Metadata Myths and Madness!

Picturepark’s “Metadata Myths and Madness!” digital asset management webinar features John Horodyski speaking about metadata best practices, the value of metadata standards and more.

Digital Asset Management Webinar

Webinar covers metadata standards, how much metadata makes sense, and more.

RECORDING DATE July 25, 2012

GUESTS John Horodyski (DAM Education Principal)

HOST David Diamond

Want to continue the discussion? Visit the YouTube page for this digital asset management webinar and leave a comment. John and David will continue to monitor the page.

Webinar Questions and Answers

A number of questions came in during the webinar that could not be answered due to time constraints. Those questions, along with answers from our panelists, can be found below.

I’m used to having a Quality Control step for all art production. Are you aware of a standard for QC for metadata? (John Horodyski) There is a great white paper, “Metadata Quality Control Best Practices,” from the Project Performance Corporation that you should download.

When talking about DRM, is it dangerous to use protected keywords from e.g. competitors for my own website to show up on searches for that search term?

(David Diamond) This question is less about DRM than it is about SEO and copyright law. With regard to search engine optimization, it would be considered a “black hat” SEO practice to stuff your pages with a competitor’s name just to hijack their traffic. This sort of practice is frowned upon by Google and it might actually reduce your search results ranking. If you try this in an AdWords campaign, it’s unlikely you’ll even be able to get the ad approved. If you use a vendor’s name in a valid product comparison, that’s a use you could better defend, assuming the content is fair and legitimate. When in doubt, always consult an attorney about matters like these. Keep in mind that “legal” and “illegal” aren’t the only considerations: Can you afford to defend yourself in court if a case is filed? Even if you do prevail, you might do so at the expense of your company’s case reserves.

How do you best balance user (client) customization requests and “exceptions” with regard to metadata consistency and integrity?

(John Horodyski) It’s always a fine balance to get this right … you do need to listen to your clients, but be reasonable in what you can offer both in terms of the “intellectual” management of the data in your Metadata Model, as well as what the DAM software can provide.

What if there is no standard for your particular subject matter? Can you create your own?

(John Horodyski) As I mentioned [in the webinar], there are many good standards available to you, and some of the good ones are listed on my website ( You should start with the Dublin Core for a great foundation and then see what else from the other standards applies well to your organization.

Will John answer the question about how to get the folks adding assets to “tag” them so others can find them?

(John Horodyski) A good workflow process will help institute rules and procedures for asset tagging at asset creation, and throughout any approval workflows in the asset life cycle. There may be many “creative” ways in which to encourage good tagging and being a good DAM and Metadata evangelist helps!

If I have metadata but no DAM yet, and the metadata is messy, should I import what I have or start from scratch? What is the threshold?

(John Horodyski) If you believe your metadata is messy, then YES, take the time now to review that and clean it up before the DAM comes on the scene … you will learn more that way!

(David Diamond) Also evaluate why that metadata is a mess. If it came from what will become a regular source of metadata for you, you might need to move the “clean up” operation upstream so that by the time the data comes to you, it’s in better shape. But, as John says, definitely do this before the metadata gets into the DAM. Tools like Excel and other data processing gadgets provide much better “data cleansing” capabilities that most DAMs.

Do you advocate extending the schema of an existing standard to suit your organization’s needs? (PRISM / XMP, etc?)

(John Horodyski) I have seen many organizations do this … if it works for you and your organization, then do it.

(David Diamond) Keep in mind that an “extended” metadata standard is no longer really a metadata standard. If you think you’ll need to exchange data with other departments or organizations that are based on the actual standard, it might be a good idea to somehow flag all your extended fields. Either put them on a separate layout that identifies them as not being part of the stdnard, or perhaps name them with a prefix/postfix. Example, “Location” might the name used for the official field while “_Dan’s Location Notes” identifies the field you’ve added yourself. When sorted, these fields will also sort separately.

I think we’ve done a pretty good job of keeping our metadata clean, but users still complain they can’t find things. What do you recommend for someone who’s run out of ideas to make things better?

(John Horodyski) I would encourage more communication with your users and engage them in creative ways. Perhaps, more training sessions, have them participate in user interface / term review sessions on a quarterly basis, etc. … let them be part of the process and the solutions.

(David Diamond) When metadata editors think exactly like metadata searchers, everyone is happy. Where’s the disconnect? I agree with John: get those users in to some meetings. Put them at a table with your editors, put a stuffed bunny on the table and ask them each, “What is this to you?” See how they think. It’s your job to connect their minds, so you need to get to know them better.

We recently switched from an old DAM that claimed to support XMP. Our new DAM supports XMP too, but it’s not the same. Any ideas why?

(John Horodyski) It appears that you are having some Vendor technology issues. I would need to see more to better answer your question.

(David Diamond) Also keep in mind that metadata standards evolve over time. Your old DAM might claim to support a standard, but perhaps not the most current version of that standard. Though not technically a metadata standard, XMP is a great example of something that’s grown over the years.

I don’t understand what metadata has to do with DRM. How can just changing a metadata value change rights?

(John Horodyski) DRM is a very specific technology / solution / process. If you do not have a formal DRM solution in place, you can use fields such as “Rights”, “Contracts”, “Licensing” to help manage the assets.

(David Diamond) DAM vendors love to say that they “support DRM,” which is basically a slippery way of saying that metadata fields are a great way to communicate directives, such as those that might be part of a DRM process. But metadata values don’t do much by way of policing actual use—they are not copy protection. Once the asset leaves the DAM, metadata isn’t going to protect it.

My wannabe contributors are on the administrative side. How do I get them to stop making suggestions about the system itself?

(John Horodyski) Great question … but you can’t really stop them. I would suggest that you continue to receive their feedback and thank them for their contributions. You never know, one of those suggestions might actually stick!

(David Diamond) Tell them you’re starting to fall in love them and you need some distance.

My DAM uses up a user license if I use any automation, so I don’t. Is this normal for all systems?

(John Horodyski) No, this is not normal; I would push back on your DAM vendor for this.

(David Diamond) Let me do some vendor defense here! When automation is performing a read/write operation, it’s fair that it use up a license to do that. After all, automation is performing the function of a user. But if that license isn’t released the moment the task is complete, go into every public forum you can find and tell the world your vendor is ripping you off.