I have over a hundred individual maps saved as pdfs for a land use plan that we would like to tag for 508 compliance. We select all the elements on the map page with the Touch Up Order tool, group them into a figure, apply an alt tag and then save. When we ask Acrobat to read out loud from the pdf map, it reads the alt tag first then proceeds to go through and read all the gobblygook on the map - "-145 degrees west, 55 degrees north, etc." How can we prevent it from reading the gobblydegook on the map and read only the alt tag? Everything I read online leads me to believe that once you have the text elements grouped as a figure, its only supposed to read the alt tag for the figure. What are we doing wrong? Could this be a bug? We are using Acrobat Pro 9.0.
"...once you have the text elements grouped as a figure, its only supposed to read the alt tag for the figure." — not true.
No "bug"; rather, it is expected behavior.
<Figure> is one of three Illustrative Elements and is intended specifically for those parts of PDF page content that are images that support conveyance of the document's semantic content but, as an image, cannot be directly rendered as "readable". So, the Alternate Text description.
The "gobblygook" is real content of the PDF that makes up a critical component of the document's semantic content. ROL is going to process any textual content present (tagged or not). ROL is certainly a helpful tool for getting a "sense" of things. It isn't adequate for use as a determinative "accessibility" or "S508" tool.
To properly tag the PDF such real content requires that it be provided appropriate PDF Tags (elements) that observe the correct tag semantics which are discussed in ISO 32000-1 & by mid-year ISO 14289-1 (the PDF/UA ISO Standard that is in an "FDIS" status).