Subscribers to David Riecks’ Controlled Vocabulary group may have seen a post by David himself: why app developers / providers have such a dim view of their users intellectual property. This describes an exchange he had with developers on EyeEm, including this response:
“it’s definitely not a technical challenge. We just never saw the value in copying exif data over to every thumbnail that we generate. The metadata is all preserved on the original uploaded photo (which is not available via a public url). That said, you make some valid points here regarding ownership and copyright. I’ll add this to our product wishlist.” [Read More]
The embedded metadata stripping issue is becoming an increasingly hot political potato where some kind of order needs to be introduced. Given the current plans for orphan works legislation (in numerous countries across the world) where a licence to use an image can be acquired if there is no obvious attribution, retaining embedded metadata seems to be essential, in my view. It’s analogous to re-touching the image without the consent of whoever supplied it. If you did that, you could reasonably expect to get sued if found out and it should be the same for the non-visible portions of the file also.
Equally, however, there are many people who don’t appreciate that photos taken on smartphones, tablets etc even have metadata within them that can provide a lot of background data about where the photo was taken, the time/date and a variety of other details that can have a privacy impact.
Whether or not privacy is the real reason the developers strip the data is hard to say. It’s also probably less a case of the developers having a dim view of IPR, more likely that the default behaviour of whatever components and process they use to prep the images for display on their site does it anyway and no one either thought or bothered to check. My own experience with software people who are inexperienced with DAM (and any kind of media processing related development work) is that they probably lack an awareness of the issues to understand why they need to at least consider this.
It should go without saying that any DAM vendors who have technical staff working on their systems should drum this into them from day 1 (and never stop either). Outside this more specialist realm, a preferable long-term solution would be some kind of mandatory option (checkbox, radio or whatever) that asks you whether you want your embedded metadata preserved when you upload, ideally with a link to a short ‘help’ article that explains the implications of either choice. Those who are image professionals and definitely want embedded metadata retained can elect for that option, others who are more concerned about their privacy can pick the other one and have potentially sensitive data erased. In both cases there should be an audit available about who chose what and when that providers are required to retain for the duration of time an image remains on their web property.
If, as part of orphan work legislation, this choice was mandatory for any system which allowed user supplied media, then there is a clear and unequivocal election of one route or another which can be tracked and used as evidence if litigation was subsequently pursued. I’m not overly familiar with proposals for handling orphan works and the arguments and counter-arguments for or against, but, as someone with a software background myself, this seems to at least be the starting point for dealing with this issue more explicitly than the current free for all we have now.
Unfortunately, I suspect that this will only be forced on to the agenda when there is some major lawsuit where the consequences are severe enough for providers to worry about the impact this will have on their businesses.
- Metadata Automation Webinar Recording From New Jersey DAM Meetup Group
- Can Enterprise Taxonomy Management Survive Analyst Reticence - And Does Anyone Else Care Anyway?
- The Role Of Taxonomy Governance In DAM Interoperability Initiatives
- Google's Visual Case Study Of The Perils And Politics Of Automated Metadata
- The Perils And Politics Of Automated Metadata Generation