Some of the images on Wikipedia and Wikimedia Commons are public domain works, but many are not and can only be used if proper attribution is given.

If you use a copied image on your site, you should be honest and provide a link to the page you copied it from, even if it's in the public domain.

But as far as I know, all the thieves who steal images and use them without permission or attribution usually don't get punished by Google or anyone else, at least at the present time.

Use of images is quite a minefield for SEO.

I use some wikipedia images, always credited to the original picture even when GNU, but I'm reducing them as part of our 'post-panda improvements' because I suspect they could be used as a clue by google that a site has been created using wiki information rather than having original content.

There are several sites which have loads of great photos at low prices so I'm using them for replacements when possible and I can't take the photo myself.

That said, it is often suggested that there is a case for linking to authority sites as a way of improving a sites SEO, so it is possible that removing the links to wikipedia is a bad thing.

That is without even starting on the issue of whether having images is a good thing because it enhances user experience - or a bad thing because it slows down pageload time (I have a travel site and choose to use lots of images despite the slower page load times, but I'm pretty sure it counts against me - but hopefully the pages attract more links because of it)

As aristotle says, the most common and most successful option often seems to be just to steal pictures without giving credit, and I haven't seen any evidence at all that sites stealing our own images are penalised in any way. I'm not recommending it though!

Overall it's a tough call, I'll be interested to hear other thoughts

As aristotle says, the most common and most successful option often seems to be just to steal pictures without giving credit, and I haven't seen any evidence at all that sites stealing our own images are penalised in any way.

Do you think so? I am somehow getting the feeling that google has included something of this sort as one of their quality signals, but they are somehow making a mistake in determining the right owner.

I never look at my image results much but just have, prompted by your question.

The three first examples I looked at were showing my images on other sites (ie stolen) on page 1, but not my own images. So I think you're right that they aren't attributing them correctly.

Another possibility is that sites hit by panda are losing both page AND image rankings - my originals of the three images above are on my main site which was hit by panda, but further testing suggests that images on my 'non-panda' sites are still showing correctly in the results.

PS there is a kind of beautiful irony if my site was hit because it was slow-loading because of the images, yet even the images get penalised as a result...

If you use a copied image on your site, you should be honest and provide a link to the page you copied it from, even if it's in the public domain.

(was offline, sorry for late reply) I always attribute images, as a professional writer you won't find me copying anything... but don't agree with the above as is not necessary for these images, they are free use and for most of them you can't even find the actual photographers name for credit, they have a log-in name as attribute (their Google account).

I didn't think about being flagged as a 'Wikipedia content user', that's a good point but due to the amount of content on Wiki and extent of topics covered I'd think Google would have an incredibly difficult time making any connection between the info on Wiki and info on a site who may have used some in writing an article (not talking copying here, but as a source).

Google has a problem identifying duplicate content and not penalizing the original creator of the work (though of course the first article published is the original) from what I've read here, so not sure how effective they would be in trying to penalize folks who use Wiki as a source at times... or why they would even try. Duplicate content, copying Wiki content of course... even though it may be allowed by Wiki's Creative Commons ShareAlike License as long as is attributed.

This may be similar to many issues of a 'percentage' potential problem. If all images on a site were Wiki images Google would have motivation to penalize in support of those who create unique works but hopefully using a few doesn't mark you in any way.

I replace Wiki images when I can (travel to an area and take my own or use other original images).