Wikipedia:We shouldn't be able to figure out your opinions
Since Wikipedia's articles must be written from a neutral point of view, we shouldn't be able to figure out your opinions based solely on your edits. If we disregard talk page comments, user page bios, userboxes, or anything else outside of a mainspace edit, every user should be inscrutable.
Acceptable edits aren't always acceptable
Wikipedia articles should always be written from a neutral point-of-view. Some edits, particularly on contentious subjects, are easily identifiable as being partisan, and are usually removed swiftly. Sometimes, however, there might be a series of edits in which each individual edit is perfectly acceptable, but collectively could reveal an editor's personal views on the subject.
These editors tend not break any rules (and indeed do their best to work within Wikipedia's rules and guidelines), but it can be clear from studying their long-term editing patterns that articles under their care eventually skew towards a certain viewpoint. It might not even be deliberate or malicious by the user – in some cases a user's personal opinions might creep through into their edits more than average.
Example
Consider, for example, an editor who makes an edit to the article for the fictional ideology of "anarcho-statism", stating that it has been criticised for being contradictory. The editor provides a reference to support this assertion, which complies with Wikipedia's rules for verifiability. So far, so good. The editor then makes another edit, saying that human rights tend to be diminished under anarcho-statism. Again, as required, the editor adds a reference. In the next edit, the editor copyedits a sentence that discusses positive aspects of anarcho-statism, drastically reducing the sentence's length. Finally, the editor completely removes a paragraph discussing people who support the ideology, calling it "undue weight".
Individually, none of these edits breaks any of Wikipedia's rules, and might be perfectly acceptable in isolation. However, these four edits seem to indicate a broader pattern: this user probably opposes anarcho-statism and is here to paint it in a bad light.
Be inscrutable
It's arguable if the above example can be considered malicious or bad-faith editing. It's certainly not vandalism, and it can't easily be called disruptive. The user will likely justify their edits as being reasonable and in line with the project's guidelines − and they might even be right.
What can be more concerning is if a low-traffic article (and so, one that isn't on many people's watchlists) is edited long-term by such a user. What you might see then is a top-to-bottom rewrite of an article which paints the subject in an entirely different light and which hasn't been challenged along the way by other editors.
A single snapshot of such users' edits don't reveal much, but put together we begin to understand their biases, which are filtering through to their edits. And so, as a tip to all editors:
Be inscrutable.
See also