ALEG

What I've done

• Worked exclusively on the works/expressions/manifestations user interface. The user interface is largely now done and workable, although lots of little things aren't working yet, and some big things (such as actually storing what has been entered in the database!), but there haven't been any huge hurdles, just lots and lots of little ones.

  A large job was the restructuring of the file which describes the contents of the user interface. This file contains "instructions" to the program which runs in the web browser to build the form used to edit the work/expression/manifestation. The previous structure worked well with the agent - it is relatively simple, with birth and death events and relationships with pseuds. However, with works things are much more complicated. Following the FRBR/Indecs models (as we are), we have a deep structure in the data:

  • Work
    • Creation (Event)
    • Realisation(s) (Event)
      • Expression
        • Embodiment(s) (Event)
          • Manifestation
            • Contains part of relationship(s) (optional)
        • Contained within relationship(s) (optional)

  plus the review-of/reviewed-by relationships, topics, etc

  Revealing this complexity on the form, tempting as it maybe, makes the form unusable, so levels must be collapsed (and re-expanded).

• Discussions on use of diacritics and formatting of search results

• Document how ALEG will handle some tricky cases - The "Poets of the Month" works from the mid 1970's and "Down the Lake with Half a Chook". These are amongst the most "difficult" cases Tessa and Kathy can come up with, so if we think the proposed data model can handle these, we'll be happy!

• Continue implementation of the work/expression/manifestation maintenance interface.

• Put in place a temporary topic selection mechanism - soon topics will be in a hierarchy, but they aren't yet (pending new thesaurus structure).

• Convert the author/title as subject data, currently just stored as text (it needs to be made 'live')

• Kerry has sent me the WA, LAW and BAL datasets ready for loading.

• Start implementation of the search result changes discussed this week.

• OK. This week's most nagging worry is related to how the attributes/events of works are inherited by expressions, and how those of expressions are inherited by manifestations. The work user interface allows selection of expressions and manifestations, not just works. But most of the time, expressions and manifestations don't have their own attributes (eg, author, date), sometimes they do. When searching for an expression, I'm sure you'll want to enter the title of the work or the expression, and I'm struggling to implement this efficiently....

Links of the week

• The Case Against Micropayments - Clay Shirky

  Micropayment systems have not failed because of poor implementation; they have failed because they are a bad idea. Furthermore, since their weakness is systemic, they will continue to fail in the future.

• Hal's Legacy: 2001's Computer as Dream and Reality, excerpt from "From 2001 to 2001: Common Sense and the Mind of HAL" - Doug Lenat

---

Home > Weekly Reports
Kent Fitch
k.fitch@adfa.edu.au
5 January 2001