I noticed a great tweet earlier today pointing me at an article on UX Booth. I conceed that I don’t visit that site often enough – the articles on the site are always insightful and informative. Moreover they’re also chock-full of real-world examples.
Have a read of their blog post Handling user errors with care: Getting users back on track.
Whilst a number of the points are relating to how the UI will visualise the error information, I personally have taken two points from the article:
- Error descriptions are important. It’s too easy as a developer – particularly with the built-in validators in ASP.NET – to have a simple regex to validate an email address, then to complain that the email address fails with no guidance as to why. In this example, perhaps the approach should have multiple to validate the different parts that may fail, providing more accurate error descriptions should it fail.
- Empty required fields may need prompting. To quote the article: The site visitor has not forgotten this field, they are often reluctant to share this personal information and wants to try and move forward anonymously. Having no data is different from invalid data and it needs a different error message.
To this end I’m starting to look at how I can incorporate standardised validation objects (RegularExpressionValidator, etc) to hook in with much more flexible client-side libraries such as the jQuery validator library. It’s one I’ve used in the past but still don’t have an off-the-shelf library I can use. Once this is sorted and has been used on a few sites I may look to roll the code into my common libraries.
Edit: The tweet was from userfocus – if you don’t follow them then you really should. Great synopsis of a variety of usability blogs.