I'm sure this will be the unpopular opinion but... if you're a professional you should take some minimal amount of time to learn the jargon for your industry. Software (just like every other industry) has a bunch of abbreviations like HTTP, DOM, CPU, a11y, etc. They've been around for decades, and blog posts like this won't get people to start typing out full names like Central Processing Unit or Accessibility if they believe their audience is familiar with the term. Learn the jargon once and get on with your career.
Since the parent commenter probably didn’t read the conclusion, here it is:
> [...] By being less accessible, the word a11y denotes a more technical approach to accessibility, not necessarily from those who use accessibility tools, but those who create them.
> That’s why I choose to use the word a11y here, and why I choose the hashtag #a11y on social media. And I’d recommend you do the same to avoid creating noise for people looking to improve their personal accessible experience.
Let us avoid tilting at strawpersons. The author does not claim industry professionals shouldn’t know jargon. From tfa:
So for general use, I recommend against using a11y, yet I’ve named this site with this acronym. There are a few reasons I made this decision. Firstly, WCAG 3.1.1 does permit domain-specific jargon so long as it we define it. So, take this post as a definition. Additionally, I expect most people who have found this site will have found it because they are specifically looking for something accessibility related.
Yes, to be clear, I agree with the author. But I bet the author also thinks it's an unpopular opinion, else they wouldn't feel the need to write a blog post defending their choice.
Jargon isn't a problem, the problem is using jargon in places where you should be able to predict your audience isn't going to understand it.
A11y and WCAG around frontend web developers is usually fine but risky elsewhere, the same way "MVP" is fine in general HackerNews posts but not elsewhere.
Even "accessibility" is jargon and a lot of people outside of tech won't know what it means or what a screen reader is.
I don't think a11y and i18n are an attempt to be clever, they're just really long words to type correctly and take up space so they get shortened, like in Git commits and pull request discussions.
I will never use these stupid abbreviations with the numbers in the middle. Not because of accessibility (although, that's a good reason) but because it's stupid as hell.
I disagree. I've always thought that "i18n" was pretty much the worst abbreviation option. At least with acronyms and more traditional abbreviations, the text of it gives you a mnemonic or clue to remind you of its meaning. These sorts of things give you nothing.
It sounds like the author first says that the word "a11y" is not accessible because it's jargon, but then says that jargon is ok as long as it's defined somewhere. He then proceeds to compare the alternative word "accessibility" with "A11y" and finds they are slightly different in meaning. So it sounds like if we want to use the word "a11y", just make sure it's defined somewhere, and know that there are no exact alternatives. Is that accurate?
Personally a lot of this a11y stuff is complicated and confusing to me. I am trying my best to be a responsible dev, but it feels overwhelming sometimes.
Furthermore he begins by saying "accessibility" is too long to use but "a11y" is too terse. But also use it anyway.
I quite dislike this post because my impression is that it just boils down to "no option is adequate to me, and I'm unwilling to present an alternative".
> Personally a lot of this a11y stuff is complicated and confusing to me. I am trying my best to be a responsible dev, but it feels overwhelming sometimes.
Microsoft also has a great interactive test that can probe your website for you and highlight issues: https://accessibilityinsights.io/
I think similar but less powerful checks are also built into some IDEs (Jetbrains) and linters (Eslint).
Edit: I am a sighted user with color vision. I can't promise these fixes are enough to make your site super accessible, but it's a lot better than nothing!
Well this certainly fits in with the standard for discourse around accessibility, which is mostly a bunch of people inventing a person in their head who will have an issue with something someone else made and attacking the creator/developer/designer for not considering it. Meanwhile the first guy gets to pat himself on the back for being a good person.
"a11y might be confusing to people with autism" -- autistic people are extremely represented in tech. Of all the things autistic people struggle with, it's not technical jargon within their own industry.
"a11y" is not accessible to me. I only saw the term for the first time a little while ago on HN, and I still haven't memorized what it means. Like the other absurd <letter><number><letter> abbreviations, I'll get it memorized eventually.
But really, I wish we'd all just agree to stop with this kind of abbreviating. It's impossible to decipher what it means on its face. You have to memorize it.
> Some people with disabilities such as autism can struggle to understand non-literal language.
Your paraphrase:
> autistic people are extremely represented in tech.
You made a switch-a-roo. The author mentioned one group, and you took a subset of that group who don’t have the problem and conclude it’s not a problem.
I had never heard of a11y; I assumed it was some kind of gender identity.
> Say you’re designing for someone with one arm. There are roughly 20,000 people in the US with one arm. But if you add up the numbers of people with one arm, people with a temporary wrist injury or a broken arm, and folks with one free hand in a specific circumstance (like new parents lugging a baby around), you’re at 20 million in the U.S. alone.
> The common abbreviation of accessibility is a11y.
What?
> We take the A and Y from the beginning and end of accessibility, and 11 for the number of letters in between. This abbreviation also creates a pleasing homophone for ‘ally.’
It took me way to long to figure out what this story is about and that it's not about feminism "allies" as men who align with feministic view are often called. It's all very confusing and the overtures on how clear it is feel out of touch with reality.
Don't get me wrong, I'm in full support of this. Of course I am, as it could happen to anyone with no fault of their ownm, to find themselves in a situation needing help.
A11y, really? This reeks of marketing. But we're talking about it, so it's effective.
Yes, but they are s4d. It's like l33t grew up and made a mess of localization, internationalization, and translation. The numeronym thing feels so exclusive, like a16z. I wonder, if the international web folks like it, because it's their own language. It's true they've seen use since before 2000, but they've also been declining since 2008 (each peaking along with l33t except for a11y, which peaked during WWI, if you believe GoogNGram).
The WCAG website generates a lot of grumbling within the accessibility community. It isn't well laid-out and the blocks of text are not particularly reader-friendly. It would be great if they built in readability enhancements to make it more accessible to people with dyslexia, ADHD, or vision impairments.
It would also then be able to serve as an example for how to implement such augmentations on text-heavy websites.
The problem with numeronyms is they don’t look like acronyms or abbreviations, so people don’t know what to make of them when they first come across them. At least if someone is speaking in regular acronyms, you can understand why you can’t understand.
I thought it was a play on i18n and l10n -- which I think is in line with the idea that a11y has a connotation of accessibility for computers (vs parks or public spaces)
It is exactly that - the same kind of abbreviation as i18n and l10n. As this article points out though, WCAG guideline 3.1.3 advises against weird acronyms like this one.
a11y, i18n, l10n are all reminiscent of a software anti-pattern - optimizing writability over readability.
if (condition); then 1; else 0;
is 31 characters long, while
condition ? 1 : 0
is 17 characters long, or about 45% shorter. However, the first option imposes less cognitive load on the (code) reader. The only code syntax one needs to know is the if-then-else construct, which is taught in every introductory programming course. The second option requires specific knowledge of the ternary operator, which a significant number of programmers are not familiar with. The syntax of the ternary operator also varies by programming language, whereas if-then-else is far more ubiquitous.
Shorthand is useful when taking notes or when working with limited space. In the context of discussing a11y, i18n, and l10n in online documentation, neither of these things are true.
> However, the first option imposes less cognitive load on the (code) reader.
Not in this trivial case; cognitive load here is dominated by effort to identify meaningful parts and reject the noise. It rings true in case of more complex conditionals and expressions, where it quickly gets hard to identify at first glance where the three parts start and end. However:
> The second option requires specific knowledge of the ternary operator, which a significant number of programmers are not familiar with.
Those programmers should then familiarize themselves with it. For those reading this comment right now, here's a primer:
That's it. The more important insight is that the whole thing is an expression, which always has a value, unlike regular if/else construct. This of course matters only for languages which, for some incomprehensible reason, needlessly complicate things by having "statements" and "expressions" be distinct things.
Anyway.
The larger point is that ternary operator is both very useful and very basic part of programming. Refusing to use it when appropriate, on the basis of "think of the juniors" is like refusing to sell paints and paintbrushes because hey, some adults never advanced past crayons and are artistically still 2 year old. And sure, if it's only about your hobby, then you do you - but when you want to do it professionally, you have to respect the fact that programming is a profession - as in, you're expected to learn stuff.
> The syntax of the ternary operator also varies by programming language, whereas if-then-else is far more ubiquitous.
Nah, arguably it's the if/then/else syntax that differs more often between languages. Starting with "then" thing being nonexistent in most.
The extreme aversion to learning, bordering on antiintellectualism, that's plaguing this industry, is not helping in making better software. It's helping companies in solving problems by throwing large amounts of easily replaceable fresh juniors at them.
Those were all really good points, and I agree with most of what you said. Do you agree with my original thesis though, "writability over readability" is a software anti-pattern and also applies to numeronyms such as a11y and i18n?
> The extreme aversion to learning, bordering on antiintellectualism, that's plaguing this industry, is not helping in making better software. It's helping companies in solving problems by throwing large amounts of easily replaceable fresh juniors at them.
I feel the same way about any business or tool that advertises itself as "write less code!". It's not the quantity of typing that is difficult about a programmer's job; it's the cognitive load and context that must be held at any given time. I'm always interested in tools that can help reduce cognitive load, but I am also fearful of leaky abstractions. If a tool wants to inject itself between myself and the machine, the abstraction better be flawless and rooted in first principles.
I think that using plain language is good! There's actually a thing about agencies not using cryptic or confusing communications - https://www.plainlanguage.gov/law/ :)
now, if you're being so plain as to obscure, instead of [say complex and precise word, then explain it in simple terms], then you're taking the piss.
