New versions of Liferea refuse to parse any feed that fails to validate, even for relatively "minor" problems (the libxml2 recovery facility is no longer used; besides, it abandons the rest of the feed when it hits such problems). I don't want to use Google Reader, since I don't like the interface.
Typically bad feeds have things like high-bit chars or bare ampersands. Thankfully, there's a "conversion filter" feature that you can use to work around the bad feeds. On the two bad feeds, I run this filter:
[moz@pent ~]$ cat bin/fix-ampersands
sed 's/\o226/&/g' | sed 's/& /\&/g' | sed 's/\o243/GBP/g'