Learning R - progression path from apprentice to guru

Learning R - progression path from apprentice to guru

Inspired by this thread on stackoverflow about learning Python, I want to
ask about how to constantly improve my R programming skills. I am a
stat-focused social science PhD who has fallen in love with R for a year
and who is looking to push my R skills to the level of package writers.
While I understand the importance of "learning by doing," I also feel that
we need a road map to even know what is possible to do, and hence, learn.
(For example, I wish I knew about ggplot2 / plyr much sooner). Another
important reason for a road map is that I want to have a vision of good
coding practices / techniques rather than learning inefficiently by fixing
each bug / inelegant piece of code.
Like the stackoverflow thread,
I don't want to know how to QUICKLY learn R
Nor do I want to find out the best way to get acquainted with the language
Finally, I don't want to know a 'one trick that does it all' approach.
A good road map may look like this:
Read this, pay attention to that kind of details
Code for so manytime/problems/lines of code
Then, read this (eg: this or that book), but this time, pay attention to this
Tackle a few real-life problems
Then, proceed to reading Y.
Be sure to grasp these concepts
Code for X time
Come back to such and such basics or move further to...
I'm sure that, like me, many are more than happy to spend serious time
with R. We just need a road map for that journey.