Great blog!

I read this book by Nielsen, and his "Designing Web Usability" a while ago and both have been very influential. Not just in terms of web design -- I don't do much of that -- but regarding design and publishing style generally. Nielsen is sound and sensible for the most part.

Unfortunately, and this comes through very strongly in "Designing Web Usability", he is also extremely conceited and has very, very fixed ideas. His notions of hyperlinking, for example, on which he is utterly rigid. Hyperlinks must be solidly underlined, and in blue, always. The search button must be near the top -- sort of like a car's steering wheel: the horn must be in the center, that's where God wanted it to be, not on a lever to the side as Ford once attempted. But what would Nielsen say about stereo controls on the steering wheel? Good? Bad? Usable?

Nielsen is worth reading if only to understand fundamental rules of web design. Today, with so many web authoring programs so easily available, everyone is a designer and yet very few understand design, or the limitations and advantages of the medium (in contrast to print especially). Too many designers throw everything, including the kitchen sink, into their designs. Nielsen correctly points out that much of this is rubbish and only alienates the viewer. His comments on how people read on the Internet (they don't read, they scan) are revelations. As other authors explain, in reading print, the eye moves across the page, left to right and then down, line by line, till the page is turned. On the Internet, the eye is relatively static and the page scrolls, up and down. This is not how humans read at all and that is why so many of us who use computers extensively still work better on paper. I am a practicing lawyer, and, despite all my work on a computer, I still find it easier to work on paper.

Having said all that, however, I must say that, at some point, I began to feel that much of what Nielsen says can safely be ignored and, sometimes, he is plain wrong. Take the two examples above. I don't any longer believe that links must be in blue and underlined. Increasingly, Internet users are able to easily distinguish links and follow them even when they are not so identified. In a trade-off between design and usability, with a given design you can break the Nielsen rule on linking without any sacrifice of usability.

Similarly, the search button. Why should it be at the top? Where it should be placed depends on the nature of the site. I went off to bookofjoe, and, frankly, my only comment is that perhaps the "syndicate this site (XML)" link could be dropped lower down the page; but the rest can be left as it is. It's completely logical: on that page, one wants to know who you are; to get a quick link to send you an email (you're sure you want this?); to see your archives; and, since this is a blog, not a formal research site, the search button is, in the natural progression of things, rightly placed below the archives. Please stick with the design: it's clean, it's simple, it's eminently readable and it's very, very usable. Forget Nielsen.

Nielsen is very much 'old guard' when it comes to these matters. He tends to see his 'rules' of web usability as punctuation. Breaking them, he seems to claim, leads to a drop in usability. This is quite evidently incorrect. The Internet is constantly evolving. Users are increasingly adept, comfortable and proficient. They don't need the hand-holding Nielsen advises. See the redesigned Hotmail inbox, or Yahoo's mail interface, for example. Many of Nielsen's rules aren't followed at all and yet there's no real drop in usability.

I still do what I call a basic-Nielsen test on my own designs, though. The test is really simple. Forget you're the designer. You're an outsider. You come in. Can you quickly identify (1) what the site's about; (2) what you're likely to find; (3) dig deeper if you need to; (4) hit the right links on the page accurately?

On this test, I have one major misgiving about http://www.gautampatel.com/, a site where I've dumped lots and lots of Nielsen's edicts and, on his analysis, would proably score zero -- it's not easy to distinguish between the link to Amazon and the link to the full post. I need to sort that out. On the other hand, http://www.mcavity.com seems to be okay, and working well, and the one I did for the Bombay Bar Association, http://www.bombaybar.com, has lots of users who are quite comfortable with it.