Having read this question I have to raise doubts what essential difference there is between posting elementary research on skeptics.SE or publishing it on some random blog and next referencing it here.

In the end, expertise is required to judge research (no matter where it's from) for its quality; that's what being a skeptic is all about.

Claiming that a skeptic is just about being an expert in referencing would disqualify a lot of the answers on this site, because a lot of answers are simply referencing logically solid but technically unauthoritative sources (if this is disputed I can compile a list).

On top of that a lot of answers are all about the interpretation of external data (e.g. this answer) which is also nothing more or less than original research, no matter how many sources for individual facts you cite.

What I am trying to make clear is that on one hand the original research policy from a skeptics point of view is quite poor and from a SE point of view is quite vague. And when policies are vague and don't have clear boundaries at the very least mods shouldn't delete answers left and right, but rather the community should be allowed to vote their agreement or disagreement.

Additionally there is a very simple alternative that already now applied to some extend: Instead of counting references, judging whether answers are checkable. After all, verifiable, checkable answers is the thing that should be the life and blood of a skeptic. And if you don't believe an answer is verifiable: downvote and comment why you think so, that way others will downvote as well! Which after all a lot of users agree with.

Which for that matter is another problem this SE has, the extend to which the mods use their powers is totally disproportionate compared to other SE giving this site far less of a community feeling (which at least for me is always one of the reasons I try to stay away from posting on this site).

Oh well, just some thoughts for you guys to consider, do with them whatever you wish.


2 Answers 2


In this case I think the answer did provide additional evidence that's helpful and I would support undeleting it. Research itself isn't bad and questions about whether or not a image is real just don't have peer reviewed papers that you can cite.

  • I would support undeletion if it did show that one original image was compressed twice and the other original was compressed once. I'm not sure how we can prove that either is the original image and thus that this analysis is meaningful. If you think of it, showing that the images are the original is basically the whole answer.
    – Sklivvz
    Commented Sep 1, 2014 at 7:47
  • 3
    @Sklivvz : We don't have to show that either is the original. The answer is about whether a certain picture is the original. Showing that the picture isn't the original is enough. If you edit an image and then then export as .jpeg you add an additional step of compression.
    – Christian
    Commented Sep 1, 2014 at 8:06
  • 1
    Let's say that the original is the "we kill children" one and it's a tiff image. The it gets shopped with the Star of David and saved as JPEG. Another copy of the original gets saved and compressed twice as JPEG. This is another possibility compatible with the findings.
    – Sklivvz
    Commented Sep 1, 2014 at 8:20
  • 4
    In a case like this a single piece of evidence is not enough to make a decision about whether which image is real. To make that decision we need to look at multiple pieces of evidence. The answer provides an additional piece of evidence that's useful. The answer also passes the "more evidence than the question" test.
    – Christian
    Commented Sep 2, 2014 at 17:03

If I can use your deleted answer as an example, you were providing personal expertise, using an online photo analysis tool.

The problems with that are:

  • that we can't verify whether the tool is reliable
  • that your interpretation of the results is reliable

What we can verify is that given the photos we get similar results with it. Also, you could provide evidence that the tool is a valid tool to use -- this is not normally necessary with more notoriously reliable sources of course.

That leaves out the last point: your interpretation. Right or wrong, it's an argument from authority. If you were to write a peer reviewed paper, you would be asked to reference it.

Even then, a peer reviewed paper is... Reviewed by people with the appropriate expertise. They effectively vouch that your research is at least viable. The problem is that only a minority of people know forensic analysis here to be able to judge.

More in general, we need answers to be verifiable and that our community of expert be able to judge the quality of it.

We are a community of experts in assessing authoritativeness, not photography.

If you point at a blog, your answer will normally be down voted, but not necessarily deleted. The reference is not authoritative. If you point to your own blog it would be deleted as self promotion, maybe. If the answer is massively down voted it could be deleted because of that.

Finally regarding moderation here: yes. Mods are hands on, they delete stuff. This site has been created with this specific intent by SE. We are different, and proudly so. There is no other way that a site where everything is based on references would work.

This said: stuff gets deleted on all sites. I have had questions and answers with hundreds of up votes deleted on stack overflow, where moderation is certainly much lighter than here.

  • 1
    How does DVK's answer not suffer the same problems?
    – user5582
    Commented Aug 20, 2014 at 21:59
  • It points to primary evidence. It has the original image!
    – Sklivvz
    Commented Aug 20, 2014 at 22:01
  • 1
    It argues that that is the original image, but how are we to know? We have to accept the analysis that DVK presented.
    – user5582
    Commented Aug 20, 2014 at 22:02
  • No, it has an image we can judge to be strong or weak evidence. He makes a case on why it is strong. There is no doubt that the image is evidence. The other answer simply applies expertise to prove one is the original. It doesn't provide any evidence that it is.
    – Sklivvz
    Commented Aug 20, 2014 at 22:05
  • 2
    How can you say that the results of the tool is not evidence? It has the output from fotoforensics that we can judge to be strong or weak evidence. He makes a case on why it is strong. There is no doubt the tool's results are evidence. It seems you just don't understand the results of that tool.
    – user5582
    Commented Aug 20, 2014 at 22:06
  • 1
    Remove @DVK's answer and the other one would not stand on its own feet. I do understand the results, what do they prove, besides that one of the two images is more/badly/doubly compressed?
    – Sklivvz
    Commented Aug 20, 2014 at 22:10
  • 1
    He makes a speculative deduction based on the output. That deduction is not supported by evidence, as both @Oddthinking and I remarked.
    – Sklivvz
    Commented Aug 20, 2014 at 22:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .