Anyone who patrols recent changes should watch out for red flags in edit summaries. This particular problem rarely comes from vandals; in fact, some vandals write very honest edit summaries, such as "blanking the page", "replacing the page with crap" or bizarre or offensive statements. Though sometimes automatic edit summaries will betray the intent of the vandal when the vandal writes a misleading edit summary; but as automatic edit summaries are done by a robot, vandals can figure out ways to mislead the robot as well.
 | This page in a nutshell: Weasel words as well as POV can occur in edit summaries, too. The author(s) of this essay believe that writing NPOV edit summaries is as essential as in the articles themselves. |
This problem with edit summaries often comes from established users who want to push their own POVs (points of view). Also, there are well-meaning users who unintentionally write misleading edit summaries. For example, an editor may make huge changes to the article but summarize this as "tweak format". Hence, the kinds of edit summaries discussed here should encourage a look at the actual edit regardless of whether or not they come from established users.
Let it be clear that indiscriminate inclusionism is not being advocated. Wikipedia is not a "dumping ground" for random facts and thoughts. But indiscriminate deletionism is even more of a problem, because it makes it harder for the community to have calm, thoughtful discussions as to what belongs and what doesn't.
One view is that Wikipedia should contain:
- Everything a typical person would expect in a general reference work, and
- Nothing a typical person would only expect in a specialized reference work.
For more general advice on edit summaries, see WP:EDSUM.
On the face of it, that sounds good. "Deleting trivia." Deleting the unimportant, the inconsequential. The colour of the napkins at a celebrity wedding.
There are several problems with this, the first being that not everyone agrees on what constitutes trivia. When someone "deletes trivia", without giving any more specific explanation, it is possible that they're trying to skirt consensus and remove facts which most other editors agree belong and are relevant to the article in question. Depending on the wedding in question, if WP:reliable sources state that the colour of the napkins had some important symbolism, then reporting the colour of the napkins and giving a quick sense of the commentary regarding this hue may be pertinent to the reader.
Let's get some perspective on trivia. Originally, there was the "trivium", which consisted of grammar, logic and rhetoric; and the quadrivium, which consisted of arithmetic, geometry, music, and astronomy. The trivium was supposed to be easier than the quadrivium, and eventually the word "trivial" came to mean, through semantic drift: "unimportant."
What is unimportant? And to what is something unimportant? Let's say, for the sake of argument, that the only thing that matters is survival. I'm not the only one who thinks Wikipedia should not be used for medical diagnosis, nor for operating a nuclear reactor. From this point of view, then
Everything in Wikipedia is trivial!
But surely also I'm not the only person who thinks Wikipedia content should not be deleted, even if it can't be used as a reference in hospital surgeries and nuclear power plants.
Instead of using a vague, yet loaded word, like "trivia", it would be more productive to expend just a few more keystrokes explaining what was actually done. For example, an article quotes a letter from Charles Darwin, and someone removes it, instead of just putting "Deleting trivia" in the edit summary, it would be far better to put "Deleting long Darwin letter because it adds too much detail that is not directly relevant to this article."
Note that Wikipedia has a policy, WP:TRIVIA, which provides ways to deal with information that might be relevant to the article but is currently presented as if it were a collection of random factoids.
A patroller should check first that the deleted section really was trivia, and second, that there was a better reason for deleting it, such as it being irrelevant to the article. A common target for this are "In popular culture" sections. Again, Wikipedia policy provides guidelines for rescuing suitable content from such sections.
The exception here is for unverified information about living persons. Other than that, unverified and/or poorly spelled or worded content can and should remain in Wikipedia so that if someone better equipped to verify it, spell-check it or reword it comes along, they can do so without having to dig in the article history. Remember what WP:PRESERVE says: "Fix problems if you can, flag them if you can't." For most unverified information, use a fact tag. See Template:Fact for more information on flagging unverified content.
This is a red flag, as no one owns Wikipedia articles, not the original writer, not the editors who have worked the most on the article, and not even Jimbo Wales.
Another red flag is if an editor states that changes must be discussed, but the issue itself is not raised on the talk page (note: this editor may have raised the issue in an archived discussion). If someone is going to demand that others discuss on the talk page before making any changes to an article, fairness demands that she too discuss it on the talk page. I.e., if one is going to use something like WP:BRD as an edit summary, a discussion should be started immediately by the person indicating such, and ideally the section should be linked in this same edit summary.
Don't get me wrong, vandalism is still a problem on Wikipedia. But it is much more of a mild annoyance, and not a significant threat to Wikipedia's credibility anymore. There never was any disagreement that vandalism is bad and should be reverted, compared to the great disagreement over trivia. So an edit summary that just says "reverting vandalism" should be an even bigger red flag than "removing trivia". As a compromise between accurate edit summaries and denying vandals the satisfaction of acknowledgement, the developers have programmed Wikipedia to provide a default edit summary for undoing vandalism along the lines of:
"Undid revision 0123456789 by User:ABCDEFGH to version 0123456788 by User:Example_One"
Anti-vandalism bots give similar edit summaries, also avoiding use of the word "vandal" or "vandalism". To see that word used in an edit summary most likely signals an attempt at deception: such an editor is hoping that those who would question a removal of "trivia" would not check up on a removal of "vandalism".
If an edit is not vandalism, then it should be an improvement to the article. Using the word "improving" therefore just wastes edit summary characters, and an edit summary only has a few more characters than a tweet. Also, one should suspect that there should be more to such an edit summary, along the lines of "to my liking, rather than to the misguided consensus." Similarly, but to a lesser degree, it goes without saying that an edit summary that isn't adding something should be fixing something, so "fixing" is often a sign of novice editors whose actions should be reviewed, or of vandals trying to cover their tracks.
This can happen either maliciously or innocently. A malicious use would be to include or delete a large chunk of text but pass it off as spelling corrections. But it can also happen innocently, such as when someone uses mass replace to expand abbreviations, or to create abbreviations. If you use mass replace in an edit, it wouldn't hurt for your edit summary to mention it!
Robots can be exempted from saying in edit summaries what the tags that were applied are. But people should be held to a higher standard when it comes to tagging. By actually saying what the tags are in the edit summary, it shows other editors that you actually gave some thought to what the tags say and that they really do apply to the article. See WP:RESPTAG for advice on choosing the most specific tag for a given problem.
It is common for novice, questionable and bogus edits to have a summary phrased in the first person. Normally one would omit the first person pronoun.
