers and developers on the team. They can help empower people who do not code to make changes that can be automatically updating live work.

I think you can start now. You may not have or need what you could term “a fully-fledged Design System” but this small step will help move towards one in the future and give you instant benefits of consistency and maintainability now. If you want more Design Tokens, as well as the links that are dotted around this article I also maintain a GitHub repo of Awesome Design Tokens which I try to keep updated with links to tools, articles, examples, videos, and anything else that’s related to Design Tokens.

About the author

Stuart Robson is a freelance front-end developer and design systems advocate who curates the design systems newsletter - news.design.systems

More articles by Stuart

Rachel Andrew guides us through a tour of the last fifteen years in CSS layout, as manifested in articles here on 24 ways. From the days when Internet Explorer 6 was de rigueur, right up to the modern age of evergreen browsers, the only thing you can be sure of is that the web never stands still for long.

I’ve written nine articles in the 15 years of 24 ways, and all but one of those articles had something to do with CSS. In this last year of the project, I thought I would take a look back at those CSS articles. It’s been an interesting journey, and by reading through my words from the last 15 years I discovered not only how much the web platform has evolved - but how my own thinking has shifted with it.

2005: CSS layout starting points

Latest web browser versions: Internet Explorer 6 (at this point 4 years old), IE5.1 Mac, Netscape 8, Firefox 1.5, Safari 2

Fifteen years ago, my contributions to 24 ways started with a piece about CSS layout. That article explored something I had been using in my own work. In 2005, most of the work I was doing was building websites from Photoshop files delivered to me by my design agency clients. I’d built up a set of robust, tried-and-tested CSS layouts to use to implement these. My starting point when approaching any project was to take a look at the static comps and figure out which layout I would use:

• Liquid, multiple column with no footer
• Liquid, multiple column with footer
• Fixed width, centred

At that point, there were still many sites being shipped with table-based layouts. We had learned how to use floats to create columns some four years earlier, however layout was still a difficult and often fragile thing. By developing patterns that I knew worked, where I had figured out any strange bugs, I saved myself a lot of time.

Of course, I wasn’t the only person thinking in this way. The two sites from which the early CSS for layout enthusiasts took most of their inspiration, had a library of patterns for CSS layout. The Noodle Incident little boxes is still online, glish.com/css is sadly only available at the Internet Archive.

This thinking was taken to a much greater extreme in 2011, when Twitter Bootstrap launched and starting with an entire framework for layout and much more became commonplace across the industry. While I understand the concern many folk have about every website ending up looking the same, back in 2005 I was a pragmatist. That has not changed. I’ve always built websites and run businesses alongside evangelizing web standards and contributing to the platform. I’m all about getting the job done, paying the bills, balancing that with trying to make things better so we don’t need to make as many compromises in the future. If that means picking from one of a number of patterns, that is often a very reasonable approach. Not everything needs to be a creative outpouring.

Today however, CSS Grid Layout and Flexbox mean that we can take a much more fluid approach to developing layouts. This enables the practical and the creative alike. The need for layout starting points - whether simple like mine, or a full framework like Bootstrap - seems to be decreasing, however in their place comes an interest in component libraries. This approach to development partly enabled by the fact that new layout makes it possible to drop a component into the middle of a layout without blowing the whole thing up.

2006: Faster Development with CSS Constants

Latest web browser versions: Internet Explorer 7, Netscape 8.1, Firefox 2, Safari 2

My article in 2006 was once again taken from the work I was doing as a developer. I’ve always been as much, if not more of a backend developer than a frontend one. In 2006, I was working in PHP on custom CMS implementations. These would also usually include the front-end work. Along with several other people in the industry I’d been experimenting with ways to use CSS “constants” as we all seemed to call them, by processing the CSS with our server-side language of choice.

The use case was mostly for development, although as a CMS developer, I could see the potential of allowing these values to be updated via the CMS. Perhaps to allow a content editor to change a color scheme.

Also in 2006, the first version of Sass was released, created by Hampton Catlin and Natalie Weizenbaum. Sass, LESS and other pre-processors began to give us a more streamlined and elegant way to achieve variables in CSS.

In 2009, the need for pre-processors purely for variables is disappearing. CSS now has Custom Properties - something I did not foresee in 2006. These “CSS Variables” are far more powerful than swapping out a value in a build process. They can be changed dynamically, based on something changing in the environment, rather than being statically set at build time.

2009: Cleaner Code with CSS3 Selectors

Latest web browser versions: Internet Explorer 8, Firefox 3.5, Safari 4, Chrome 3

After a break from writing for 24 ways, in 2009 I wrote this piece about CSS3 Selectors, complete with jQuery fallbacks due to the fact that some of these selectors were not usable in Internet Explorer 8.

Today these useful selectors have wide browser support, we also have a large number of new selectors which are part of the Level 4 specification. The changes section of the Level 4 spec gives an excellent rundown of what has been added over the years. Browser support for these newer selectors is more inconsistent, MDN has an excellent list with the page for each selector detailing current browser support and usage examples.

2012: Giving Content Priority with CSS3 Grid Layout

Latest web browser versions: Internet Explorer 10, Firefox 17, Safari 6, Chrome 23

My 2012 piece was at the beginning of my interest in the CSS Grid Layout specification. Earlier in 2012 I had attended a workshop given by Bert Bos, in which he demonstrated some early stage CSS modules, including the CSS Grid Layout specification. I soon discovered that there would be an implementation of Grid in IE10, the new browser shipped in September of 2012 and I set about learning how to use Grid Layout. This article was based on what I had learned.

The problem of source versus visual order

As a CMS developer I immediately linked the ability to lay out items and prioritize content, to the CMS and content editors. I was keen to find ways to allow content editors to prioritize content across breakpoints, and I felt that Grid Layout might allow us to do that. As it turned out, we are still some way away from that goal. While Grid does allow us to separate visual display from source order, it can come at a cost. Non-visual browsers, and the tab order of the document follow the source and not the visual display. This makes it easy to create a disconnected and difficult to use experience if we essentially jumble up the display of elements, moving them away from how they appear in the document. I still think that an issue we need to solve is how to allow developers to indicate that the visual display should be considered the correct order rather than the document order.

The Grid Specification moved on

Some of the issues in this early version of the grid spec were apparent in my article. I needed to use a pre-processor, to calculate the columns an element would span. This was partly due to the fact that the early grid specifications did not have a concept of the gap property. In addition the initial spec did not include auto-placement and therefore each item had to be explicitly placed onto the grid. The basics of the final specification were there, however over the years that followed the specification was refined and developed. We got gaps, and auto-placement, and the grid-template-areas property was introduced. By the time Grid shipped in Firefox, Chrome, and Safari many of the sticky things I had encountered when writing this article were resolved.

2015: Grid, Flexbox, Box Alignment: Our New System for Layout

Latest web browser versions: Edge 13, Firefox 43, Safari 9, Chrome 47

Grid still hadn’t shipped in more browsers but the specification had moved on. We had support for gaps, with the grid-row-gap, grid-column-gap and grid-gap properties. My own thinking about the specification, and the related specifications had developed. I had started teaching grid not as a standalone module, but alongside Flexbox and Box Alignment. I was trying to demonstrate how these modules worked together to create a layout system for modern web development.

Another place my thinking had moved on since my initial Grid article in 2012, was in terms of content reordering and accessibility. In July of 2015 I wrote an article entitled, Modern CSS Layout, Power and Responsibility in which I outlined these concerns.

Some things change, and some stay the same. The grid- prefixed gap properties were ultimately moved into the Box Alignment specification in order that they could be defined for Flex layout and any other layout method which in future required gaps. What I did not expect, was that four years on I would still be being asked about Grid versus Flexbox:

“A question I keep being asked is whether CSS grid layout and flexbox are competing layout systems, as though it might be possible to back the loser in a CSS layout competition. The reality, however, is that these two methods will sit together as one system for doing layout on the web, each method playing to certain strengths and serving particular layout tasks.”

2016: What next for CSS Grid Layout?

Latest web browser versions: Edge 15, Firefox 50, Safari 10, Chrome 55

In 2016, we still didn’t have Grid in browsers, and I was increasingly looking like I was selling CSS vaporware. However, with the spec at Candidate Recommendation, and it looking likely that we would have grid in at least two browsers in the spring, I wrote an article about what might come next for grid.

The main subject was the subgrid feature, which had by that point been removed from the Level 1 specification. The CSS Working Group were still trying to decide whether a version of subgrid locked to both dimensions would be acceptable. In this version we would have declared display: subgrid on the grid item, after which its rows and columns would be locked to the tracks of the parent. I am very glad that it was ultimately decided to allow for one-dimensional subgrids. This means that you can use the column tracks of the parent, yet have an implicit grid for the rows. This enables patterns such as the one I described in A design pattern solved by subgrid. At the end of 2019, we don’t yet have wide browser support for subgrid, however Firefox has already shipped the value in Firefox 71. Hopefully other browsers will follow suit.

Level 2 of the grid specification ultimately became all about adding support for subgrid, and so we don’t yet have any of the other features I mentioned in that piece. All of those features are detailed in issues in the CSS Working Group Github repo, and aren’t forgotten about. As we come to decide features for Level 3, perhaps some of them will make the cut.

It was worth waiting for subgrid, as the one-dimensional version gives us so much more power, and as I take a look back over these 24 ways articles it really underlines how much of a long game contributing to the platform is. I mentioned in the closing paragraph of my 2016 article that you should not feel ignored if your idea or use case is not immediately discussed and added to a spec, and that is still the case. Those of us involved in specifying CSS, and in implementing CSS in browsers care very much about your feedback. We have to balance that with the need for this stuff to be right.

2017: Christmas Gifts for Your Future Self: Testing the Web Platform

Latest web browser versions: Edge 16, Firefox 57, Safari 11, Chrome 63

In 2017 I stepped away from directly talking about layout, and instead published an article about testing. Not about testing your own code, but about the Web Platform Tests project, and how contributing to the tests which help to ensure interoperability between browsers could benefit the platform - and you.

This article is still relevant today as it was two years ago. I’m often asked by people how they can get involved with CSS, and testing is a great place to start. Specifications need tests in order to progress to become Recommendations, therefore contributing tests can materially help the progress of a spec. You can also help to free up the time of spec editors, to make edits to their specs, by contributing tests they might otherwise need to work on.

The Web Platform Tests project has recently got new and improved documentation. If you have some time to spare and would like to help, take a look and see if you can identify some places that are in need of tests. You will learn a lot about the CSS specs you are testing while doing so, and you can feel that you are making a useful and much-needed contribution to the development of the web platform.

2018: Researching a Property in the CSS Specifications

Latest web browser versions: Edge 17, Firefox 64, Safari 12, Chrome 71

I almost stayed away from layout in my 2018 piece, however I did feature the Grid Layout property grid-auto-rows in this article. If you want to understand how to dig up all the details of a CSS property, then this article is still useful.

One thing that has changed since I began writing for 24 ways, is the amount of great information available to help you learn CSS. Whether you are someone who prefers to read like me, or a person who learns best from video, or by following along with a tutorial, it’s all out there for you. You don’t have to rely on understanding the specifications, though I would encourage everyone to become familiar with doing so, if just to be able to fact check a tutorial which seems to be doing something other than the resulting code.

2019: And that’s a wrap

Latest web browser versions: Edge 18, Firefox 71, Safari 12, Chrome 79

This year is the final countdown for 24 ways. With so many other publications creating great content, perhaps there is less of a need for an avalanche of writing in the closing days of each year. The archive will stay as a history of what was important, what we were thinking, and the problems of the day - many of which we have now solved in ways that the authors could never have imagined at the time. I can see through my articles how my thinking evolved over the years, and I’m as excited about what comes next as I was back in 2005, wondering how to make CSS layout easier.

About the author

Rachel Andrew is a Director of edgeofmyseat.com, a UK web development consultancy and creators of the small content management system, Perch; a W3C Invited Expert to the CSS Working Group; and Editor in Chief of Smashing Magazine. She is the author of a number of books including The New CSS Layout for A Book Apart and a Google Developer Expert for Web Technologies.

She curates a popular email newsletter on CSS Layout, and is passing on her layout knowledge over at her CSS Layout Workshop.

When not writing about business and technology on her blog at rachelandrew.co.uk or speaking at conferences, you will usually find Rachel running up and down one of the giant hills in Bristol, or attempting to land a small aeroplane while training for her Pilot’s license.

More articles by Rachel

Drew McLellan puts the chairs up on the tables, sweeps the floor, and closes off our season, and indeed the entire 24 ways project with a look back at what it’s meant to run this site as a site project, and what impact side projects can have on the work we do. Will the last one out turn off Christmas the lights?

Brought to you by The CSS Layout Workshop. Does developing layouts with CSS seem like hard work? How much time could you save without all the trial and error? Are you ready to really learn CSS layout?

Fifteen years ago, on a bit of a whim, I decided it would be fun to have a Web Standards version of something like the Perl Advent calendar. A simple website with a new tip or trick each day leading the readers through December up until Christmas.

I emailed a bunch of friends that kept web design and development themed blogs (remember those?) suggesting the idea and asking if they’d like to contribute. My vision had been that each post would be a couple of paragraphs of information. A small nugget of an idea, or a tip, or a suggestion. What happened was something really amazing. I began to receive really insightful blog posts containing some of the most valuable writing I’d seen online all year.

Look at this piece from Ethan Marcotte on Centered Tabs with CSS, or this detailed piece on scripting block quotes from Jeremy Keith. I was blown away, and the scene was set.

Collaboration

What I hadn’t anticipated in 2005 was that this little side project would turn into a fixture of the industry calendar, would introduce me to a raft of field experts, and would have me working with an eclectic team of collaborators for fifteen long seasons.

And that last point is crucial. I’ve by no means produced this alone. Rachel Andrew has been a constant supporter in helping each year to see the light of day and producing our ebooks. After a couple of years, Brian Suda stepped in to help me plan and select authors. In 2008, I managed to persuade Tim Van Damme to replace my very basic site design with something altogether more fitting. In 2010, Anna Debenham came on board initially to help with the production of articles, but rapidly became a co-producer working with me on all aspects of the content. Owen Gregory joined up that same year to help with the proofing and editing of articles, and for many years did a fantastic job writing the home page article teasers, which are now but a shadow of their former selves.

Also in 2010, we produced a book in collaboration with Five Simple Steps and raising funds in the memory of Remy and Julie’s daughter, Tia Sharp.

In 2013, Paul Robert Lloyd stepped up to the plate to provide us with the design you see today, which not only subtly shifts colours between each day, but across the years as well. Compare the reds of 2005 to the purples of 2019, and the warm tones of a Day 1 to its correspondingly cool Day 24. It’s a terrific piece of work.

In 2014 we won a Net Award for Best Collaborative Project at a fancy ceremony in London. Many past authors were there, and as it was an aware for our collaborative efforts, we all posed with the glassware for photos.

Looking back, looking forward

But even I, Sea Captain Belly Button am not enough of a navel gazer to just be writing an article just about this website. As we draw our fifteenth and final year to a close, it’s important to reflect on what can be learned. Not from the articles (so much!) or from the folly of committing to a nightly publishing schedule for a month every year for fifteen years (don’t do it!) but from the value in starting something not because you have to, but just because you want to. From scratching an itch. From working with a friend just because you love spending time with them. Or for doing something because you see the opportunity for good.

As web designers and developers, we have the opportunity to turn the skills we use in our profession to so many different purposes. In doing so you never know what good might come from it.

Seeing the good

This week I asked around to find out what good others have seen from their side projects. Long time 24 ways contributor Simon Willison had this to say:

Basically every job I’ve ever had relates back to a side-project in some way

— Simon Willison (@simonw) December 17, 2019

Simon went onto explain how it was a website side project that got him his first job in tech. After that, his personal blog lead him to getting a job at Lawrence Journal-World where he created Django. On his honeymoon, Simon and his new wife (and 24 ways contributor) Natalie Downe created Lanyrd, and Simon’s more recent Datasette project landed him a JSK Fellowship at Stanford. That’s an impressive record of side projects, for sure.

Others had similar stories. My good friend Meri Williams is currently CTO of challenger bank Monzo, as well as being a trustee at Stonewall and Chair of The Lead Developer conference.

I got asked to write the book you tech reviewed off the back of a Meetup talk.

Chairing @TheLeadDev has led to me getting to hire & work with so many new brilliant people, as well as getting me multiple CTO gigs (both perm & interim)

Got the gig chairing Lead Dev after meeting @RuthYarnit at a dev meet-up in the basement of a pub in Reading (which I think @drewm you also spoke at / introduced me to!)

Leading LGBTQ employee network at P&G eventually led to my applying for board role at Stonewall

— Meri Williams ???????????? (@Geek_Manager) December 18, 2019

Again, an impressive list of achievements, and I’m sure both Simon and Meri would have eventally found other routes to their individual success, but the reality is they did it through side projects. Through being present and active, contributing a little to their communities, and receiving so much more back in return.

Of course, not all projects have to be directly related to the web or software to be fulfilling. Of course they don’t. Mark Small and Jack Shoulder embraced their love of a good rear end and created MuseumBums, informally cataloging perfect posteriors for your perusing pleasure. On its success, Mark says:

Ok! We’ve been profiled in the Cambridge Independent, the Sun and the Metro; raised money for Prostate Cancer UK; created a little community of museum fans who aren’t afraid to be a bit silly online; and we’ve got a list of big ideas for developing our ????????️???? offer further! ????

— Marky Smallstice ???????????? (@thehistoryb0y) December 17, 2019

Jack adds:

We’ve also got a shout out on @BBCRadio4 and helped a beloved museum achieve record numbers of visitors.

Wow. It’s been a *year*

— Jack Frost (@jackshoulder) December 17, 2019

I had so many heartwarming responses to my request for stories, I really recommend you go over to the thread on Twitter and read it. It’s been one of my favourite set of replies in a long time.

Focussing on what’s important

As the years progressed, more and more publications sprang up both at Christmas and throughout the year with how-to articles explaining techniques. As a natural response, 24 ways started mixing up solution-based articles with bigger picture takes on a wider range of topics, but always with a practical takeaway to impress your friends.

After the embarrassment of white dudes that dominated the early years, we actively sought to open the opportunity to write to a wider and more diverse range of experts. While I don’t think we ever got as much racial diversity in our lineup as I would have liked to have achieved, I’m very proud that each season has been closely gender-balanced since 2012. This is something that was never forced or remotely hard to achieve, all it took was an awareness of the potential for bias.

Calling time

With all the benefits that side projects can bring, it’s also important to be mindful of downsides. Not every project will take flight, and those that do can also start to consume valuable time. That’s fine while it’s fun and you’re seeing the benefits, but it’s neither fun or healthy long-term to have no time away from something that might otherwise be your job.

Spending time with family, friends, and loved ones is equally important especially at this time of year. Just as anyone who does a lot of sport or fitness will tell you about the value of rest days between your activities to let the body recover, time away from ‘work’ is important to do the same for your brain.

Having run this site every Christmas for 15 seasons, it’s time to take a breather and give it a rest. Who knows if we might return in the future, but no promises. It’s been a good run, and an absolute privilege to provide this small tradition to the community I love.

So from me and the whole 24 ways family, Happy Christmas to all, and to all a good night.

About the author

Drew McLellan is a developer and content management consultant from Bristol, England. He’s the lead developer for the popular Perch and Perch Runway content management systems, and public speaking portfolio site Notist. Drew was formerly Group Lead at the Web Standards Project, and a Search Innovation engineer at Yahoo!. When not publishing 24 ways, he keeps a personal site about web development, takes photos, tweets a lot and tries to stay upright on his bicycle.

More articles by Drew

Cambridge, Mass. : National Bureau of Economic Research, 2020

Cambridge, Mass. : National Bureau of Economic Research, 2020

Cambridge, Mass. : National Bureau of Economic Research, 2020

Cambridge, Mass. : National Bureau of Economic Research, 2020

New York : Springer, [2007]

Cambridge, Mass. : National Bureau of Economic Research, 2020

Dewey Library - HX236.5.S6293 2020

Dewey Library - HV7561.H85 2020

Dewey Library - HV6431.J674 2019

Dewey Library - U39.C76 2020

Dewey Library - HV6250.4.W65 B69 2019

Dewey Library - HV6692.T39 L48 2019

Dewey Library - JC599.A8 P53 2019

Dewey Library - JC573.2.U6 W54 2019

Dewey Library - UA668.I49 2019

Dewey Library - HV6626.2.S59 2019

Online Resource

Online Resource

Barker Library - UA23.H538 2019

Online Resource

Dewey Library - JK1896.W37 2019

Dewey Library - JC251.A74 A83 2019

Dewey Library - JC573.2.E852 B88 2020

Online Resource

Online Resource

Dewey Library - JC261.A7 S74 2020

Dewey Library - UA23.B415 2019

Dewey Library - HV8079.B62 A44 2019

Dewey Library - JK468.I6 T76 2019

Dewey Library - JK511.K66 2019

Dewey Library - HV9950.C79 2019

Online Resource

Dewey Library - JC423.G74 2019