- Scenario A: Spend $X to make free content accessible to 99% of people. - Scenario B: Spend $X * Y to make free content accessible to 100% of people. - Scena

- Scenario A: Spend $X to make free content accessible to 99% of people. - Scenario B: Spend $X * Y to make free content accessible to 100% of people. - Scenario C: You wish to do A but are forced to do B via legislation. $X * Y is greater than your budget to make said content available $Z, so you don't make the content available. I know of more than one organization that has selected this option with free content and with some paid content (in which case $Z is the budget to create it based on the expected revenue) so this is not a theoretical scenario. Now, Y can vary quite a bit depending on the content and depending on what you're trying to make it accessible to. In the case of making a transcript available, you may have to pay a transcriber. In the case of an app, you may need to buy a screen reader to test with. There are free ones, but they're often not as full featured as the popular one that most folks are using. So, you'll need to buy the $1,100 screen reader (JAWS with the non-Home license) to test against in addition to testing the free open source one. You may also need to pay someone to do more standardized testing with as well. It's not an insignificant amount of money or work in many cases. I've run into it with PortableApps.com. I had to buy a newer version of our toolkit to be able to get it working with screen readers at all and rewrite multiple controls. I don't have JAWS yet to test against but I do test against NVDA. There are ways to mitigate some of this. There are multiple groups that transcribe for free to help make content available accessibly. And if you make free open source software that's useful, you can probably find many impaired users willing to help you test as I have for PortableApps.com. On a personal note, I very much believe in making everything I create accessible.