When generating the report, we need to know whether the Text item has rich text (html) or not. That is because we need to avoid generating HTML from that item's text when it is already HTML (otherwise, what the user sees is the actual HTML code, naturally. This is illustrated below). We might be able to heuristically guess that (e.g. for support of the Text items in existing report devices from older RiverWare versions RiverWare).