-MonDB table hierarchy
-explanation of table relationship diagram
-relationship model
-contents of each table
-particular tables requiring extra explanation
-MONUNITS
-"multi-level" info
-document all info at appropriate LEVEL
-FIELDATA
-standardization of format
-non-normalized (and why)
-use of EXTRACT function
-why order of actions when using MonDB is important
-how to make changes to TAXA, TAXONCOD, etc.
-how/when (not) to create new taxa, taxon codes,
etc.
-how to design queries (based on table hierarchy)
-PDOXWIN environment (theory)
-object-based
-objects (=entities)
-properties (=attributes)
-containership (objects' relationships to one another)
-event-driven
-events (=actions)
-focus (links events to objects)
-normalization (database theory)
-what is normalization
-simply put, "max-out # of rows (fewer columns)"
-why normalize?
-integrity
-nonredundancy
-ease of querying
-why NOT normalize?
-data entry
-examples of (non-)normalization
-normalized
-all tables but FIELDATA
-(MONUNITS breaks strict rules: allows null components in key)
-non-normalized
-FIELDATA
-"EXTRACT" function creates normalized
tables from FIELDATA
-working with PDOXWIN (mechanics)
-the toolbars
-the Project Viewer
-working with tables
-demo
-create a simple table (work with it)
-sort it
-add key (then work with table more)
-add 2ndary index
-using filters to sort/select displayed data
-create related table (work with both tables)
-add table lookup (work with both tables)
-add RI (work with both tables)
-create 3rd table, add TL/RI
-demo lack of cascaded update(, delete)
-useful example
-add an attribute to MonDB infrastructure (e.g. "fenced")
-create a table for a custom data set type (field method)
-when/why (not) to do this
-how: steps
-copy FIELDATA
-decide whether to maintain FIELDATA/other compatibility
-working with reports
-basic: SR/MR/tableframe
-group bands (incl. sums)
-calculated fields
-graphs
-using filters to sort/select displayed data
-query-based
-(mention data model)
-working with forms
-on-screen
-for data entry
-field layout
-single-field
-tableframe
-MRO
-value-added
-buttons (PDOXWIN7 wizards)
-to display info
-calculated fields
-graphs
-as a report
-using filters to sort/select displayed data
-(mention data model)
-using :Standard:UTILS.FSL
-working with queries
-create a simple query ("data dump")
-add selection criteria
-other PDOXWIN elements (mention but ignore)
-scripts
-SQL
-libraries
-relational database concepts
-tables (entity type definitions) are related (via common info)
-table concepts
-field (=column)
-record (=row)
-field type
-validity checks
-indexes & keys
-table relationships
-lookup
-dependency forced (1-way) only at DE
-referential integrity
-dependency enforced 2-way always
-MonDB installation
-PDOXWIN setup
-program directory structure
-CONFIG.SYS settings
-network setup
-MonDB data directory structure
-MonDB aliases
-MonDB updates
-why update?
-resources
-"official"
-HIMONDBA-L
-online list of taxon codes/species
-unofficial
-MB/USFWS' online island distributions
-each other! (via HIMONDBA-L, HIMONDB-L, etc.)
-MonDB troubleshooting
-what to do (FAQ)
-who to call/where to go
NOTES TO PT:
-prepare a detailed handout!
-work out flow so theory/hands-on work together & presentation makes sense
-define time slots for each
-Casanova reservations (Sunday Nov. 23, 6pm, ~10 pax)
-export to spreadsheet
-writeup of explanation of FIELDATA fields; and explain why it's hard to do some queries
-prepare discussion/demo of EXTRACT function (get extracted PE95 from Coleen onto diskette ahead of time for demo purposes)
-PERHAPS: skip the table creation; just start w/ Monitoring Database; how to use what we have; explain table linkages; what order to do things in; the order that changes must be made in; loc. taxon codes; how to change loc. taxa to real taxa (new taxon; existing taxon); MONUNITS/DATACOLS entry tricks; duplicate script will be set up some day; do custom table creation last; day 1: current mondb setup; day 2: custom tables; ArcView linkage demo; rules for new taxon codes; not short-term projects; consistent, integrity for the long haul; make decisions re: precision NOW rather than later when 1-time crunch, 2-you can never know/remember details (can't even remember 1 wk after field!), 3-consistency over time.
-questions from Marie B.: I am really interested
in how people are using their data, once they get it into the
database. I could use some more help on the graphing stuff, and
would like to know if and how folks are running statistics on
the data. Also, are you going to be discussion how to set up a
monitoring plan so that it fits the model you have constructed?