While Landow's rules are often broad and highly relevant to any discussion of hypertext rhetoric, they betray a particular top-down approach to readerly comprehension and writerly practice. This is suggested in several ways, including the reason that they are presented as 'rules' in the first place (many of these rules appear as prohibitions more than enabling or creative acts or possibilities), and also in the manner in which the very language of the rules contain numerous assumptions about the risk that confusion or ambiguity represents for the reader. These rules, by their very language, encourage their acceptance as canonical, yet the assumptions upon which they appear to be based, for instance the presence of ambiguity and the resulting distress for readers, remain unexamined.

For instance, a dramatically different view is offered by Shields when he points out:

It is through both clearly marked and embedded links that webpages are constructed as apparently stable displays on a computer screen. However, links also disrupt the easy flow of a text or webpage to force viewers into an awareness of the constructed quality of webpages: they send the viewer elsewhere; they break up the authorial control of texts and supplement and problematize what has been displayed or written. (145-6)

Shields continues shortly after, "I will argue that links always disrupt the static quality of a webpage; they move us away to other pages, or up and down a page" (146).

When using links writers ought to assume and embrace the risk associated with linking, while the general program of usability proposes a writing that reduces this risk. The former recognises risk as integral to linking and writes with this risk; the latter domesticates this risk through a language of naturalisation which has been hypostatised by Nielsen.