4 ms·

> There are plenty of things we could do that would help certain people in a big way but we don't because the cost would be prohibitive. We certainly can't do

> There are plenty of things we could do that would help certain people in a big way but we don't because the cost would be prohibitive. We certainly can't do everything (and what we can do is subject to constraints of scale), but there are key phrases here: readily achievable, reasonable accommodation. There are a lot of things that are required that are quite expensive, and the burden for providing those is entirely on the facility, service, employer, etc. You still can't apply a strict cost-benefit analysis, because that would fail almost every time. That's what the ADA is for: to compel (necessitate!) accommodations that would not be made otherwise. Take a look at an ADA checklist sometime: http://www.adachecklist.org/doc/fullchecklist/ada-checklist.pdf http://www.adachecklist.org/doc/fullchecklist/ada-checklist.... > It's a matter of how many other people do and how much it would cost to support them. That's not the point. The point is that you're aware there are users different from yourself, so you take them into consideration. Empathy, not market share. That's what accessibility is all about. > The first version isn't going to be particularly accessible That doesn't have to be true, if you are aware of the simple things you can do to make pages more accessible. Then it's not difficult to put those in as you're building things. As I keep saying, if accessibility is incorporated into the process from the beginning, then it is much easier to ensure ongoing support. It's not hard. > it's worth the effort to go back and do it This kind of mindset is exactly the problem. Going back and re-evaluating your accessibility from scratch is a complete waste of time and resources [1]. Take an afternoon off from faffing around on HN, read up on accessible design, apply those few simple steps as you're building each page. Believe me, you'll feel a lot better about it knowing that your pages are accessible: there is something satisfying as an engineer knowing that your product is resilient and usable in various circumstances. [1] Most new public construction is required to be ADA compliant at the outset, because there are laws on the books that mandate it. Now imagine if architects didn't actually consider accessibility when designing the facility. Or if they were completely clueless about accessibility issues. That situation is unfortunately exactly what we have with software developers today -- and devs have it even easier, so there is not really an excuse not to make even a passing attempt at accessibility from the outset.

AnthonyMouse 11y ago
> but there are key phrases here: readily achievable, reasonable accommodation. In other words, the cost/benefit has to be reasonable. > There are a lot of things that are required that are quite expensive, and the burden for providing those is entirely on the facility, service, employer, etc. Suppose you have something like a wheelchair ramp, which might cost $2000. A particular store has maybe 10 customers who need it, the customers each spend ~$200/month in the store, the store's margin is ~10%, so the store pays for the ramp in less than a year. A rational owner of that store is going to build the ramp regardless of whether the law requires it. Now consider private homes. The people who need ramps would have installed them regardless and everywhere else they would rarely if ever be used. But it would still cost the same per home as it would for a store with a lot of profitable business. So the law [in most places] doesn't require ramps in private homes. Here's the problem with the law. Very small shops are in the same boat as private residences. The cost of the ramp is fixed but the value depends on how many people go there and need it. For very small shops that number is often zero. And the cost of the ramp is nothing to Walmart because they do a lot of business, but the same cost is real money to a tiny shop owner. It's the sort of thing that, repeated a few of times in similar contexts, will put the tiny shop out of business. So there is a real case to be made that very small shops shouldn't be required to build ramps. Because it's all cost/benefit. There is no use building a ramp to a shuttered shop. > That's not the point. The point is that you're aware there are users different from yourself, so you take them into consideration. Empathy, not market share. That's what accessibility is all about. Inefficiency is not empathy. Doing the thing that results in the least waste leaves more resources that can be used to improve peoples' lives. Imagine the inefficiently-spent money was instead put toward medical research to cure the people affected. Or do any other thing that benefits them with a higher cost/benefit ratio than the more wasteful thing. How is that not more empathetic than wasting resources that could have been better used to help people? > Going back and re-evaluating your accessibility from scratch is a complete waste of time and resources Only if you already know you'll have to. If you're today's Facebook, you know that. If you're a small startup, you don't. You're probably going to fail and then it isn't going to matter. It can be more efficient to not spend resources doing it right the first time when there is a high probability of failure, because the high cost of doing it over is only paid if you succeed, and in that case you can afford it.

AnthonyMouse 11y ago

> but there are key phrases here: readily achievable, reasonable accommodation. In other words, the cost/benefit has to be reasonable. > There are a lot of things that are required that are quite expensive, and the burden for providing those is entirely on the facility, service, employer, etc. Suppose you have something like a wheelchair ramp, which might cost $2000. A particular store has maybe 10 customers who need it, the customers each spend ~$200/month in the store, the store's margin is ~10%, so the store pays for the ramp in less than a year. A rational owner of that store is going to build the ramp regardless of whether the law requires it. Now consider private homes. The people who need ramps would have installed them regardless and everywhere else they would rarely if ever be used. But it would still cost the same per home as it would for a store with a lot of profitable business. So the law [in most places] doesn't require ramps in private homes. Here's the problem with the law. Very small shops are in the same boat as private residences. The cost of the ramp is fixed but the value depends on how many people go there and need it. For very small shops that number is often zero. And the cost of the ramp is nothing to Walmart because they do a lot of business, but the same cost is real money to a tiny shop owner. It's the sort of thing that, repeated a few of times in similar contexts, will put the tiny shop out of business. So there is a real case to be made that very small shops shouldn't be required to build ramps. Because it's all cost/benefit. There is no use building a ramp to a shuttered shop. > That's not the point. The point is that you're aware there are users different from yourself, so you take them into consideration. Empathy, not market share. That's what accessibility is all about. Inefficiency is not empathy. Doing the thing that results in the least waste leaves more resources that can be used to improve peoples' lives. Imagine the inefficiently-spent money was instead put toward medical research to cure the people affected. Or do any other thing that benefits them with a higher cost/benefit ratio than the more wasteful thing. How is that not more empathetic than wasting resources that could have been better used to help people? > Going back and re-evaluating your accessibility from scratch is a complete waste of time and resources Only if you already know you'll have to. If you're today's Facebook, you know that. If you're a small startup, you don't. You're probably going to fail and then it isn't going to matter. It can be more efficient to not spend resources doing it right the first time when there is a high probability of failure, because the high cost of doing it over is only paid if you succeed, and in that case you can afford it.