compared with
Current by Vladimir Alexiev
on Dec 09, 2013 15:58.

Key
This line was removed.
This word was removed. This word was added.
This line was added.

Changes (15)

View Page History

h2. FR Dependency Graph
FR-weave.pl also generates a dependency graph (processed with graphviz dot) showing which CRM props and sub-FRs are used in the implementation of each FR. Legend:
- This lets you see which CRM properties are "bundled" into which FRs. Similar info about the old implementation is at [FR Implementation-old#FR Summary (Short Variant)] or [FR Implementation-old#FR Summary (Long Variant)]
- It was an indispensable debugging aid while creating th FRs.
- plain text: 51 source classes/properties
- filled rectangles: 13 intermediate sub-FRs
- rectangles: 19 target FRs

FR-graph.gv is processed with graphviz dot to create [^FR-graph.png]:
This lets you see which CRM properties are "bundled" into which FRs.
- It was an indispensable debugging aid while creating th FRs (that there are no disconnected properties, each FR uses all source properties as expected, etc).
- Similar info about the old implementation is at [FR Implementation-old#FR Summary (Short Variant)] or [FR Implementation-old#FR Summary (Long Variant)]

[^FR-graph.png] (full resolution)
!FR-graph.png|width=1000!

I don't believe any of the problems reported on that page are fixed by BM

To cater to these inconsistencies, the FRs loop by (P9,P10,skos:broader)* which means:
{info}
If present at an event, a thing is considered to also be present at:
- sub-events (P9)
- super-periods (P10)

But as you see, the hierarchy is navigated *up and down*, which violates principles of covariance.
{warn:title=If someone does the CRM property}
If a data provider uses P9_consists_of (instead of skos:broader) to connect Periods, the following false inference will result:
{warning}
If a data provider uses P9_consists_of instead of skos:broader to connect Periods (whcih is the *right* thing), the following false inference will result:
- GAA42731 was present at Ancient Egypt, therefore it was present at all sub-period thereof, its sub-periods, eg 26th Dynasty
{warning}

{section}
{column}
In 2 above we have asked the production sub-events to be directly connected to the object:.
Then we'd navigate upward and the P9 link is never used:
!FR12_was_present_at_data1.png!
Then we'd navigate upward and the P9 link is never used
{column}
{column}