> Let's see if I understand... > > xdocs/index.html > index.rss|index.atom > > You want that index.html contains a link in the header to the feed files. > > Is this correct? Correct. > Some thoughts: > > In forrest there should be only one file with a given name in a given > dir, as there should be only one source. FWIIW: The two I create are not xdocs, so: src/doc/content/xdocs/index.html src/doc/content/xdocs/index.rss|index.atom > xdocs/index.html > index-feed.rss|index-feed.atom If the above doesn't change your opinion I'm fine w/ changing my names. Heck rss.rss and atom.atom suit me fine. With auto-discovery a user ought never type them in... > Secondly, what's the relationship between index.html and the feeds? Can > one be generated from the other? If so, we can have the same source for > both. If not, the first comment applies. Let's get concrete: http://lsd.student.utwente.nl/gump/index.html http://lsd.student.utwente.nl/gump/index.rss http://lsd.student.utwente.nl/gump/index.atom They are all the same index (sorta) first one human readable, second two computer readable. BTW: Here is an example of such links. Feed aware browser know to find a feed for the page from these.