About the SA Blog Network



The art of science and the science of art.
Symbiartic HomeAboutContact

Mash-Up This! Science Communication’s Image Problem

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

The promised Information Economy based on creative culture is a sweatshop.

Award winning illustrators, fine artists, photographers, cartoonists and animators are routinely ripped off, mashed-up, and reshared without attribution, let alone money. “But it’s always been this way!” “Good luck changing the whole internet!”  It wasn’t supposed to be this way, and creators don’t have to put up with it.

Okay, to start talking about copyright, let’s all agree to this: there’s law and then there’s culture. The internet is a culture-changing dynamo for communication and some laws haven’t kept up. I’m an illustrator and a  fan of Creative Commons myself. But that doesn’t mean I should treat other artists’ work the way I hope people will treat mine. I have to respect their boundaries.

As members of the online culture, we don’t have to accept that image theft will always be the dominant way of sharing visual information online: culture matures. Expectations change. But right now, large portions of science communication online are part of the problem. Science bloggers, willing to cite scientific papers with utmost care, rip and run with images grabbed from Google Image Search and Wikipedia.

Compound Eye Vs. I F*cking Love Science

Yesterday, my fellow SciAm blogger, entomologist and photographer Alex Wild became fed up with the popular (over 5 millions fans) I F*cking Love Science page when its creator ripped off one of his images and shared it without attribution. His post, Facebook’s “I F*cking Love Science” does not f*cking love artists kicked over a hornet’s nest online. In it, Alex took a look at the most recent 100 images, and how many had attribution and found it appalling. I agree. Here are my thoughts about the controversy.

  • Very recently, I F*cking Love Science’s creator came under a lot of nasty, horrible, all-too-common sexism and misogyny when fans found out the popular page was run by a woman. It’s not right, and emotions are high in defense of Elise Andrew in the face of the outrageous attacks. But that’s not what Wild is doing: his post was a rant, but it was a measured one about a common practice on IFLS, not about Andrew herself.
  • One of the key reasons for sharing and discussing this publicly is precisely to raise consciousness about proper image use with the fans of the IFLS page. I’ve done this before, and it often works. I’ve tried it in the Pharyngula comment thread, on DeviantArt, and on paleo-art blogs. You’re not just having a conversation with the site author: you’re having a conversation with anyone looking on.
  • Accusing Wild of being rude by not quietly contacted Andrew is placing the blame in the wrong place: IFLS was apparently not trying too hard to contact any of the artists before using their work.
  • Generally, I like the idea of IFLS: it’s the one science feed most of my non-sciencey friends on Facebook might repost from. It’s good outreach. But in this day and age of attribution, and with IFLS being treated like a business there’s no excuse. This is not idle sharing among friends.

Personal Brand and Insulting People

Something both wonderful and hindering about our online culture is how personalities get tied up in brands. Most art commissions I have received are not only because I’m up to the skill level of the illustration: they’re also about my blog and personality. People want art because of the artist, moreso than ever. How many people felt they had some personal connection with Picasso, with O’Keefe? With how easy it is to reach out via Twitter, Facebook or old-fashioned email, it doesn’t compare.

The downside of this, all this pouring of our artistic and personal selves online, is whenever someone who has made something is criticised, we take it personally.

Time to Take a Look at Science Blogs

I did something like this back in 2010, and Alex Wild did excellent work yesterday looking at the past 100 images on IFLS for attribution. I’ve used and looked at the most recent posts on each of the networks. Let’s see how blog posts on major networks shake out:

Sample from 1am to 7am 24 April, 2013. Click to enlarge. Yes, you may reproduce this, but please link back to this blog post.

The legend is in the first set: red means no attribution, yellow means credit without a link, and blue is the good one, credit + link. We should be seeing lots of blue. I sampled 10 recent posts from 10 networks: they may not all add up to 10, because some blog posts didn’t use images at all.

All the samples were taken between 1am and 7am this morning. I have kept a list of each url which I’ve forwarded to Bora Zivkovic, our editor.  If anyone is dying to see it, lemme know.

Caveat: I know this isn’t scientific: it’s a sliver, a glimpse in time. I looked at 100 science blog posts from 10 networks. Perhaps a wider sample over 2 years would reveal something more comprehensive. At the same time, this tiny sample, slightly more than an anecdote tells some stories.

I know that at SciAm, we discuss proper image use a lot. So it’s nice to see that it’s working. I  may guess some networks like Phenomena have a mandate not to link to image sources. Perhaps to keep people on the site? But again, 10 blog posts from that network is hardly representative of their output.

Over the past several years, I have stopped reading some science blogs and even avoided some of the people at conferences for their lack of care about image credit. Maybe I should be opening a dialogue at every turn, but it’s hard balancing the importance of the issue without being self-proclaimed internet police.

All is Not Lost

Online culture has been moving to better and better attribution models while at the same time it has been easier to rip off more and more media. A particular case of frission is with Pinterest, the site that allows pinning from any website with lovely links directly back to the source but who’s Terms of Service say you should only post things which you have permission to use – something that would kill their whole site if it was actually followed. (Kalliopi Monoyios and I wrote about Pinterest’s strange Terms of Service here, here, and the resolution here.)

Yet: retweeting; resharing on Tumblr, G+ and Facebook; pinning linked-back images on Pinterest – all of these are ways that attribution can be traced. They’re usually not sufficient on their own, granted. And attribution this way can be abused: Tumblr doesn’t demand a source link, it’s optional; you can pin from search engines, thwarting the link-back on Pinterest; but they represent attempts to give credit where credit is due.

Tools like Google Reverse Image Search, and the new Imgembed (which I look forward to exploring) help artists protect their visuals, and help anyone search for the right author. (You can usually find the correct image creator in seconds.) Art and illustration college and university programs really need to start teaching students how to manage their online presence.


The problem is bigger than I F*cking Love Science, but that’s a big site and it ticked off a photographer with a strong online following. And so the messy discussions of the past couple of days. A lot of what I’ve said in this post I’ve said elsewhere. We’re going to keep having this conversation until the tools get easier to use and the culture shifts back to appreciating images, and not treating them like bubble gum cards.

The internet is the most powerful visual medium in history, while making and sharing images is easier than at any other time in history. We need to respect these two powerful visual forces.

- -

Further reading:

[A lot of people have informed and challenged my thinking about these issues over the years, and so with the trepidation of someone worried he's forgetting someone, I'd like to thank Eric Orchard, Bora Zivkovic, Kalliopi Monoyios, Alex Wild, Katie McKissick, Cory Doctorow and Matthew Inman.]
“Searching for Copyright” images were derived from a digital painting I made and then manipulated with the Halftone app. © Glendon Mellow, feel free to reuse.

Glendon Mellow About the Author: Glendon Mellow is a fine artist, illustrator and tattoo designer working in oil and digital media based in Toronto, Canada. He tweets @FlyingTrilobite and is on Instagram. You can see Glendon's work-in-progress at The Flying Trilobite blog and portfolio at Follow on Twitter @symbiartic.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 14 Comments

Add Comment
  1. 1. beatrice 12:08 pm 04/24/2013

    Thanks for the recap, Glendon. The internet has brought about many cultural shifts. Next up: caring about where your f*cking images came from! :D

    Link to this
  2. 2. Glendon Mellow 12:18 pm 04/24/2013

    Thank you, Beatrice. I know this ended up really long, but amidst the problems, there are signs of hope for proper attribution: but it’s going to take culture as well as technology to work. Your piece about YouTube gave me a lot of food for thought.

    Link to this
  3. 3. Glendon Mellow 12:49 pm 04/24/2013

    I should clarify something else: The thing that makes me happy about Imgembed and Tineye (and even Klout and Kred) is that it means there’s money to be made in preserving attribution to people. It pays off to find the original innovators.

    Link to this
  4. 4. bgriswold 7:06 pm 04/24/2013

    I will be very interested in your thoughts on Imgembed. Looks like a solution for this that want to share, but not for those interested in protecting themselves from sharing without permission. So it might help a subset of artists into blogging. I am still trying to figure out how it helps other artists…

    Link to this
  5. 5. edyong209 9:19 pm 04/24/2013

    For the record, Phenomena has no mandate to avoid linking out. Our posts are linked up as per usual. The main reason for the lack of links is that the template seemed to be stopping us from linking from the captions in the main images. I *think* we’ve sorted that out now, so shouldn’t be a problem in the future.

    That aside, thanks for putting together this valuable analysis. It’s a problem, and one that should be trivial to fix. I’m proud that Phenomena has nothing in the red at least – we can build on that.

    Link to this
  6. 6. Glendon Mellow 6:56 am 04/25/2013

    bgriswold, Imgembed does look potentially pretty useful. I don’t know when I’ll be writing about it. In the meantime, Matt Shipman directed me to this comprehensive review.

    Thanks Ed. I didn’t think National Geographic would be too worried about bounce rates. And nothing in the red is excellent in my books. I tried to stress above that ultimately this sample is too small: I mean *I* don’t post on Symbiartic every day and we have 3 bloggers. None of the data had my own posts. I looked at 10 recent posts per network on a particular day.

    The other issue too is that these are all science blogs that are on networks. There are tons of excellent high-quality independents such as Deep Sea News. Would a larger survey see networked blogs in better shape than independents? Conceivably there are network standards, but I don’t know.

    So why use numbers at all? Because holding up a mirror, even a tiny warped one, gets people to pay attention.

    Link to this
  7. 7. scicurious 8:47 am 04/25/2013

    Thanks for this Glendon! I’ve really thought about linking images over the past two years, all because of you. I hope I’ve gotten better!

    I had a question about figures. I post figures from the papers I talk about all the time, usually because they are funded with public monies and in the public domain. I don’t directly cite UNDER the figure, because it’s part of the paper, and I cite the paper at the end and the beginning. Should I also cite under the paper with a link to the figure when I can get access? Or is where I cite right now enough?

    Link to this
  8. 8. Glendon Mellow 10:13 am 04/25/2013

    Thanks Sci! One thing everyone should remember is though I write about this stuff, there’s no hard and fast rule. I’m trying to look for image attribution that is
    1. There at all
    2. accurate and clear, usually name and a link
    3. respectful of both the image creator and readers who may want to learn more.

    When it was obvious form the text that figures and diagrams were form papers, I counted those as blue in my data above. It would be helpful just to note in the caption, “image from paper-name, researchers’ names et al” in a similar way to a reference. It takes the guess work out when scanning through images for anyone who might want to reference the visual later.

    Link to this
  9. 9. NerdyChristie 1:28 pm 04/25/2013

    Glendon — how does this apply to images bought from a site like iStockPhoto or 123RF? Do you link to where you bought it?

    Link to this
  10. 10. Glendon Mellow 1:49 pm 04/25/2013

    Christie, I think linking to stock images is important for a couple of reasons.
    1. it shows transparency and that it is where you bought it (as in, it lets readers know you didn’t rip it off)
    2. It’s advertising for stock photo sites (I’m not a fan of those sites but they gotta pay their employees too)

    Sometimes there can be a stylistic reason not to put the image credit in the caption and to put it at the bottom of the post with the other references and links. Scicurious in particular is a master of the funny image caption, and credit in the caption would break the flow. Easily finding the credit below, at the bottom of the post would seem to me to be a good trade off.

    Link to this
  11. 11. Lunar Camel Co. 10:06 pm 04/25/2013

    I’m so happy to see someone writing about this in a clear and thoughtful way and backing it up with a bit of research. I’ve been blogging on and off since 2006 and I’ve heard all sorts of crap excuses over the years from people using my photos without attribution. The one I’ve probably heard most often is from people who took the time to embed one of my photos from Flickr but not to actually read my Creative Commons license, which prohibits commercial use: “But I don’t get paid.” They’re writing for sites that are covered with ads, sometimes even sites affiliated with print magazines, but because they personally don’t get paid, they think it’s not commercial use. I will never not be amazed by how ignorant that is.

    As for imgembed, this is the first I’ve heard of it. It looks interesting but there are aspects of it that bother me. FIrst, they’re taking 30% of revenue and that’s a lot. That’s more than what an agent’s cut would be if they were negotiating a deal for a photographer, no? It’s not hard to justify paying that much in exchange for actual representation, for an actual service being performed by someone looking out for your interests, but for a digital clearinghouse that seems excessive. Second, they appear to have built it into their model that anything that gets less than 10,000 clicks is free. If 9,000 culturally important people see something, that just doesn’t count? That seems arbitrary, and, perhaps more importantly, where is the analogue in IP law? How many people have seen the Richard Prince paintings (or the Patrick Cariou photos they’re based on) that are the subject of the lawsuit covered in today’s New York Times? There’s no such arbitrary number of viewers that’s part of the court’s analysis in a case like that. In effect imgembed is asking copyright holders to give away rights that the law does not require them to give away.

    Link to this
  12. 12. Glendon Mellow 7:35 am 04/26/2013

    Thanks Lunar Camel!

    I have some trepidation about Imgembed too – but I’ll try to check it out from the inside.

    Link to this
  13. 13. paigekbrown 1:37 pm 03/20/2014

    I just noticed that SciLogs is in there for no attribution! I work very hard to provide attribution for images – I hope we do better in the future!

    Link to this
  14. 14. Glendon Mellow 1:45 pm 03/20/2014

    Paige, remember: this is a year old and an extremely small sample size!

    Entirely possible that had I done this on a different day it would have looked very different. Perhaps a new post checking in is a good idea?

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article