-* \[[!inline pages="..." trail=yes]] currently tries to work out
- what pages are in the trail, and their order, at scan time. That
- won't work, because matching a pagespec at scan time is
- unreliable - pages we want might not have been scanned yet! I
- haven't worked out a solution for this. I think
- \[[!inline pagenames="..." trail=yes]]
- ([[patch_required|todo/inline_plugin:_specifying_ordered_page_names]])
- would be safe, though.
+Demo:
+
+* [in use on entries in my blog](http://smcv.pseudorandom.co.uk/)
+* [a demo trail based on links](http://demo.hosted.pseudorandom.co.uk/trail/)
+* [a demo hybrid trail/inline](http://demo.hosted.pseudorandom.co.uk/trail2/)
+
+The page `e` is in both demo trails, to demonstrate how a page in more than
+one trail looks.
+
+The `smcv/trail2` branch is an older version of `trail3` which used typed links
+as its data structure, resulting in timing-related limitations (it couldn't
+select pages for the trail by using pagespecs, because pagespecs can't be
+evaluated correctly until the scan stage has finished).
+
+Updated, November 2011 (`trail3`):
+
+* reinstated `inline` integration ([[report]] integration would probably be
+ pretty easy too, if this gets merged)
+* switched from typed links back to a custom data structure to avoid
+ chicken/egg problems with ordering
+* create typed links too, as a side-effect, but not when using an inline
+* regression test with nearly full coverage
+* CSS for the default anti-theme and all built-in themes (it looks nicest
+ in the default anti-theme and in actiontabs - the demo uses actiontabs)
+
+Updated, March 2012 (`trail3-integrated`):
+
+* replaced `\[[!trailinline]]` with `\[[!inline trail=yes]]`
+* added a `build_affected` hook so it doesn't have to use `inject`
+ (optional commit, can be omitted)
+
+Known bugs:
+
+* the blueview and goldtype CSS nearly work, but the alignment is a bit off