Prefer Excel spreadsheets with FIT/FitLibrary

Since the invention of  FIT, now called FitLibrary, in 2002 a couple of new acceptance test frameworks with improved features appeared on the scene like FitNesse or Concordion.

But FitLibrary still remains a serious candidate to me, because tests can be written with Microsoft Word or Excel. Domain experts of the customer are usually accustomed to Word or Excel. I don't think you can convince a domain expert with no IT background to write or comment test specifications in Wiki markup like in FitNesse. Besides, you'd have to set up and run a FitNesse server that can be accessed by your acceptance testers or customer. With FitLibrary test documents can easily be exchanged by e-mail, e.g.

FitLibrary supports two document formats for test specifications: HTML or Microsoft Excel.
If you consider to use FitLibrary for acceptance tests, you shouldn't write your tests with Microsoft Word and save them as HTML file, because there are at least two pitfalls:

1.  If the cell's content gets too long, Word will insert a newline character somewhere in the middle of the cell's content, which you won't see in Word or a Web browser, but will likely break your test. In some cases you can work around this in your fixture code and strip off newline characters from the expected value. I found this behaviour of arbitrarily inserted newline characters in SeaMonkey's HTML editor, too.
2. If the rows of a table aren't exactly the same width, Word will insert placeholder cells or sort of compensation rows, which aren't visible. This might produce red cells in FIT's test reports.