My previous article on user expectation has really got me thinking, and unfortunately for me that means I’ve been asking myself (and anyone nearby who didn’t have the good sense to run screaming from the building) whether we are really giving users what they want.
The crux of the issue for me is the potentially negative implications of experience and user expectation which conspire to create a certain mental model of a website.
Are we really going to let the bad old days of nested tables for layout and mindless tag soup websites define the future of the web? Should the sins of the father be visited upon the son?
There have been some interesting studies done on the expectations of blind users as to the structure of information on a page (e.g. http://www.usability.com.au/resources/source-order.cfm), but they have invariably failed to address the simple and more important issue of want.
Although the work undertaken by Roger Hudson, Russ Weakley and Lisa Miller is no doubt well-intentioned, its failure is in its phrasing. Their study is couched in terms of prior experience and expectation rather than asking the user what they actually want, and the preferences utlimately expressed by their participants are ambiguous at best - there is exactly a third of all respondents in each of the three categories:
More importantly they identify that the key to making things easier for the user is good coding - using headers properly, providing good descriptions, using semantically marked up code and providing good quality links.
A more pertinent line of questioning would be to ask how users actually want their content delivering, not how their experience has led them expect it to be delivered.
Given prior experience of English beer, foreign visitors may have come to expect a warm flat pint delivered by a surly barman with questionable personal hygiene, so it would not be unreasonable for them to assume that this is the way English beer should be. On the other hand I demand a delicious frosty beverage of impeccable character and taste delivered by a personable and fragrant member of staff. My tastebuds have thanked me for my refusal to bow to expectation, and landlords across this fine country have seen their revenues rise or fall accordingly. Now that’s progress.
If the web is to progress and become the ubiquitous information source that it has been evolving towards then there is a definite and compelling argument for deliberate and considered change.
It is sometimes neccessary to abandon that which went before in order to make way for a clearer, easier, more accessible future. We abandoned the complex system of pounds, shillings and pence in the UK in favour of a vastly simpler decimal system despite the upheaval and confusion it caused - should we not be similarly brave in replacing the legacy of the fledgling days of the web with a more robust, studied, and usable model?
Unfortunately the issue is not that simple. By re-ordering the structure of our HTML documents we may make things better for users of screen readers and other assistive technologies, but we run the risk of making the same pages less usable for people using other emerging devices such as mobile phones where screen size and resolution is more restrictive and support for CSS is less mature than in desktop browsers. Accessibility is about universal access for all, not positive discrimination towards a particular group yet too often we see accessibility considerations which only really consider a fairly specific subset of the user group, namely the blind or visually impaired.
The “web standards movement” has been very successful in both making developers more aware of the benefits of standards-compliant build techniques and simultaneously pressuring browser manufacturers into adopting full and uniform support for web standards in their software products, yet handheld devices and screenreaders have fallen behind.
Here we have a double opportunity to both lead and compel progress - support for specific stylesheets for “handheld” browsers is poor and screen readers languish behind the leading visual browsers in their support for clean semantic code. Perhaps some specific advocacy in these areas would help to bring them back up to spec.
Unfortunately many mobile phones are now getting more and more capable and are being shipped with fully-fledged XHTML browsers which eschew the poor cousin that is the handheld stylesheet as they consider themselves to be “proper” browsers, but until they are capable of displaying the same resolution as a desktop monitor I am not convinced that this is a helpful or constructive position.
I would suggest that we take the revolutionary step of asking users to throw away their expectations and tell us what they genuinely want.
Who knows - they may just have an idea or two. Even if it turns out that they really do want things just as they are then you’ve got yourself a nice piece of quanifiable user reaseach with which to shut people like me up.
So far one person has argued with us about ‘What do you want?’. Read what they've said and then add to our woes using the form below.
Love what we’ve said? Think we’re talking nonsense? Don’t worry about being polite, just let us have it. We’re not afraid of telling you that you’re talking crap, so don’t be afraid of telling us the same.
While I agree with your sentiments, you may find that users find it hard to seperate what they want and what they expect - they simply want something that they can use with the least hassle, and if their expectations are met then it will be, even if the solution is not the most elegant in the long run.
Without being able to name specifics, I am sure there is a massive list of great products, innovations and ideas somewhere that all failed despite being the best solution possible, simply because users found that it was too different to what they already had to bother making the effort to adopt to.
Like everything else in the world, the route followed will be the easiest, not the best.