Excellent authors and editors do what is right and what is good for their point of view. That's why they are authors and editors; they are creating and structuring information for readers. When great authors are asked the question (as in the Paris Review series) "Whom do you write for?" the usual answer is "myself." Or, similarly, Paul Rand, the great designer, offered his clients a choice: they could take it or leave it. This is my view. Some consider that this is elitist. Well that is what it takes to do good work. And besides I'm trying to make everyone part of the elite by writing and teaching!

Whether such noble thoughts apply to a newsletter is something to think about. But you do want to tell your audience new and provocative things that might not always please them--and many in your audience may welcome something fresh. So display some intellectual leadership. And exactly how much are you willing to pander?

Of course you need to think of your audience and also to pass some kind of market test. But why not let people vote with their subscriptions and not about every damn single article? Subscribers decide on the basis of the entire publication, and quite fully express their views by whether they continue to read the publication.

When you do such finely detailed market testing, you not only bring a lot of suspicious data into the editorial process but also, alas, a lot of marketing types. These marketeers will intervene in the editorial process--and corrupt the integrity of that process. Some in your audience will in turn detect the resulting lack of integrity and see through your pandering.

In addition, omnipresent rating systems contribute to an already over-developed sense of entitlement found among at least some users.

Finally, there are enormous problems with the data here. You have no sampling plan; you have, in fact, not a random sample but rather a sample of lunatics. The strong-minded, opinionated, touchy, and easily offended are much more likely to bother to respond to the questions. If you allow user comments along with the ratings, you may well alienate your writers by showing them crank mail and abusive flames about their work.

Are you going to let lunatics and marketeers micromanage your editorial policy?

-- Edward Tufte