Version 1.1.0
  Rearanged using "db" shared library, getting things working under
  debian system.
Version 1.0.1
  KanjiTable--added highlights, info to def page
  Speeded up sort
Version 1.0.0
  KanjiTest--fixed info splash page
  KanjiTable--added button to bottom of screen
Version 0.9.9
  KanjiTest--Rewrote generation to pay attention to similar characters
  KanjiTest--removed FrmPopupForm to fix memory problem--SERIOUS
  KanjiTest--fixed save state (not quite)
Version 0.9.8
  Added user tracking to KanjiTable.  Highlights trouble kanji when sorted
  by frequency.
  Both romaji and kana databases are included
  8th column bug probably fixed
  Version number included in info screen
  Bigger letters in Kanjitest
  Menus used in kanjiTable                    
  More characters shown on screen in 9 column mode
  "Merit" used more fully in determining if characters should be added to
    the drill (experimental) 
Version 0.9.7
  Added Spahn & Hadimitzky, Grade, by rewriting indexing system.  Sort
  optimized as much as possible, but still somewhat slow.

  Added KanjiTest to distribution
Version 0.9.6
  Fixed(?) find bug, which caused fatal exeption.  _Seems_ to work 
Version 0.9.5
  Added frequency based index, for use in new "KanjiTest" program
Version 0.9.4
  Fixed another "Find" bug by removing "Search for SJIS code" and using
  "StrStr" on "StrAsTyped"
  Rewrote font generation code to create database
  Fixed font generation to correct for height of characters such as "ichi".

Version 0.9.3
  Rewrote database generation code to use database for bitmaps.
  Rewrote kana-romaji translation code--no more rixyuu.

Version 0.9.2
  Added "search for SJIS code"--not sure how useful it is.  Meant to
  interact with JStroke
  No more temporary allocation of dynamic memory for indices

Version 0.9.1
  Fixed bug in writing out database
  Temporary index stored in data memory
  String pointers eliminated (indices used instead)

Version 0.9.0
  Initial, non-public, release
