1. change - a folder for pages that mutate TM per #1.1 and subpages
1. overview - current top level public page for this app sub-section
• CMSspec - public include page, via forms.jsp
• CMStest - public include page, via forms.jsp
• CMSarch - semi-public page for Nada, JAL?
•  wholePart, internal only - on Ideas & _gestault management
2. forms.txt (of JSP) - allows entry of a TERM or FORM - includes spec
• Note: first call made to Listener.getNextStep() sets ndx.listener
• Right now, this happens only from forms.jsp!! => a bug!?
3. terms.txt (of JSP) - resolves name conflicts  - NEEDS definer added?
4. Obsolete specs on above, useful as ideas or history, in time order  
• topic - w/ id and a#, shows new w/ pending changes
• assoc -w/ id and a#, shows new w/ pending changes
• commands - needed for isa/ako/aso + has, forget, push/pop
2. query - a folder for pages that display underlying model
1. listing.txt (of JSP) - list of all topics meeting a specified query
1. listing-help.txt - included specs - see below
   
2. listing-0a,b,c,d,e.txt - for listing of all associtation types
3. listing-0,1,2,3,4,5.txt - for listing various queries
4. boot.txt - to execute the logic of focus=null reboot
2. review.txt (of JSP) - to see a topic w/ all associated roles a players
• review-help.txt - included specs
  
• review-hdr - for formatting the main chart
3. format - doc only - explains layout and CHRC types
• sun.txt obsolete - line-formatted <tbody> rows, easy to parse
• query.txt  obsolete - is just a listing of static links to topics
  
3. shared - a folder for shared support pages model
1. error.jsp displays a message per its inputs
2. index.jsp lists query pages & change pages; copyright
3. navhdr.txt - adds links to top-level pages
4. ontology.txt - in 2/9 JSP; lists major EARLY features of XTM/CTM
5. inglish - 2/8, on a GOOD blend of LATER Inglish concepts to add
6. ctmdemo - 2/20, on how I could use LTM to enhance CTM context
7. term.jsp displays list of links to somewhere - wrong place?
8. ensureIndexSet.txt - find via session, after override check via t0={name} 
4. charting - holds published web docs (and UN!) we don't want to lose
   
1. index.html lists 2 query pages and 2 change pages
2. usecases.htm - in three parts for agents
3. semantic site map - one application for this stuff?
5. biopax - first sent JAL mail about 10/22/04
1. index - a/o 2/20, I now openly seek fund and help
   
2. usecase - to support the argument below
   
3. perturbing:- this is easier than non-FOL
   
6. OTHER PROMOS - often modified subsets of folder #4, with local notes
   
1. scripts - for demos and introductions
   
2. context - for root? links MODELER design page to #4
   
3. miles - sent about 9/17/04
4. mit - sent about 10/31/04

• 12/05 - add 3.8 = shared/ensureIndexSet for new "machine mode" (must still try t0=)
  
• 12/05 - add auto-detect of folders in 2.1.4 and use the above; support it in 2.1
• 12/06 - update 1.2, 1.3, 2.2 to support 3.8 - adds machinemodel (still in theory only)
• 12/06 - delete charter/x-query as cleanup task, and post this stuff to Lexikos
  

• 5/14 - done 5/17 (in Eclipse only) - Lextend can dump WORDS Scripts!
  
• 5/15 - done 5/17 - make HAS handle Locator types such as  CTM's :sid
• 5/18 - redo basic scripts, CMStest/spec pages, Java updates
• 5/18 - split up latest CTM script into parts; add exampls to VERB models
• 5/18 - add upper ontology via scripts; resolving _thing, :object issues

• 6/02 - update to TMAPI SP1 - see bug fic changes
• 6/02 - add and used special TMAPI utils (downloaded)
• 6/02 - open and try out latest omnigator (downloaded)
• 
  
• 5/14 - IndexData & History support CTM as "empty" file loading
• 5/14 - add variables so scripts better mimic CG-like SITUATIONS
• 
  
• 5/14 - add AKA and auto-define JSP, so Terms can be defined on the fly
• 5/13 - fix DET and related REIFY code not to use Commands
• 5/14 - adjust DET to drop the NEED for INGROUP - clearer
• 
  
• 5/14 - Script linking, so I can import SITUATION models as needed
• 5/14 - make saving files work - takes new IndexData; gets File map!
  
• 5/02 - fix the Error page to say more about itself
  

• 5/14 - done 5/14 - do a new CMStest page, with room for more scripts
• 5/14 - done 5/14 - redo CMSspec page, much cleaner, with INTRO link
• 5/14 - done 5/14 - clean up forms.txt logic to include eithe above per help
• 5/14 - done 5/15 - Post 3 above, 5 scripts, new Java source, _release pages
• 5/02 - done 5/2 - Fix/post the Error page to emit in Red

• 4/26 - done 4/28 - Post all Java, and JSPs and up to Supplehost
• 4/26 - done 4/28 - terms.jsp must allow look as well as select
• 4/26 - done 4/28 - CMSspecs get posted in simplified pre-release shape
  
• 4/17 - done 4/28 - Scripts to publish a few demos - before Friday's BPC group
  
• 4/26 - done 4/28 - Change Management page, and BioPAX only gets linked to it
• 4/26 - done 4/27 - CurrentUser - init within themes; shows up in pending
• 4/26 - done 4/27 - IT and THIS; HAS now requires one; created timestamp
• 4/24 - done 4/25 - Add next compiled code for HAS - cannot really pre-release without it
• 4/24 - done 4/25 - Fix and integrate terms.jsp - so it can be demonstrated even with Types
• 4/24 - done 4/25 - Embed specs into forms.jsp - so I can show it to BPC, and sales clients
• 4/17 - done 4/23 - Add Scripts.java, whose utils can "read in" next test string (as from a file)
• 4/17 - done 4/23 - split Form,java up into utilities with Ideas and Nature, and call Scripts
  
• 4/17 - done 4/23 - except not in Ideas - Forms,java on simplest possible syntax by calling Ideas\
• 4/17 - done 4/23 - Make Scripts read named text files
  
• 4/16 - done 4/23 - Add change/forms.jsp, which gets a web form for Terms/Forms data entry.

1. fixed MISSING links in the page header(s) to listing, focus = null, 0 (with arg1=1, 2)
   
2. fixed MISSING shared/index.jsp (welcome) page body to set listing.jsp args
   
3. fixed BROKEN page footer that still includes Portland phone number
   

1. mostly fixed BROKEN display (via boot.txt) for the current file that is loaded
2. HELP text needs to get reviewed, fixed, and culled of bad links/comments
   
3. (defer), loading files must be expanded to better cross network space

1. simplified shared navhdr to cite only the ontology page and sample contexts
   
2. fixed review and format pages, mostly to handle new Java classes
3. updated 11 listing files - which I hope to refactor quite soon
• changes will involve 5 (shared?) JSPs for each data type
• listing.txt can include them, not hard-coded doc files
• these page initially should mostly cite the XTM 1.0 specs,
• LATER, I'll add notes for WORDS that cite CTM models
  

• boot.txt must move to shared; listing.txt fixes this, and next item
  
• topicmaps.txt must exist; listing.txt must fix code for this as well
  
• focus 0 should get treated as below, for code symmetry
• focus 1 will therefore start to exist; so fix up listing.txt, 0b -> 1a
  

• focus = null => clear off session to boot (and reset focus = 0; arg1=1)
• MISSING: if no arg1, include a (admin-editable) welcome-page body
• it lists static listing page table row links with args in them
• it is stubbed by a simpler boot, with a default TM being built
  
• if arg1 and/or arg2 use them to decide file from which how to reboot
  
• focus = 0 => list of links to classes of subgraphs (with aspects?) per arg1
1. association classes (via listing.jsp?focus=1&arg1={the class})
2. topic classes  (via listing.jsp?focus=2&arg1={the class})
3. (defer) aspect classes  (via listing.jsp?focus=3&arg1={the class})
• focus = 1 => listing of assoc [MISSING = subtypes and] instances for the type in arg1
• focus = 2 => listing of topic subtypes [BROKEN and/] or instances for the type in arg1
• (defer) focus = 3 => all topic instances, subkinds with the aspect class in arg1
  

0 - HTTP requests, passed on to JSP componentst

• Some acts may be ad hoc, custom to Miles
• query acts for main UI, naming editor and topic ID
• update, edit, approval acts for expansion scripts
• latter involve 1-N WORDS idioms, passed in ISA parms
• text file output = list of CHRC-specific lines per act, ID
• multiple-line answers legal for any query
• *** if we dump a context later, this is what we need
• expect coded error notices for user on any act

• secure JSP pages signify the act; import local portlets
• always provide the main output
• bundle calls to WORDS forms
• portlets are invisible, page components
• may be based at first on Java code only
• means no/limited nesting of forms?
• *** may also use JScheme for acts, not security?

• similar to Omnigator, but merely a listing of entity aspects
• instead of association view, get all roles in each (up to size limit)
• scopes on topics or roles is needed on each CHRC
• *** TMQL queries can also be passed in as desired as special acts
• this is advanced; takes special logic and pricing
• without it and LTM needs, could use TinyTM too?

• ad hoc frame-idioms to efficiently upload new Entity, Role, Prop, Name
• adding articles or entities new to the TM itself
• adding props/roles/names from slot-like form-parms
• Keep the above super simple, but allow slot-lists
• Miles will ike command-like nature parms
• call this WORDS, not JScheme, to get latter into build
• *** later, we can add posted or read-in JSchme uploads
• changes (a separate speech act)
• may reuse idioms to notate each change
• may perhaps, (later) also use HAS for updates
• deletions (a separate act) mark specific roles/props/ents as deleted
• all above go into one daily CSVLog, scoped, for persistence/approval
• tuples of ((time, editor) scope thing aspect value(s)) changes in log
• any editor's changes lock an old entity in the TM until all are made
• at that point (before the entity unlock), the editor accepts new values

• This is fully general **** logic, and allows scopes to be added in Modeler
• Only the UI must change, as storage layer must let rescoping be normal
• Miles may get a special scope used in his TM that drives all approval
• TomCat chron job runs new export overnight; renames daily/weekly logs

• retrieval logic can display any full topic from TM and mark delta's
• basic plan is to accept each entity's full set of changes all at one
• in it, all/no entity changes accepted, and results/entity logged for wk
• line-item rejection is also possible in the servlet posting?
• simply causes the main loop to skip some flagged changes
• editor notify can be automated by UI as well, if desired
• approver should NOT make edits directly (as approver)
• may play both roles on another PC/browser
• role is session based, assumed separate
• this UI should finalize changes to drop scoping (or aspect/ents)
• weekly CSVlog should mostly work like outbox did?

• this does bulk-uploads, and is mostly in UI layer code - to build up XTM
• adding approval can come later, if initial uploads use normal/other scope
• *** we can add simple constraint checking (hourly) per early results

• Miles need hear of this only a future hourly option on populate/edit
• *** using JScheme gets us a testbed for inferencing script