I gotta say, I was *very* happy to see the Spotter in Pharo 4. This approach, the grand unified auto-completing fuzzy find-by-anything box, is one of the greatest programming UX advances in recent memory. There's a reason there's now a version of it in every editor or IDE I know of. That's not to say there's no place for a Browser approach. But what's interesting to me is that I think the Spotter represents the 80% case. 80% of the time, we have a good idea of what we're looking for and we just want to get to it quickly. Browsers are more helpful for the 20% of the time when we're new to the code and we're exploring. On Tue, May 19, 2015 at 12:42 AM Joachim Tuchel <jtuchel@objektfabrik.de> wrote:

We're in agreement here. It is nog about the panes per se. It's about the fact that once you reached an artefact to read or edit, there is no need for the navigation to cover a lot of space on screen. Multiplied by the number of open browsers. Until you commence your journey through your code.

Then there is the problem of navigation being much more than just packages and class hierarchies. There is the question of why and how did I get here? That area is uncovered in most IDEs - not only Smalltalk.

Another problem area is the longevity of a need to have some artefact open. How often do we close 10, 15, 20 inspectors at once, because we don't need them any more? When we look for the place where something started, we travel from implementor to implementor, via some senders until we finally get to where we can answer our question af hand. Sometimes we reach a dead end, need to go back a step or two and try another branch. But does it really mean we need each stop of the path as an open window? Forever? Or are there better ways to navigate?

I am getting so off-topic it hurts, but these questions need an answer, and the Smalltalk community should be one that can solve them much more easily than others because our language lets us concentrate on the problem.

Joachim