As reported on David Riecks’ Controlled Vocabulary group (and elsewhere) the IPTC’s Embedded Metadata Manifesto Initiative has repeated a test it carried out of popular social media sites three years ago and found that many are still stripping embedded metadata from images uploaded to them. The Social Media Sites Photo Metadata Test Results are on their website. There is also a press release issued by the IPTC about this issue including this quote from Michael Steidl, Managing Director of IPTC:
“There are many important reasons to embed and preserve metadata – to protect copyrights, ensure proper licensing, track image use, smooth workflow, and make them searchable on- or offline. If users provide captions, dates, a copyright notice and the creator within their images, that data shouldn’t be removed when sharing them on social media websites without their knowledge.” [Read More]
In addition, there is also this by David Riecks:
“Because many of the social media sites are essentially free, users become the product, and not necessarily the customers. Users are often not aware of these practices. There should be a sweet spot between these social sites preserving all metadata and removing it all. I’d like to see more engineers working together to find solutions.” [Read More]
We wrote about this subject on DAM News back in 2013. Many of the social media sites continue to strip metadata for two key reasons. The first is because the development personnel responsible for providing the photo upload, storage and manipulation features are likely to be using software components and/or techniques which inadvertently remove the embedded metadata and they haven’t realised. The other is that they fear more the possibility of litigation based on the content of the embedded metadata itself, e.g. uploaded photos being used as evidence in court cases with all the consequent legal expense and complexity which that entails.
In 2013, I described the situation with embedded metadata and social media as a ‘free for all’ and outlined a simple functional framework where users had the choice to either strip or keep the embedded metadata, depending on their preferences. The fact that nothing like this has been implemented suggests that the social media sites essentially don’t care about it, probably because people continue to keep uploading material anyway and the aforementioned legal issues probably encourage stripping rather than retaining embedded metadata. I have drawn attention to David’s quote because it emphasises the true nature of the relationship if you are a user of social media, i.e. that you are the product (or arguably more like an unpaid supplier of digital assets).
In the short to medium term, I cannot see this changing. If you are the copyright owner of digital photography, you should probably get used to the routine pillaging of your back-catalogue when/if it gets posted on social media sites (as well as stripping of metadata by the operators). When the situation might change is if any kind of interoperability standards for digital assets become more widespread since embedded metadata with identifiers that link digital assets back to their source might become more important (and valuable) at that point. How long that will take to come to pass (based on current progress with establishing DAM interopability standards) is not at all clear, however.