The New York Times has led a great discussion about excerpts calling out a few sites whose business is largely based on excerpts.   As we reported last November, the impact is meaningful with the average publisher missing out in over $150k of revenue a year.

The reaction has been been noble and passionate; however, the discussion is largely around fuzzy fair excerpting  concepts such as intent that can’t be measured.

We’d like to submit another concept — one that ultimately can be quantified:  Excerpts should be considered fair if the reader of the excerpt still has a reason to read the original article.

Here’s our proposal to measure whether an excerpt is fair.

1. The excerpt must contain a link. A no-brainer but you’d be surprised how many sites lift content without linking back.  In our studies, this ranges from 30-40%.
2. The excerpt must be less than 50% of the original article. The actual percentage is debatable, but it’s hard to argue that the average reader will want to click through to read the original article if over half of it has been excerpted.
3. The excerpt must be less than 100 words. Again, the number is debatable and probably too high for some blog posts, but when used in combination with the % of original article, it can provide a good guideline.

The real question is what to do when an excerpt is unfair – should Google be automatically informed to remove it from their index?   Should AdSense, DoubleClick and Yahoo be informed to pull ads from these excerpts?    What do you think?