Metadata Monday: Tips for the Extremely Paranoid

I ran across this paper in CyberGuerilla called "The Danger of Metadata" that discusses the potential pitfalls of file metadata. If you can look past the paranoid author's foul mouth, there are some very astute points about how metadata can be used against you, from GPS information to serial numbers, leaving you vulnerable to trickery.

There are also detailed instructions on removing auto populated metadata from different file types from PDF to Word docs and JPG's. While copyright metadata is important, there are other instances where you might desire the removal of all metadata.

While I am not suggesting covering up criminal behavior, this article is worth a read merely to see the other radical side of the metadata spectrum.

 

 

5 Things Thursday: DAMs, Taxonomies, Librarians - Tips Galore!

5 things is back right on time!

1. What are 10 common mistakes when developing a taxonomy?
2. Would you like 5 principles of innovation?
3. How about Screwy Decimal's 10 tips for librarians?
4. What are the top skills a DAM professional should acquire?
5. Where to find all kinds of metadata standards - both here and here.

Metadata Monday: Metadata Matters

Have I ever mentioned a fabulous blog called Metadata Matters? I think I have in passing, but I wanted to take some time on a Metadata Monday to expand on my admiration for the content. This blog features very technical information, primarily appealing to the traditional cataloger set - those involved day-to-day with MARC21 and RDF and AACR2 and Dublin Core.

I like the way this blog takes fairly complex concepts and breaks them down. There are plentiful diagrams and examples of code. From linked data to simple and informative conference reports, metadata is the star of the show. This is a great resource to add to your RSS feed of choice.

The latest entry is A Manifesto for Managing Metadata in an Open World. Since metadata goes through the create/publish/consume/integrate cycle, there are many rules to employ for consistency and accuracy. Similar to geometric theorems, this manifesto lays it all on the table. Metadata slackers beware! After all, findability depends on the quality of the metadata.

 

5 Things Thursday: Taxonomy, Indexes, Project Management

Here are five things to muse:

1. Can librarians (or anyone) do more with less?
2. Awesome notes on taxonomy from Seth Earley's Confab 2012 Workshop.
3. Interested in the role of librarian as project manager?
4. How big are the indexes of web scale discovery services and how does that affect search?
5. Is Google building the world's largest database of links to pirated content?

Metadata Monday: Taxonomy of Pop Songs

This blog post about creating a taxonomy of 90's pop songs is interesting for a number of reasons. Although my pop song taxonomy would more likely lean towards the 80's (examples to follow), the idea of creating a taxonomy or an organizational structure to organize personal information, like music, in a relational way could prove very useful for recommending and remembering certain songs or bands.

Kind of like your own personal Pandora without the occasional bizarre tangents, this type of thinking is less algorithm and more user input focused.

Consider the author's example, noting representative cases especially:

Summer Jam Band
Representative cases: Smash Mouth, Sugar Ray
Random, enjoyable single: I'm not sure if the one I'm thinking of is "Every Morning" or "Someday."

At the very end of the nineties, for a few sun-drenched years, "Songs with a kitschy beach party in the video, probably" inexplicably became a genre of pop music. I'm not complaining about this, but it happened, and Smash Mouth and Sugar Ray are the class of the group.

If I had to guess, I'd wager that in the distant past this species broke off and developed independently from the rest of the third-wave ska-punk bands (not pictured) and ended up, in a musical case of the Galapogos Effect, evolving a bunch of weird, purpose-made appendages, like the acoustic-guitar-with-a-DJ-effect-on-it and the rawk organ and the checked bucket hat.

Presumably the island where they were marooned had a bunch of tiki torches on it.

Now here is a more fleshed out example with more fields from my personal audiofiles:

Moody British 80's Pop

Representative cases: The Smiths, The Cure, Echo & the Bunnymen, OMD

Related to: British Pop with Clever Lyrics (see Squeeze, XTC) and Clever Men in Suits (see ABC, Spandau Ballet)

Imagine if you could populate musical relationships based on personal metadata tags. It is kind of like imagining that the Netflix recommendation algorithm mentioned last Monday actually allowed for user input (aside from ratings) to generate more ideas.

 

5 Things Thursday: Grateful Dead, Digital Audit, High Def Film

Here are five things which seemed pretty interesting this week:

1. Does the scent of patchouli make you long for a peek at the Grateful Dead archive online?
2. Do you worry about deviating from taxonomy standards?
3. Have you audited your digital assets lately?
4. Want tips on attending a virtual conference?
5. Learn more about the Warner Bros archives and conversion to HD.

Metadata Monday: How Netflix Gets Tags

Ever wonder how the Netflix suggestion engine works? Each film is tagged with at least 100 data points. These tags are created not by metadata elves, but by people like Jordon Canning described in this article.

She watches movies and enters data into a spreadsheet on a laptop tracking everything from the characters occupations to the amount of smoking in the film. Some are straightforward and others are subjective  -- how violent was the movie on a scale of one to five?

While this certainly clears up how the data is created, it doesn't clear up some of my lingering questions such as why certain films show up in my recommendations (algorithms are wacky) or why IMDB and Netflix don't join forces or integrate data.

 

 

 

5 Things Thursday: Hitchcock, Summer Reading, Zines, DAM Maturity

Here are 5 things for July 5th:

1. Are you a Hitchcock fan? His 1925 directorial debut has been restored.
2. The very funny Jen Lancaster's summer reading suggestions.
3. Want to be a zine librarian?
4. How mature is your agency's DAM system?
5. How does gamification work as a rewards system?

Metadata Monday: Clean and Tidy

The first line of the abstract for this excellent article on D-Lib Magazine says it all:

"As digital library collections grow in size, metadata issues such as inconsistencies, incompleteness and quality become increasingly difficult to manage over time."

I have spent the better part of the past year managing a clean-up project of the digital collection I oversee. Since the primary goal of any digital system is to retrieve relevant assets when users search, extraneous items and messy metadata adversely affect results.

This project at the University of Houston involved a four step pilot audit including a literature review, planning phase and ultimately correction and changes to targeted metadata. Since the project was handled by mainly by interns, there were frequent consultations with professional staff to direct progress.

My project was similar although with a slightly different order of steps. The planning phase occurred before we engaged temporary help to archive assets and clean-up fields. I am still working on changes to several new custom fields.

This paper is valuable for anyone undertaking an audit. Many systems are maturing and continued usefulness and growth relies on periodic assessments of the quality of the metadata used.