I am drafting an article for the RI:Technology blog on Screen readers and Search Engines, and was reviewing a paper a colleague wrote about Search. He mentioned sIFR as a technique “to bring content to search engines”. I asked another colleague about this, as I’d always just considered sIFR as a “stylability” technique.
We started talking about the weight factor of search engines, whether content written to a page and then sIFRized would be weighted more heavily than the alt text of an image. I hadn’t really thought about that before. I then mentioned a habit I have of placing images within a heading tag, i.e.
<h1><img alt="descriptive text" /></h1>. Toby asked if this really worked, if the alt text would be considered the header. I realized that I’d never really verified it before.
So I took a quick look at an example using FANGS, and learned that alternate text, really isn’t. Turning off images would cause the alt text to display as the appropriate heading level, but at least for FANGS, the alt text does not get surfaced as the heading (the text should show up before the colon in the screen shot below)
Now, obviously FANGS is an emulator, and it’s possible that a screen reader would access that alt text. But this gave me something to consider..