Why are substantial edits rejected when they don't also address all, potential minor, issues?

Question

Some reviewers of suggested edits appear way to restrictive when it comes to approving an edit.

See for example

• This suggested edit (Related meta post)
• And this one (Disclaimer, this one is mine)

The reviewer stats are also worrysome

• Scott has approved 2795 edit suggestions and rejected 1430 edit suggestions and improved 374 edit suggestions
• robinCTS has approved 27 edit suggestions and rejected 118 edit suggestions and improved 36 edit suggestions
• Run5k has approved 3215 edit suggestions and rejected 2579 edit suggestions and improved 301 edit suggestions

All three users rejected a large amount of edits, and only improved a small fraction.

A short sample of the reviews of those reviewers shows that many edits where rejected because they address not all issues of the post. It appears some reviewers try to educate the users to only perform complete substantial edits. Thing the reasoning behind that is flawed and actually harmful: Most editors are not able to spot all mistakes and flaws of a post. This restrictive approach will cause frustration and drive away potential contributors.

Has superuser a "Review suggested edits" rejection issue?

Answer 1

While I agree with your point in your answer of "Every edit improving a post should get accepted." the problem is that what constitutes an improvement is a subjective thing.

Just changing the phrasing of a title may not qualify as an "improvement" to a great deal of people. Yes, you changed the title to a complete grammatically correct question, but was it unclear or ambiguous to begin with? Not really as it was succinctly stating the topic of the problem.

This is also why we have the requirement of 3 people to review suggested edits. Group consensus rather than individual.

Many "old" users of Super User and the review queues still remember the old reject reason of "too minor" which stated

This edit is too minor; suggested edits should be substantive improvements addressing multiple issues in the post.

Back in the day this was an effective training that we should be rejecting these sorts of minor edits. That reason is gone now, but there has been little in the way of trying to pull people back away from that trained behaviour. Many still believe that edits should be substantive and useful and not cause unnecessary churn on the front page.

Granted on high volume sites like Stack Overflow edit churn is quickly lost in in the torrent of questions on the front page, but on lower volume sites it can make people wonder "why in the neck was this dragged back up" and leave them questioning whether the change really was an improvement at all as the cost of dragging it up weighs against the benefit of the edit.

To address your final sentence

Trivial edits are discouraged, but that does not mean that you have to reject them.

Not every user has the free time to click the "improve" or "reject and edit" buttons either. Many users are contributing a few spare moments of their time inbetween other things and even patrolling the suggested edit queue is an appreciated and commendable use of their time. If we lost all the people checking the queues in their brief spare moments then I suspect the queues would be significantly longer as a result.

Rejecting an edit is valid if it results in no quantifiable gain and trivial edits are generally rejected for a good reason. Just because you don't like seeing your edits rejected isn't a good reason.

Answer 2

Edits should make a post better. But what does "better" mean? It isn't a simple matter.

At one extreme, take a canonical post that gets heavy, ongoing traffic and contributes a really good question or answer. A single typo may be glaring and detract from every use. Polishing such a post with a minor fix can contribute value to the site.

At the other extreme, take a garbage post that not only contributes nothing of value, but also contains extensive errors of spelling, grammar, etc. Maybe it's a rant, or a completely misguided answer. Fixing all of the English mistakes still won't make it a useful post, so it isn't really an improvement.

Take another example, say a trivial question from 7 years ago that was of interest to only a few people and received an answer. But the matter is no longer relevant, and the thread has been buried and inactive since the day it was written. Someone does a search on a bad grammar pattern or common typo, looking for posts to polish, and stumbles across that thread. They could polish the heck out of it, turning it into pristine English, but the thread is no longer of use to anyone. Is it worth the time of three reviewers to review the edits, or worth bumping it to the main page where it will take oxygen from current questions?

One more example--a question that is closed or on-hold. Any edit automatically puts it in the reopen queue where five people will need to spend time reviewing the question to see whether it merits reopening. If an edit fixes things like spelling and grammar, but doesn't address the closure issue, a lot of people will waste time reviewing a question that will have better English when it is deleted. That isn't really an improvement.

"Better" should be from a site perspective, not just the text in the abstract.

Also, we get questions where the author has obviously not invested any effort before coming here asking other people to invest their time responding. Those questions aren't well received. If someone is going to do edits that three other people will have to spend time reviewing, the same logic should apply. They should put in the effort to fix as much as they can of what is needed. The push-back is generally when extensive fixes are obviously needed, but the editor just does a few items of low-hanging fruit.

Answer 3

I think that is the result of a mindset which robinCTS nicely demonstrates in this answer:

As a side note, suggested edits also need to be substantive. So, for example, if the only issue with a post is the typo "teh" (should have been "the"), a suggested edit to fix that will be rejected. (The underlying reason has to do with wasting reviewers' time and bumping the question to the top of the Questions Page, as explained further down.)

This reasoning is flawed. If the typo is the only flaw of the post, the edit should simply get accepted, as you do not safe any time by pressing rejected. If there are more flaws, consider improving the edit, someone will eventually have to do it anyway later. And if the reviewer improved the edit, then this is not "wasting the reviewers' time".

The stats of the three users given above indicate that they try to educate the users to only perform complete substantial edits. It is wishful thinking that this will cause the desired result. Most editors are not able to spot all mistakes and flaws of a post. This restrictive approach will cause frustration and drive away potential contributors.

Every edit improving a post should get accepted. The edit should simply get improved if there are more things to improve. Reject only if it is obviously a destructive change. Trivial edits are discouraged, but that does not mean that you have to reject them.