How does a mistake happen, and how do you stop it happening again?

  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  

Last week I made a mistake, and quite a few people picked it up. I put an image of a lily where an orchid should have been. It seems a basic mistake so how did it happen?

Every post on this blog now should have an image associated with it. The reason we do this is that posts with images get more engagement (shares, likes, retweets, clicks) than posts without images. For more information see the infographic below. The way I look for images is first in the paper. If there’s a diagram, then I’m most likely to choose that as people like diagrams. After that, it’ll be a photo or a graph.

The problem with choosing a graph is that when you schedule a few posts at once, all graphs start to look the same. So sometimes I look for images from further afield. Wikipedia has images available for re-use. Flickr has a lot of images, but only some are available for re-use. These days I’m generally only comfortable with picking images where commercial use is allowed. The Annals of Botany Company that owns this blog is a charity. I could argue this is non-commercial use but, given so many Annals of Botany papers are behind a paywall for a year, it’s very debatable. This reliance on a commercial licence can make pickings very thin when you’re looking for something obscure.

In the case of the Corsican orchids, there were photos of Corsican landmarks I could have used, but they didn’t seem right. I found an image with an orchid and Corsica label and here’s where the mistake happened.

A lot of people suffer from imposter syndrome, the idea that you’re not at the same level as your peers. However, given I have zero botanical qualifications, I don’t think the term imposter syndrome applies to me. I thought the labelling was odd, but I know enough to know orchids can be odd too. That’s why I accepted the label I was given.

So, given it’s not an orchid, why have I left the image of the lily in the post?

I thought it was important to acknowledge the mistake. While we have the image, we also have a corrected caption and a note to Scott Zona’s comment at the bottom. I think this is a bit more honest than changing the image without acknowledgement. It also means the people commenting look like they know what they’re talking about instead of leaving them looking like they’re saying an image of an orchid is not an orchid. I’m also hoping that it might encourage other people to let me know when I get something else wrong.

Ironically, this was a problem that we identified in our recent meetings and we’ve already sorted out a fix. We have images for the Content Snapshots in the journal, but these are sized for the journal and far too small for use on the web. We’re changing the author guidelines for people submitting papers so that they’ll be asked for 1080px wide images so we can use them on the blog and social media. This takes time to work through, so there’s still plenty of time for me to mess up other image captions.

This is a solution, but there’s something in Scott Zona’s comment that makes me think we can do better.

Ah, the perils of getting a photo off Flickr.

This ties in with an earlier comment from Ian Street.

And IDK how possible this is, but if AoB were willing to start an image cache of plants that could be used by others, I think it’d be an amazing resource for the plant science blogging community to have pictures of the plants we’re trying to write about as that is one of the hardest things to source.

AoB may be in a position to do something.

I’m thinking to contact authors when they submit images to see if they’d be happy with them going up to an AoBBlog Flickr account under a BY-SA licence. If they are, I can upload them with a link back to the source paper. If there’s a clear species in the image, I could also machine tag the image with the relevant taxonomic data and put the photo into the Encylopaedia of Life image pool. This would make it searchable for anyone looking for a reliably identified image of that plant. The EOL pool is already impressively big, so I think AoB could be more help by encouraging scientists to contribute to that rather than setting up our own pool of photos.

Thank you to everyone who pointed out my mistake. I’m sure there will be others.* If you spot a problem then please let us know here.

As promised, here’s the information from MDG Advertising on images in posts. It’s long.

its-all-about-images-infographic_1000

* If I were placing a bet, my money would go on scheduling. Our automatic scheduling system broke down last week. The way we run alerts on social media now is overlapping cycles of posts and tweets, with links to other sites thrown into the mix. There’s a possibility for something to go impressively and very publicly wrong.
Update: I should add Buffer’s customer service have been a huge help in making a lot of these problems much easier to handle.


  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  •  
  • Great post, and a great idea on creating a bank of images for use.

    One tiny problem. It’s against the Flickr Terms of Service to upload an image to an account you control that you did not take yourself. Even if you have permission or even if you own the copyright on the image, if you did not click the shutter yourself, you’ve violated their terms of service.

    The photographer could add the image to the EOL pool themselves and you could tag it with something that will help you find images destined for the AOB Blog.

    • Thanks for pointing that out. I might need to re-think workflow. I can see authors ticking a box to agree to have their images used. Putting more barriers in place, like signing up for an account, then uploading and machine tagging yourself would reduce uptake. Wikimedia Commons might be a possibility, but that would be less useful than the EOL pool I think. Update: Doh! I’ve just seen iNaturalist is another way to contribute to EOL, and is an excellent site in its own right. I can try reaching out to them.