On 03/16/2011 02:21 PM, Scott Kitterman wrote:
On Wednesday, March 16, 2011 09:01:59 am Thorsten Wilms wrote:
Sometimes the problem may be certain users stubbornness rather than anything else, especially if you design for the long term. So the answer may have to be wrapped up in a strategy to "sell" it.It depends on how important your current user base is to you. Particularly when there are alternatives available, such radical restructuring is more likely to result in existing users leaving than them immediately jumping to the new paradigm (see KDE 4.0 for example). It may be that you'll get them back in the long run, but many will be gone for good.
That's exactly why I mentioned strategy. To expand, you would ideally design where you want to end up (caveat: any "end" will only be a step, once you are there or past it), then determine how there may be a transition and how to communicate the benefits.
*Luckily*, the implementation effort tends to be so huge, that you have to go step by step, anyway. I think more thought should be put into tools and processes, so design experimentation can be accelerated and opened up to more people. Everyone who hasn't yet, should take a good look at a Smalltalk implementation like Pharo. If we could have that level of malleability at runtime, without being trapped in an image file ...
-- Thorsten Wilms thorwil's design for free software: http://thorwil.wordpress.com/