Increasingly, free feeds from news organisations in Germany contain per article pay walls. These articles are hard to recognize ahead of time. Some of the protection is really weak, and can be circumvented with browser plugins (ROT13 in case of Spiegel Online after 1000 chars or so).
These articles read badly in the preview in Android reader. And it comes as a surprise that they are paywalled, often I cannot tell at all, e.g. from web interface.
What I would love to see, is the ability to be able to filter out these articles entirely. They have no tags, nothing to recognize, but I believe there are phrases in the text to look for. The trainer seems to not allow this though.
Let me know what to do. I was even thinking of mirroring the feeds and doing the filtering myself, but that feels kind of wasteful.