1.- Schema change requests: #92 (E. Yu): Drop the significant_event table. ----------------------------------------------------------------- It seems unclear if we really have use for this table. (It is not populated here in SCEDC) is it at the NCEDC? I see no clear policy about how we would use this table and there seems to be no requests for its use. ----------------------------------------------------------------- ------------------------------------------------------------------------ ==> Approved. ------------------------------------------------------------------------ #113 (E. Yu): Need to use same waveform sequence name (NC uses waseq), (SC uses wfseq). ----------------------------------------------------------------- It looks like we are not using the same sequence name for the waveform table? Also the filename sequence - NC has it as fiseq, SC uses fileseq. ----------------------------------------------------------------- ------------------------------------------------------------------------ ==> Approved. Caltech agreed to change their sequence names. ------------------------------------------------------------------------ #93 (A. Walter): Origin.bogusflag should be made into a NOT NULL column. ----------------------------------------------------------------- Applications should enter a value for origin.bogusflag because other applications look at this field. ----------------------------------------------------------------- ------------------------------------------------------------------------ ==> Approved. Add NOT NULL constraint. Add DEFAULT VALUE of 0. ------------------------------------------------------------------------ 2.- Possible use of svn for tracking schema related documentation: * DDL scripts (SQL): a) Schema creation (PI,WF,HT,IR,SIS): - Should we just store the logical model of the schemas? NC & SC physical models are different. b) Alternate tables (RT, post processing): - Does SC already store them in svn? - NC should do the same (ChannelMap, AssocWfRc, ...). c) Views: - NC & SC have different versions of some views (md_magparms_view, ...). d) Packages: - Does SC already store them in svn? - Berkeley should do the same (Truetime, ...). * ER/Studio files (HTML). ------------------------------------------------------------------------ ==> We decided in a 1st pass to store only the logical views of the DDL scripts. Ellen voluntereed to come up with a draft for the SVN structure. ------------------------------------------------------------------------