Jeremy Huntwork wrote:
Of course once the build actually works and it has some definite structure, we can/will more readily employ the one page per commit method.
I don't understand why the status/state of the branch changes how you make changes to it. Surely after you do a test build you have a list of points that need to be fixed up. Why can't you fix each issue in separate commits? That way it's easier to see a) what was wrong and b) what was required to fix it up.
I think some of it was hesitancy to commit each change right away, which came from not quite seeing what was wrong, and making several changes quickly so we could test them - then, when we were sure about them, committing what we had.
However, point taken.
For my part, I'll make more of an effort to do one change at a time.
-- Jeremy Huntwork
-- http://linuxfromscratch.org/mailman/listinfo/lfs-book FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
