DPLA Best Practices Part I

  We all need a best practice so let's look briefly at DPLA with good reason.

Metadata made available is done under Creative Commons Zero (CCO) Public Domain Dedication 

Which means you can use the metadata

So here are the Guidelines

1. Give credit --others need to be able to reuse data---remember interoperability
2. Data can and does change-not static,keep a hyper-link to know updates
3. Don't mislead or misrepresent-need I say more
4. At your own risk-"as is" metadata offered and conform to local laws

So that is it in a nut shell...check the policy for more specific information ....

11 Things to Know About Semantic Web--ReadWrite

What you need to  know and probably already do..maybe...

1. no apology necessary for calling it Web 3.0 ..of course web doesn't upgrade like that..but there are phases..think co-lab and databases.
2. Allows structure on the "fly" adding as needed but this will take some time for this transition because the entrenchment of some tools within the structured system and semantic web will start slow decline of relational database tech.
3. Consulting fees for those with big headaches..with things like RDF or tuples.
4. Success will be judged under different categories..inherently intergrated.
5. No need for app to apply all about platforms and servers or enterprises.
6. Apps will be consumer or enterprise driven
7. Google may get a run for their money or at least slow down "stream roller" effect ..structure less important as underlying content will be structured.
8. Don't know how it is going to look and just know what it won't look like.
9. Pragmatic will be replaced by semantic platforms
10. Tagging we know but micro-formats will provide the structure we need
11. leverage of the community will happen under the semantic web..techniques from social networking will be used..we know its coming..just provide structure!!!

Flickr Image Tagging: Patterns Made Visible by Joan Beaudoin

Folksonomies!

We all tag so let's agree probably difficult find a pattern but our author has a plan...
 Two interconnected ideas in play
First look for underlying patterns or similarities..next  effectiveness of image tagging..which means a tag could cross intersections  to be anything...sounds very difficult when one thing fits into many.

Model evaluation
The evaluate and develop categories...ok can't agree on that ...could get a majority but not all to agree..but those open to interpretation caused problems..modification of categories of tags would improve the model's performance  ..user to choose best fit

Tag categories--how often used
Proofed preferences in tag categories used...yes, i would agree place name should be #1...that's how I usually id my own images no surprise..use of compound categories when the user decides a single tag will not serve their purpose.  Usage on some tag categories was nominal ..some showed significant usage such as, event tag...at a much lower usage were tags humor, poetic, number and emotion..

Problem of meaning
 We as librarians know this issue well..."in other cases the unknown tags illustrate how important contextual knowledge is to the categorization of the tags"..in addition foreign languages pose difficulty too.

Our individual differences
Various tags experienced highly individualized use.  Use of time tag  showed most variation among users that were studied..but since most image have a date stamp this might be redundant.

Flickr developments
  each improvement offers users a more vibrate tagging process...opportunity to strengthen natural language and greater accuracy in retrieval.

US
 Progress in image retrieval
clarify how personalized can be
tagging is here to stay so get with it...

Practical principles for Metadata Creation and Maintenance by J. Paul Getty Trust

Really practical and related well to classroom discussion as we move forward in our own creation fantastic journey of discovery!!!

J. Paul Getty Trust has some experiences in metadata creation...so I trust the source not the hip-hop magazine the source haha!!!

10 rule lesson

• is a core activity of memory institutions
• incremental and share the responsibility (kinda sounds like what are groups are doing?)
• rules and processes must be followed
• staff levels and skill sets must be present for a metadata strategy and successful implementation
• share reliable info among relevant units in institution remember interoperability
• no such thing as one size fits all for schema, vocabulary or data content standards.
• streamline and replace manual metadata creation think "industrial production"
• Routine part of workflow(creation of shareable, repurposable)
• Integral part of institution's metadata workflow with research and documentation
• High level of understanding and importance or value buy-in from upper management.

All this point seem to be common sense...but remember there are break downs in organizations such as, sacred cows and lack of vision.

May I speak Openly about mass digitization?

The first statement I agree with is that"it's the responsibility of non-profit, cultural heritage institutions to find ways to bridge that gap and work with the corporate world toward a public good." Mass digitization projects need that corporate mindset...and the results have been good for both parties and we are moving towards some great outcomes.

Public/private collaboration have worked for many other facets of our society so why not mass digitization.  Now that many agreements are public there is no mystery in the process.  As more libraries form partnership we should become more familiar how these partners work and will we as a library benefit from the relationship.

US National Archives has a plan for acting responsible within their  mass digitization project ensure the public's confidence in the agency.

But, how can we best work with these partners?  Trust them for what they do best and remember we as libraries have expertise too.  Remember that the end goal of these collaboration is open accessibility for  patrons, researchers and anybody else who wants access.

PS....Anybody interested in mass digitization check out Jan/Feb 2014 Archival Outlook pg.8 Large-Scale Digitization: Developing the Los Angeles Aqueduct Digital Platform .

My Phone is collecting metadata...stop it

The Ecology of Longevity

Survival of the fittest as it applies to non-biological phenomena.  Digital longevity using Darwin's theory as the framework.  The "fittest" as it applies to digital objects is determined by info pro's who actively or passively in decided the so call "lifespan" by selections making the life or death of object.

The digital objects are understood using the language of computers who "translates" for lack of a better terminology into visual or audio formats for human consumption. This computer and software which characteristic will version 1.0 pass to 2.0. or properties will survive or data survive. We as human have traits just as the digital hardware and software such as, MS Office interconnected of their various products on this platform and how some "traits" from MS Office 2007 are inherited into MS Office 2013.  What does this mechanical process imitation of a biological phenomena mean for digital hand me downs? View the digital formats as a living organism that evolves but requires us to create preservation tools that outlive and become evolutionary in application.

As we look at our system think about open source is represented and how this process has basically evolved on various paths by creator, users and how characteristics are directly inherited. Also, many technically advancement mimic organic evolution by (1)creating different designs and (2) existing designed are optimized for competition.

This article posit that an evolution process exist within technology and the strongest applications survive.  The ones who make it determine the future just like the victor writes history and gets the spoils which are "econosphere" of consumers and longevity among  formats.