Web usability guru Jakob Nielsen has come under fire for his latest suggestions on building mobile websites. Nielsen’s controversial advice can be distilled down to this nugget: “good mobile user experience requires a different design than what’s needed to satisfy desktop users. Two designs, two sites, and cross-linking to make it all work.”

Among the many negative responses to Nielsen’s latest Alertbox post is that of web developer and mobile specialist Josh Clarke, who calls Nielsen’s questionable advice “180-degrees backward.”

Indeed, much of Nielsen’s advice may seem like a shockingly bad idea not just to developers, but to anyone who uses a mobile device to browse the web.

This isn’t the first time Nielsen has taken a contrarian stand about mobile websites. Last year he suggested that a mobile-first approach to design was wrong because “PCs will remain important,” which is, at best, a false dilemma since a mobile-first design doesn’t mean ignoring the desktop. Just because you’re focused on the future doesn’t mean you’re ignoring the past.

Nielsen’s latest advice has similar false dichotomies. For example much of Nielsen’s advice rests on the premise that a single site cannot serve the wants and needs of both mobile and desktop users. In fact you don’t need to choose between mobile and desktop, the page can adapt and serve the needs of both users. There are plenty of examples of sites that do just that with responsive designs. To be sure there are plenty of websites that claim to be mobile-friendly and obviously aren’t. But that doesn’t mean the solution is to toss out the whole idea of responsive design and go back to separate websites for every device.

In fact what Nielsen considers one of the “main guidelines” for a successful mobile website is something that many people would probably consider the most irritating thing about mobile sites: “If mobile users arrive at your full site’s URL, auto-redirect them to your mobile site” (emphasis in original).

The problem with this advice is that, as Clark puts it, “Nielsen is confusing device context with user intent.” In other words, just because someone visits your site on a small screen doesn’t mean they won’t want access to all the same things they would see on a slightly larger screen. It may be necessary to rearrange elements for smaller screens, but hiding them is a bad idea.

“All that we can really know about mobile users is that they’re on a small screen, and we can’t divine user intent from that,” writes Clark.

Nielsen, however, does just that, divining that — because a user has a small screen — they will want to do less on your site. He suggests you serve up a limited site and then offer a link to the full site “for those (few) users who need special features that are found only on the full site.”

We suggest you don’t do that. You can do better than that.

You can use responsive design patterns to make sure that the same content is always available, but that the experience is tailored to the device at hand. In other words, responsive design means your site works just as well on mobile as it does on the desktop. If it doesn’t that means something is wrong with your site, not the whole approach. Sure, there will be times when a separate mobile site is the right way to go, but those times will likely be few and far between.

Nielsen’s argument is based on copious research and we have no doubt he found plenty of horribly designed websites that completely fail on mobile devices to justify his recommendations. But just because Nielsen is finding a lot of poorly made mobile websites does not, as Clark writes, “mean we should punt on creating great, full-featured mobile experiences.”