Re: Ideas for Recommendations for Report

Looking at http://www.w3.org/2005/Incubator/lld/wiki/Draft_issues_page#From_MARC_to_RDF I wonder if there is useful work to be done in gathering examples, documentation and code for transformations that are already out there? I'm hoping a project I'm working on currently will release some code and docs (http://lucero-project.info/), and the COMET project at Cambridge (http://cul-comet.blogspot.com/) I hope will have some stuff as well. I also understand that Talis are looking at the possibility of releasing code that does MARC->RDF transformation.

Having worked on a transformation for Lucero my own feeling is that discursive documentation that gives strategies and highlights issues is the best starting point. Local issues and motivations can mean that code from one transformation is not immediately transferrable to another project - however, I think that snippets of code showing how specific fields have been processed would be useful (often not so much the mappings, although that's useful, but things like regular expressions to extract more structured information from fields like 300 etc.)

Does this sound useful to others, and is it a viable recommendation?

Owen

Owen Stephens
Owen Stephens Consulting
Web: http://www.ostephens.com
Email: owen@ostephens.com
Telephone: 0121 288 6936

On 24 Mar 2011, at 23:50, Karen Coyle wrote:

> The working group has fleshed out text on the issues that have been identified around LLD, and now wants to gather ideas for recommendations that the report can make. Recommendations can be at various levels from general to specific, and it would be good to have a number of proposals that could result in gains in the short term.
> 
> We assume that the recommendations will evolve out of the issues. At the high level, the issues we have identified are:
> 
>    * 1.1 Linked Data is an emerging technology
>    * 1.2 Library data is expressed in library-specific formats that cannot be easily shared outside the library community
>    * 1.3 The library standards process is highly top-down and non-agile
>    * 1.4 Current library data practices are expensive (and the true costs are unmeasured)
>    * 1.5 Library ecosystem is designed for stability and resists change
>    * 1.6 Library data may have rights issues that prevent open publication
> 
> Each section has a fair amount of detail.
> 
> As a first pass, the general categories for recommendations are:
> 
>    * 2.1 Identify costs of current practices, and costs and ROI to moving to LLD
>    * 2.2 Identify issues for migration to LLD, both technical, managerial, and intellectual
>    * 2.3 Identify areas where existing library community standards and Semantic Web standards require extension or development to support LLD
>    * 2.4 Identify tools that are needed to support the creation and use of LLD
>    * 2.5 Analysis for the transformation of current library data to LLD
>          o 2.5.1 Deduplication
>    * 2.6 Cultivate a research and development environment
>    * 2.7 Create educational opportunities
>    * 2.8 Include metadata design in library and information science education
>    * 2.9 Foster a discussion about open data and rights
> 
> We expect there to be iteration between the issues and the recommendations as we work on this, so if you have a recommendation with no issue, or vice-versa, please send it in.
> 
> We are asking committee members and anyone else who wishes to begin to fill out points in the recommendations area. (We'll turn it into text as part of the editing process, so short bullets are ok if they make sense.) If you do not have edit access to the wiki, you can air your recommendations on this list and we'll gather them. Of course, discussion is encouraged. This is the real meat of our report and all ideas are welcome.
> 
> http://www.w3.org/2005/Incubator/lld/wiki/Draft_issues_page
> 
> -- 
> Karen Coyle
> kcoyle@kcoyle.net http://kcoyle.net
> ph: 1-510-540-7596
> m: 1-510-435-8234
> skype: kcoylenet
> 
>