In this lesson you
- Judged each tag against the page, even on a rule that said Passed
- Fixed a heading, a list and a line's spoken words
- Applied every change as one new version
Transcript
Passed is not the same as right
The tag editor. When every check on a document is green, is it ready for someone using a screen reader? Not always. Some checks can only ask whether something exists, not whether it is right. This notice has a green check on every step. The Tagged PDF rule says Passed. Even the note at the top says there is nothing to retag. That only means the automatic checks found nothing to question. Passed means the document has tags. A tag is a hidden label that tells a screen reader what each piece is, such as a heading or a paragraph. It doesn't mean the labels are right. Only a person, looking at the page, can judge that.
Reading the tag list
That judgement happens in the tag list, on the Tags tab of the Reading step. Each row is one piece of content a screen reader announces, from the top of the page down. The number on the left matches a numbered box on the page. The menu beside it shows the row's tag. Then come the text's size and weight. Fourteen point B means fourteen point, bold. Content ten of twenty-seven means ten pieces of content are listed. The other seventeen are structure, the boxes around them and the smaller pieces inside them. The menu on each row changes the row's tag.
Judging tags against the page
To judge a tag, compare its row with what you see on the page. Introduction is set fourteen point and bold, like the start of a section. But its tag is P, a plain paragraph. A listener who jumps from heading to heading never lands on it. Proof noticed this one, and asks about it in a card above, as the lesson on tags and headings shows. The list is a problem no card points out. Turn off the Tags layer, the labels drawn over the page, and the page shows a bulleted list with two items. Turn it back on. In the tags, each bullet and each sentence is its own paragraph. So a screen reader reads a dot, then a sentence. It never says there is a list, or how many items it has. And look at the line under the title. Its row shows the letters spread apart. The document painted a space between every letter, so a screen reader spells the line out, letter by letter.
Fixing a tag on the page
The quickest fix starts on the page itself. Select a box, and a small menu opens beside it. Its first row offers a paragraph or a heading level. The second row holds rarer tags, such as Caption, for the words under a picture. If a box shouldn't be labeled at all, Remove tag takes the label off and keeps the content. Decorative tells a screen reader to skip something on purpose, like a border. Edit opens the full editor, which we'll use next. Introduction starts a section, so choose H2, a level-two heading. The row now shows the change as staged. The suggestion card above goes away too, because its question is answered. Nothing is written to the document until you apply.
Fixing the words a screen reader says
The spaced-out line needs its words fixed, not its tag. To do that, choose the pencil on its row. This is Edit tag. The tag type is at the top. The box below holds the exact words a screen reader will say. Fix spacing joins letters that were spread apart, and keeps the gaps between words. Choose it. The line now reads, The federal rights every patient has. Always read the result against the page before you keep it. When text is garbled in other ways, AI clean up drafts a readable version for you to check. If a cleanup goes wrong, Reset to extraction brings back the original words. You don't have to close this window to check the next row. Prev and Next move through the page's rows. Two of ten, at the top, says where you are. Stage and next keeps this text and moves on, so you can check every row on the page in one pass. The next row is Introduction, already staged as H2. Choose Done to close the window.
Making the list a list
Now the list. A list needs four tags working together. L holds the whole list. Each item is an LI. Inside each item, Lbl holds the bullet, and LBody holds the words. L and LI are containers. They are boxes that hold other pieces. The tag list only shows content, so choose Content ten of twenty-seven to show all twenty-seven rows. The extra rows include Span, a small piece of text inside a paragraph. You can leave those as they are. Then go down the rows with each one's menu. The outer box becomes L, and each item box becomes LI. Each bullet becomes Lbl, and each sentence becomes LBody.
One apply, one version
Last, let's write these changes. Nine are staged now, and the document hasn't changed yet. Apply writes them all at once, as one new version, and checks the document again. Choose Apply nine tag changes. The document is now version two. On the page, the Tags layer labels each box with its tag. Introduction is H2. Each list item is one LI, with its bullet inside. So the list now shows eight content rows instead of ten. In the list, the line under the title now shows its new words in color. A screen reader reads those words instead of the painted letters. Tagged PDF still says Passed. This time, the tags are also right. The Tags layer is the quickest check of your work. Turn it off to read the page plain, and on again to see each tag. Now you can judge a page's tags against what you see. Then fix them, and apply them as one version.
Recap
That's the end of this lesson.
Maple Ridge Community Health, its documents and its people are fictional, made for these lessons.