Building accessible websites and apps is a moral obligation | Hacker News Reader

Building accessible websites and apps is a moral obligation

gomakethings.com

54 pointsawb84 comments

So let's say some random person, Bob, builds an awesome website in his free time. Bob spends hundreds of hours just so others can use his awesome idea. Now you're telling him what to do. That's not morality, that's just entitlement.

Even if Bob was to put some ads on his webpage, because he wasn't so selfless after all, then what?

> It’s not on people with disabilities to tell you how you screwed up

If you have a problem, you complain. If you don't complain, then you'll have to wait until someone else does or just live with it. That goes for everybody; it's how life works.

> If you build for the web, you have a moral obligation to make sure it works right for everyone.

No. No I don't.

It wasn't my moral obligation to build that website to begin with. It wasn't my moral obligation to make it available to anyone in the first place. It's not my obligation to translate the web page to make it available to people who don't speak my language. And it isn't my obligation to make it available or accessible to any other group of people.

It's one thing to discuss how to best support people with disabilities.

It's another thing to sit on your own high horse, and preach from on high about how other developers should do "their fucking job". I'm so sick of this gluttonous moral posturing.

It goes deeper than that. At my enterprise company, our accessibility department explicitly does not gather any data on our users with accessibility needs. We don't know who they are, what problems they are facing, or if our solutions work for them or not. The reason? It would infringe on their "rights" (What rights are these? I didn't press further).

What they do is hire a few people with visible disabilities (wheelchair or missing limbs) and then parade them around to development teams to shame us into accessibility "compliance". It is a gross process involving lots of trial-and-error.

Privacy rights, including personal medical information. Additionally, counting how many people are affected by a particular accessibility barrier too often leads to ignoring them because the numbers tend to be small.

> gross process involving lots of trial-and-error.

That can be reduced through developer education and changes in processes but doesn't that also describe browser and device testing; it's part of the job.

How do you even make a website that is not accessible for wheelchair users?

One example: You have fucking tiny up and down vote buttons that aren't separated by anything and you make that vote unchangeable.

This means people with motor control disabilities -- some of whom will be wheelchair users -- can't reliably vote using the website supplied buttons.

Simple changes are to make the buttons bigger, and to separate them with something other than whitespace, and to allow some changing of the vote.

It's an interesting discussion. Governments often require buildings and businesses to have certain accommodations for certain types of people.

Are websites different to buildings of businesses?

Mr Bob Hypothetical makes his new burger joint, and spend hundreds of hours working making it so people can eat his awesome burgers. Should there be any building or health codes he should have to follow?

I realise the barrier of entry to a website is lower than a physical business. Presumably though at somewhere between ismyideafunnyyet.xyz and amazon.com/co.uk/ca/multi-national doom machine we have to start treating them with the same or similar expectations to the local chippy.

And at least in the US, that’s how things were. Legislation changed that and forced the issue. Maybe we need an ADA update for the web?

(There was a good movie on the movement the created the ADA called Music Within for anyone wanting something to watch on the subject.)

No we don't, law updates, regulation updates, and case law has made it clear that the principles of the ADA apply to websites as "places of public accommodation."

Do you know if and how those updates translate to codes that need to be complied with?

For example we have codes for ramps that specify the maximum grade, length, minimum platform size, etc. I’m wondering if something analogous needs to exist for the web. (Actual regulatory codes, not W3 standards or guidelines.) it would be “fairly” easy for something like USDS to define something like that for use for all Govt sites, and that could be enforced across agencies. Seems tricky for how you’d handle that with non Govt sites.

Section 508, which applies to government sites (and I think their contractors) used to have its own standards but is now aligned with WCAG 2.0 AA. Most countries have done the same.

You're never going to have standards for software as clear-cut as architectural standards for ramps and such, the factors for software are too complex. There are only a few digital accessibility standards that are very specific, like color contrast between foreground and background at particular font sizes.

There is a need for more specifics from the government in specific areas. For instance WCAG 2.0 AA says videos have to have captions but do you have to wait before making a video available until the captions are ready? If you have thousands of hours of uncaptioned video, would it be sufficient to produce captions by request with a reasonable turnaround time? TV broadcasters and DVD publishers have already had rules to follow but as video has become commonplace, regulations haven't kept up.

> the barrier of entry to a website is lower than a physical business

Are you talking about as a proprietor or as a user/customer?

As a customer the access is quite different, to the point of comparing apples to oranges. To access a place of business you simply need to physically get there. The business, if open to the public, needs to make a few reasonable accommodations for special physical needs (e.g. ramps for wheelchairs, etc.)

To access an app or website, you need to have a phone or computer, internet connection, and be able to operate those things. Pretty much everything else can be accommodated by the user.

To me they are not comparable.

I'm talking about as the proprietor: creating the business online is a lot easier than creating a physical one.

Even so:

> As a customer the access is quite different, to the point of comparing apples to oranges.

> To access a place of business you simply need to physically get there

> To access an app or website, you need to have a phone or computer

> To me they are not comparable.

What lack of overlap do you see between people who can physically go to a building and people who can use a computer?

I guess, having thought about it some more, my argument would be more that a physical business benefits from community-provided resources. Streets, sidewalks, perhaps public transit, parking, services such as police and fire protection, etc. The public, having provided that infrastructure, should expect some reasonable accommodations be made with regard to accessing businesses that benefit from it.

A website is nearly entirely provided by private entities paying or being paid for by the users of the service. Government websites excepted, there is really no "public" anything about a website.

> somewhere between ismyideafunnyyet.xyz

The blog post doesn't make that distinction, nor does it seem like the author really cares. That's precisely my point. Is it a moral obligation for amazon to build their services so everyone can use them? There's probably a strong case there.

I think we need to draw a few clear lines. Some websites are inherently tied to one medium and don't make sense beyond it, others should be viewed as a piece of art, just like a drawing or a piece of music.

> I think we need to draw a few clear lines. Some websites are inherently tied to one medium and don't make sense beyond it, others should be viewed as a piece of art, just like a drawing or a piece of music.

Agree there, and it's very hard to draw these lines. Similarly to whether or not it's OK for software you publish to be a buggy mess, somewhere between your one-weekend lisp implementation you pushed to github and home security camera firmware lines need to be drawn.

Maybe morality in general is not a benchmark but rather an endless continuum. Let's say Alice builds a similar website to Bob, but took extra time to add accessibility. You could argue Alice's site is morally superior to Bob's without saying Bob is immoral.

Hackernews readers put a huge amount of stock in building websites without JS, that work in lots of browsers, that deliver tiny bundles or no bundle at all, with old tech rather than new, and that "just work", because those are the things that impact the readers themselves.

As soon as someone suggests building a site that works as well for other people the comments fill up with reasons not to.

This strikes me as selfish.

You might as well tell them, "Tough shit." Different markets and different languages, etc. don't deserve automatic accommodation, but if you fully believe in accommodating disabled people in order to give them the dignity and humanity they deserve, then you'd take on the moral imperative to make an effort at accessibility.

Maybe it shouldn't be a legal requirement for private-market web documents, but dismissing it altogether is not a morally defensible position.

Would the ADA have ever been passed under the principles you're espousing?

People with visual disabilities make up about 2.5% of the US adult population. (Source: https://nfb.org/resources/blindness-statistics#)

People who do not speak English natively make up about 20% of the US population, and around 8% of the US cannot speak English fluently. (Source: https://en.m.wikipedia.org/wiki/Languages_of_the_United_Stat... , http://blog.languageline.com/limited-english-proficient-cens...)

So if you're aiming for maximum inclusivity, you should consider translating your app into Spanish a higher priority than adding ARIA tags.

Unlike learning a language, blindness is not something which can be overcome with education.

I've tried learning multiple languages and have lived abroad and learning another language for me was not easy or fast despite years of studying and practicing the language.

No one said it was fast or easy. Given a long enough timeline and some daily effort, you can most likely learn another language.

You can't study yourself to un-blindness.

So? You still learned it. You can't learn to see if you're blind.

Just because something is possible you shouldn't go around expecting it from anybody. Learning a language just for a website is dumb and nobody is going to do it, so the website is essentially just as unaccessible to someone who doesn't speak the language as someone who's blind.

I don't think you understand that my argument wasn't necessarily contrary to yours, just stating that it's certainly easier to learn a language than it is to not be blind. Both of which are unfair expectations to thrust upon anyone.

So you can expect thousands to learn English because of your one website, but it's immoral that some disabled people will just have to miss out on your website because you can't be bothered to spend countless hours on accessibility?

Honestly, that statement is about as dumb as saying "Blindness can be overcome with science, all those blind people should just start working on artificial eyes".

If all you care about is the US.

Presumably the 2.5% figure is reasonably similar in many other countries, as opposed to the % who speak spanish, which will be higher in spain but lower in kenya.

Except disabled kenyans won't really benefit from your english site being more accessible. Translating into spanish, on the other hand, will benefit lots of people outside the USA, for example in Mexico or Spain. So by extending the scope to outside the USA, you're essentially making an even stronger point that localization > accessibility

Erm English is a fluent second / third language to many Kenyans. Or do you mean something else?

Anyway, I just picked Kenya randomly. I could have also stuck to English-speaking countries like NZ / Aus / Uk.

> localization > accessibility

Speaking as a disabled (visually impaired) user here: Chrome can translate many websites while keeping their functionality intact, to the point where for many tasks, understanding the local language of that website is not a requirement. Granted, we can say this is for simple tasks, things that we already know how to do in our native language and are much the same across the world. But I'm living in a country where the majority of people don't speak English and the majority of websites and apps don't have an option to switch them into English. I'm still learning the local language, which is a slow process, but I don't feel too bad using websites here to get things done.

Chrome can't improve the accessibility of an inaccessible website or app, full-stop. If it's inaccessible to the point of being unusable, there is nothing the browser can do. Maybe as a developer I can userscript it into compliance, or maybe I can just go to one of your competitors who either have an accessibility-first culture or have stumbled on accessibility by accident (I don't really care which). And maybe all the other blind people out there who don't know how to write JavaScript can learn it so they can make your website work for them. It's just as easy as learning how to say "Add to Cart" in Spanish, right? The other day I had to use the browser console just to check out on Lenovo's website, for crying out loud. Lenovo. Come on.

> Chrome can translate many websites while keeping their functionality intact, to the point where for many tasks, understanding the local language of that website is not a requirement.

And that feature works better if the site follows accessibility best practices! So you might as well not put text in images, avoid overly-elaborate JavaScript, and so on; it'll help foreign language speakers as well as people with visual impairments.

People can learn other languages. Disabled people cant get rid of their disabilities.

If you only look at market advantage, yes. But from a risk management perspective, non-English speakers are not a protected class so it's not legally discrimination if you don't translate it but it is if you fail to make it accessible to people with disabilities.

It's better for everyone for them to become fluent in English. People with blindness aren't refusing to become fluent in sight.

If those websites are life or business supporting or required to use (government, etc).

I don't think it's a "moral obligation" to make e.g. a gaming website accessible, any more than it is to "help the poor" or thousands of other ways we can directly do good.

In other words, if someone fails to make their hobby SaaS for passive income accessible, they haven't have some "moral failing". They just prioritized shipping basic functionality for the majority of users. It's not really that different that opening a website or business for business in one country (or part of the country) and not another.

> It's not really that different that opening a website or business for business in one country (or part of the country) and not another.

It is, because people can choose to move to a part of the country where your business operates. Maybe people don't do this as a rule, but the option is there. People can't just opt to not be disabled.

That is an absurd barrier. Move to a foreign country to get access?

Also, money is the ultimate accessibility issue that affects everyone, and that's something Facebook and friends understands, which is why they're swallowing the world.

Another huge reason people can't access things is because of linguistic or cultural access. I mean, some people say the professional or academic writing of their own culture has access issues...

> That is an absurd barrier.

It may be absurd for someone who has always lived in a location with access to the education, job opportunities and other services which allow them to live the life they want. It's not quite that absurd for the millions of people who have migrated to other countries rather than stagnating in a place they didn't choose to be born in. It's not that absurd to those who live in countries so vast that moving to another city may involve the sorts of distances that people from smaller countries can't cover without crossing a border. And it's not at all absurd for disabled people who do this because their country doesn't allow for independence.

Nevertheless, my original comment was more addressing people who move to another city/state, or away from a rural area, to gain access to services which otherwise would remain unavailable to them. That happens all the time.

Next you're going to say that people should be making themselves unpoor? Because you can't be unblind but you can be unpoor (I mean, you could even move your family to another country and learn a new language), should we now be asking why people choose to "remain poor"?

Money is the most severe access issue in the world, which is why Facebook and friends are dominating. That's why they want to subsidize everything, because even $5 is a barrier which will stop people across the world.

>It is, because people can choose to move to a part of the country where your business operates.

Yeah, I don't see the "move to a different state or country" much less of a barrier. As if people ever moved countries just to get e.g. Netflix or some chain they don't have at theirs?

This attitude is the same as every other business choosing to ignore the issue, scaled down to one person. This demonstrates why larger organisations don't do the work either.

I would have no problem if this were only a moral obligation, but it seems like it's becoming more and more a legal obligation. Legal obligations that require a developer add something frustrate me, because they essentially crush people doing things for fun.

It's also a business obligation, in my opinion, because 1) an increasing percentage of the population has some form of disability, and 2) not accommodating them has a bottom-line business impact.

There's a moral, legal, and business obligation to serve your product in an inclusive manner. We're working to get select developers IAAP (International Association of Accessibility Professionals) certified on my team.

On the legal front, see also Gomez v. GNC: https://www.levelaccess.com/federal-court-decision-gomez-v-g...

I'd say, when people build stuff for free, or in a way that it just barely covers the costs, it's even morally wrong to expect them to make their stuff more accessible. They're already doing others a favor and asking for even more is just a sign of entitlement.

I think that the baseline more obligation is not to be exclusive, i.e. whether you are working for free or not, you have an obligation not to design tools that actively further harm to a group, demean or discriminate against a protected group, etc. The duty to be actively inclusive scales with commercial reasonability. For example, it would be reasonable to expect a small business to read aloud a menu to someone with a visual disability, at minimal cost. By contrast, it would be reasonable to expect a transnational company like Walmart to spend significant sums making stores wheelchair accessible. Someone who does something for free, or for incidental income, should only have to do very low time/cost accommodations.

Here's my breakdown:

> Morality is not always relative (house example)

Building an unsafe house exposes people to active harm. When you invite someone into a house you make the implied warranty that the roof will not collapse and kill them do to your negligence.

> You’re a web professional

If you're making a new product for Google or Microsoft, I completely agree with this article. However, developing web apps is not the same as building bridges. If you are a bridge builder, you're a licensed civil engineer with a sizeable government budget. If you build web apps, you might be a 30-year senior dev, or you might be a student trying to launch a startup with a hundred dollars and no employees. Or, you might just be making a personal web page for yourself or family.

> The web is accessible out-of-the-box. We break it.

As with any term, "the web" can have multiple meanings. This statement is true of the original web implementation, "[an] HTML file with no CSS and no JavaScript." That's not the web anyone knows today. Modern definitions of the web might include PWAs,

> It’s not on people with disabilities to tell you how you screwed up

If you're Google and you "forget" to alt-tag images, again I agree by all means. But this is only true if we're talking about common knowledge stuff (which is relative to size and resources of the developer) or things that are deliberately ignored.

> It should be easier

Why is it the job of a side-project, non-profit, startup, or personal developer to conform to obscure, complex standards--why isn't it the job of the for-profit screen reader companies to make decent standards in the first place?

> you have a moral obligation to make sure it works right for everyone.

Building anything is a game of tradeoffs. If you can afford and have the expertise to do so, you absolutely should invest in a11y. Your moral obligation is to not intentionally harm users, not to spend endless amounts of time complying with every standard and best practice.

I agree to a point, but I think it goes too far at times, mostly when it comes to making _all_ images/video accessible. I can't imagine YouTube would exist if it had to have captioning for every single video to begin with, for example. (Now they auto-generate, but I don't think even today your average independent web dev could make their own videos do that, pretty sure YouTube needed a huge corpus of content to start with to develop it. And even with auto generated subs, any video that's mostly things happening without words is "dark" to a blind person.)

> I can't imagine YouTube would exist if it had to have captioning for every single video to begin with

In many instances, the rise of YouTube has been deleterious to the web. So much content that would have once been presented in text is now video, with the result that what might have been a fast skim now forces readers to move at the slower pace of the video, and it is much more difficult to parse for those with disabilities or to find in a search engine. (Too much content today has moved to images, too – it is remarkable how people who would have once written a blog consisting of longform text, are now trying to shoehorn their content into the more limited space that Instagram provides for them.)

I can never get over the fact people share images with one or two sentence "wisdoms" or "jokes" as their only content on social networks instead of just copying text.

Yea. But for the work most developers do, making their content accessible is largely just learning how to do it. UI toolkits will generally do the right thing out box. Using the semantically correct elements for your content, once you learn how to do so, is no extra development cost. For me, I belive it is the morally correct thing thing to do, for the 1 of me to spend a few hours learning in order to make N people's lives that little bit easier in the future.

The main accessibility cost is when it comes to building new UI patterns. But that's a relatively rare case, most sites will be served fine by piecing together elements from existing toolkits and modules.

Perhaps a more general state is that it's a moral obligation to think and care about inclusivity relevant to your platform, instead of just "works for me"ing it.

The application I work on is not specifically inclusive of blind people. However, it is inclusive of people who have very basic levels of general and technical education (e.g. not understanding what a percentage is), because those are the people we know use it.

I think it's understandable if your videogame doesn't work with blind people, but lots of people have simple visual imparements you could work around (colour blind modes, scalable UI for old people etc).

Similarly, giving a shit about having good contrast on your website doesn't seem like a big ask, as well as not breaking everything when someone wants to scale fonts.

I don't think you have to support every possible use case under the sun, but at least thinking about and considering some of them is a big positive step.

You could turn anything into a moral obligation, eg. I have a moral obligation not to eat pasta because there is less environmentally damaging food available and it's also probably doing some non-zero amount of harm to my body compared to something healthier so since I may feel a tiny bit worse therefore I'll have less potential energy to help someone in need or have a lower probability of assisting a disabled person when I next encounter one. So let's not play that game.

As a contract developer who provides estimates on my work I always include a section for ADA compliance, WCAG compliance, Accessibility, and a few others. I provide time estimates and explain to my clients that I consider these extremely important things to do. My company has the ability, knowledge, and experience making websites that are accessible.

Our clients often don't want to spend the money required for this task.

This seems to assume that the developer is at fault for making the choice to not make websites accessible and not management or clients who don't want people to spend time on "useless" things like aria attributes.

I preach about accessibility to every manager and client I meet. Most don't care because it "costs them money for such a small percentage of users".

Adherence to laws or special guidelines may take time and cost money, but isn't a professional working to spec what a client should be able to expect? Should a baseline of accessibility not always be provided and thus figured into the base price? Does adhering to HTML Spec for example cost a premium price?

Adhering to HTML spec does not by default make a fully accessible website. White text on a white background adheres to the spec but no one can see it. If you mean adhereing to something like WCAG then that is a premium since it requires things like aria attributes, balancing contrasting colors for foreground and background and a whole slew of other things.

Isn't it my job as a professional to tell my clients what I am doing and how much it will cost them? Isn't it my job to also explain the importance of these line items? I do this task.

Isn't it my client's professional job to then take that information, process it and see how those line items affect their bottom line? If accessibility for the blind on an Aircraft Pilot app website for Pilots who need vision to have that job and use the app is then excluded is this a moral failing on my part or the clients fault? I would say no in both cases, they have literally zero visitors who are blind visiting their website.

Your arguments seems reasonable and I heard it before, but I am not really convinced it strikes to the heart of the matter. Accessibility does include more than the HTML Spec, but adherence to the spec already provides a huge chunk of accessibility for free. And I don't feel every website needs to be 100% accessible, but people complaining about lacking accessibility often do so because developers disregard good practices and common sense. Yes, bad contrast is bad, but most disabled users know ways to fix this minor issue[0], but needing js to dynamically load basic text and image content that a screen reader than struggles with is just inexcusable and gets to the hearth of the issue, that ignoring accessibility also often means ignoring different user Plattforms and usage. A finely crafted website is not only readable by a modern browser and a screen reader, it will also very probably mostly work on older browsers, minor browsers/operating systems and so on. So every developer should have accessibility in mind when creating websites.

Of course, being competitive, a shoddy developer who just bangs two wordpress themes together will always be cheaper than good craftsmanship, but in my experience, the problem is more often targeting the wrong clients anyway.

[0] And, as sad is this is, the actual design is often not the developers responsibility, unless of course, when it is, in which case he is also to blame. ;-)

I agree with you, my only point is that accessibility costs something. It either costs time and money if you want to adhere to some standard like WCAG, it costs in hiring a better designer who doesn't make inaccessible design choices and understands basics like contrast ratios, and/or it costs in hiring a developer who has enough experience and common sense to know what is accessible and what isn't.

Accessibility costs time and money, there is no way around this. Not all companies want to pay that money. I don't think the entirety of the blame can be placed on developers.

You are right, it is not entirely free. I do see a baseline accessibility as necessary to call yourself a competent developer though. And you are right, a competent developer costs more than a bad one. But we are a community of developers here who, i would assume, think of themselves as competent. Should we not strife for a certain baseline standard instead of partaking in a race to the bottom? In the end, there is always someone that does a bad job cheaper than we can. I know there are often economical circumstances and we all cut corners one time or another, but we as the knowledgeable developers must be the one to educate the clients and put in the work, because no one else will.

> Should a baseline of accessibility not always be provided and thus figured into the base price? Does adhering to HTML Spec for example cost a premium price?

If the client doesn't give a crap about it, in the end, a competitor will offer the service without these costs added in, and will win the bidding wars. Adhering to the spec by itself does not guarantee accessibility.

It's not a question of "should it cost something?", but "does it cost something?", and it does.

Perhaps you might have more success if you abandon the practice of separating Accessibility into its own line item, and simply include it as a given in your estimate. It could be worth a try.

His competitor will provide separate quotes and win.

I tried this at one point and my competitors just excluded accessibility entirely and won the "bidding wars". This is not a solution.

I'd say the moral obligation of the developer is to be aware of the issue and be capable of delivering an accessible website if the client wants it. Asking for anything beyond that is unrealistic.

“Asking for anything beyond that [from the developer] is unrealistic.”

It seems realistic that we could (and IMO should) ask for it from the businesses though. Just in line with existing non-internet accessibility requirements we have for businesses, government organizations, public institutions, etc.

It's as much my moral obligation to build an accessible website as it is a mason's moral obligation to build a wheelchair accessible house.

> An HTML file with no CSS and no JavaScript is accessible by default.

This is not true. What about color contrast? What about alt text for images? These problems (and many others) can occur in plain html pages.

Color contrast between a default background #FFF and foreground text of #000 is pretty much the gold standard.

That’s true! But the author didn’t talk about just sticking to the defaults. He made it seem like it was as simple as not using CSS and JS. In reality, it’s more complicated, like you pointed out.

If you don't apply styling, the browser defaults for color contrast are often sufficient.

But it isn’t as though the problem is “using CSS” — it’s doing the stuff that people currently use CSS to do. Using html to do that same stuff is no better from an accessibility perspective, which is what the quote seems to imply.

Note: I’m all for web accessibility and I work in this space. I think it’s important for advocates to not overstate the case for the simplicity of compliance, which is what I think happened here.

The default blue link color is also accessible by default, w/ a contrast to white of 9.3: https://color.review/check/0000EE-FFFFFF

> This is not true. What about color contrast? What about alt text for images? These problems (and many others) can occur in plain html pages.

Another example: one must use <span lang="XX"> tags on foreign words and phrases sprinkled in a text, otherwise screenreaders can choke on them.

If there's no CSS then there is no color. It's up to the user's browser how to display it, what fonts to use, and what colors.

We have vision algorithms that are able to tell that a picture is a picture of a horse, why do we still need the alt attribute ?

We have the level of knowledge to build a tool that make website accessible, so my opinion is you need that tool. It is easier to build one accessible tool than to ask all developers of the world to put an alt attribute and respect the h1, h2 naming

Yes. And in some cases you can actually be fined and or sued for ignoring accessibility requirements for handicapped and senior users. So you could argue the government will enforce morality if you do not.

Side note, in interviews I always gauge seniority by asking about accessibility. I find it an obvious characteristic of more junior engineers to not have had to consider accessibility APIs. It is a must, and rightfully so.

>Would you say the builder has a moral obligation to not build a house that’s going to collapse on people?

This is nothing to do with moral obligation. At least for me, much more to with the cost vs return on building a house that’s going to collapse on people.

Unless I'm forced to build website that is accessible, I won't do it.

TFA gets a few things wrong - the analogy to house building is off, and I don't think the moral argument is even necessary - but ultimately if, as I infer, he's talking about web professionals, not hobbyists, those people should care and be educated about how to do this.

The house analogy bothers me because a house is not a public space the way a shop or museum is. And a lot of accessibility concerns like contrast ratios are simple and can be minor CSS changes - analogous to curb cuts in sidewalks. But if you're acting as architect for a public-facing website for a business then it absolutely makes sense without invoking any moral obligation to engineer at least a minimum amount of accessibility.

I look at this very simply. If you are a business that is marketing the app to external consumers, adding internationalization and accessibility protects you from possible legal and regulatory trouble, while also increasing the size of your potential user base.

But for side projects? No, when it comes to my side project, I'm the master, in the dictator, what I say goes. Users don't like it? Tough, go build your own app.

Sure, but you have to accept that means you are not committed to the dignity of disabled viewers.

You should build for your market. Unless you're specifically targeting a market segment that has accessibility requirements, then you should do the least amount of work possible to provide value, then iterate. When creating a product in an early stage market, most development time is better suited to discovering an MVP than adding accessibility features. Think of it as premature optimization.

Yes, it's a good thing for sites and apps to be accessible.

No, it's not an obligation of any kind (nor should it be).

For companies that have grown enough to be providing services to (and profiting off of) a large enough part of the population, it would make sense to expect them to put resources into this. For someone working for free, trying to get an idea off the ground, or servicing a very niche subset of the population (i.e. a very restricted market) this blanket statement makes no sense and the costs it would add to make it an obligation would likely only cause a decrease in quantity of what is available (which, if deemed valuable, can be made more accessible later on) rather than an increase in quality of what's available.

I thought screen readers by now would be able to adapt to the large volume of malformed web content out there instead of breaking on misused tags etc.

Chalk it up to SEO if you need to make a business case for it.

Reply on news.ycombinator.com