Monday, October 22, 2007

Design in exceptions

Thinking of the project/products I have worked on I realize that the real design was not in designing for normal cases but in “exceptions”. Imagine yourself designing a Search Page.

User types in a query = result found. Query Page + Search Results page. Job well done!!!

Simple isn’t it?

But look at the broader picture – lets say there are cases where the query is not exactly found. What do we do then? How will the system behave?

Should it show the results along with the message? Or should we not show irrelevant results and ask few more questions from the user? And also when should we do that – if the relevance is 75 or 50 or 25?

These are the complication of design. How do you deal with these numerous cases in your produce really defines how comprehensively have you thought about it. Normal cases are easier to handle they don’t take so much time and your mental resources.

Thus design is in exceptional cases. You will design with an approach – every thing will go well but suddenly there will be one case where this approach will crumble. How smartly you deal with it really defines how well the product get designed.

Wednesday, October 10, 2007

Is it worth?

I was in Sunnyvale few weeks back and I was having a discussion with Prasad over the lunch. Prasad mentioned something interesting about discussing design.

Most often we land up discussing with PMs or Engineers or other designers and the discussions land up in conflicts say about changing the link color from blue to black. Now it’s difficult to argue especially with PMs who have data to back up their claims most of the time. So as designer you loose the argument unless you have the data or a design research to back your point.

Prasad mentioned something interesting. He said when you discuss these issues also look for a new parameter “If I change this how much impact is this going to create to the users”; “Is it impacting a substantial amount of our user base or just 50 of them?”; “If its just 50 user that this change is going to impact, then is it worth doing it?”

It’s a very crucial point if you look at it deeply. Not only this ‘parameter’ can be used in design discussion but I guess also effectively during ‘designing’. It is worthwhile to look at our design and ask – is this change going to impact a large user base? Should I focus on this problem or shift to other problem which can have a greater impact?

Nice and important point.