
                    SUPER-MEMO 5 SHAREWARE

                         USER'S GUIDE

                Edited by Krzysztof Biedalak
                July 31, 1991

    CONTENTS:

    1. Introduction ...................................  2
    2. First steps with SuperMemo .....................  5
    3. Screen layout .................................. 10
    4. Menu tree ...................................... 15
    5. SuperMemo database ............................. 26
    6. Hints and tips for SuperMemo learners .......... 28
    7. Protection mechanisms .......................... 32
    8. Distribution disk files ........................ 34
    9. Outline of the optimization procedure .......... 35
   10. About SuperMemo World .......................... 37
   11. Glossary ....................................... 38
   12. References ..................................... 45

1. INTRODUCTION

The SuperMemo software implements a very  effective  learning
method - the SuperMemo method  -  allowing  to  memorize  and
retain knowledge in  the  shortest  possible  time.  Learning
aided by SuperMemo is fast, pleasant and almost unaffected by
forgetting.
   The  SuperMemo  method  is  based  on  computing   optimum
intervals between repetitions of studied knowledge. It allows
to increase the  learning  speed  from  10  to  50  times  as
compared to classical methods and  what  is  more  important,
regular work with the program ensures that the once  acquired
knowledge will constantly be maintained in learner's memory.

   The knowledge acquired  in  the  process  of  learning  is
stored in the form of long-term molecular changes in neuronal
synapses in  the  brain.  These  changes,  however,  tend  to
decline persistently, hardly dependent on  the  importance of
the  knowledge.  Repetitions  (no  matter  if  conscious   or
unconscious, intended or accidental) fix the  information  in
memory protecting knowledge from  being  forgotten.  The  old
Latin saying "repetitio est matter studiorum" indicates  that
from the very beginning of the  history  of  studying  people
realized the inevitability of forgetting and  the  importance
of repeating the once acquired knowledge.  The  only  problem
which remained  unsolved  was  how  to  compute  the  optimal
interrepetitory intervals and how to effectively automate the
process of repeating. The SuperMemo method solves the problem
of optimal timing in the process of repetitions.

In  the  SuperMemo  method,  repetitions  are  scheduled   in
intervals which have to satisfy the following criteria:
 - they must be short enough to  ensure  that  the  knowledge
   will not be forgotten,
 - they must be long enough to ensure that the time spent  on
   learning is minimized and the stimulatory effect maximized
   (higher recalling effort causes higher fixation effect).
Optimal   intervals   are   calculated   according   to   the
optimization  algorithm,  developed   by   Polish   biologist
P.A.Wozniak, and depend on:
 - the capacity of a given individual,
 - difficulty of a given piece of knowledge,
 - retention strength of a given piece of knowledge (i.e. how
   well it is remembered at the moment of the repetition).

   In the SuperMemo method, the knowledge to be memorized and
remembered is split  into  smallest  possible  pieces  called
items. Each of them is a pair consisting of  a  question  and
answer. The rule of formulating items of  maximum  simplicity
is called the minimum information principle and is  justified
by several reasons (described in other publications available
from SuperMemo World). The  most  important  reason  is  that
joining several items into a greater one requires the item to
be repeated in  intervals  that  are  optimal  for  the  most
difficult of the subitems, i.e. the remaining  subitems  have
to be repeated more often than in  the  case  when  they  are
repeated separately.
   Sets of items (question-answer pairs)  make  up  SuperMemo
databases later called just databases.
   Upon memorizing an item, the following  repetitions  occur
on strictly determined dates, with successive interrepetitory
intervals increasing in length as memory traces  become  more
and  more  durable.  With  the  assistance  of  a   computer,
calculation of optimal intervals becomes so accurate that the
rate of knowledge acquisition approaches the maximum  natural
capability of the human brain in forming memory engrams.

   The SuperMemo software line by SuperMemo World is designed
to assist learning  of  all  text-based  forms  of  knowledge
(Graphic SuperMemo in preparation).
SuperMemo programs allow the following:
 - creation  and  maintenance  of  databases  consisting  of
   question-answer pairs,
 - automatic control of knowledge repetitions  based  on  the
   calculation of optimal intervals,
 - providing the user with  mechanisms  for  supervising  the
   memorization process.
Computer applications of the SuperMemo  method,  investigated
since  December  1987,  revealed  sensational   effectiveness
especially in such areas as learning  languages  and  natural
sciences.

The following chapters describe SuperMemo 5  Semiprofessional
in more detail. Read them carefully before  you  start  using
the program.

When running SuperMemo 5 be sure that  your  CONFIG.SYS  file
allows to open at_least fifteen files (files=15).

2. FIRST STEPS WITH SUPER-MEMO

SuperMemo  5  options  can  be  accessed  through  menus.
Press Esc several times to quit any option  and  get  to  the
main menu.

The most important options offered  by  SuperMemo  5  include
creating databases, modifying them, and testing. The  studied
knowledge   is   stored   in    databases    consisting    of
question-answer pairs called items.

Creating a new database consists in  initializing  its  files
and appending new items.

To initialize a database, perform the following operations:
 - press Esc a few times if you are not in the main menu,
 - press M and I to select the  Initialize  option  from  the
   Miscells menu,
 - type the name of the new database and press Enter.

The just initialized database contains no items . In order to
append an item to the database, the following operations  are
required:
 - select Append (press A while  in  the  main  menu  or  use
   cursor keys to select the option and press Enter, use  Esc
   to get to the main menu while in any other  point  of  the
   program),
 - formulate the question using  the  SuperMemo  item  editor
   (see further in this chapter),
 - press Esc (or PgDn) to get to the answer window,
 - formulate the answer,
 - press Esc once, to append the next item, or twice, to quit
   the appending option.

Instead of creating  your  own  database  you  can  also  use
exemplary databases stored on the distribution disk (if  any)
or buy/get a database created by someone else. To perform any
operations on a database you must first open it. To do so:
 - select Open (press O while in the main menu),  which  will
   present a list of SuperMemo databases  and  subdirectories
   in the current directory (if any),
 - use the cursor keys to indicate the  database you  want to
   open and press Enter (or Esc to return to the main menu).

As soon as you have a database  containing  some  items,  the
learning process may start.
In order to memorize items stored in the database:
 - select Test,
 - press Y(es) in response to "Do you want  to  memorize  new
   items?",
 - read the first question and try to answer it (do not  type
   the answer - just tell it loudly or think of it),
 - press any key to display the correct answer,
 - compare your response with the displayed  answer,  type  a
   number from 0 to 5 as  the  response  quality,  and  press
   Enter,
 - press Enter to continue the learning process or Esc if you
   do not want to memorize any new items.

As  soon  as   you   memorize   first   items,   you   should
systematically load the SuperMemo program and get through the
Test option.

The  Test  option  makes  the  most  important  part  of  the
SuperMemo  program.  It  allows  to  master  and  retain  the
knowledge stored in the database. The  test  option  is  used
both to memorize new items from the database  and  to  repeat
items  assigned  for  repetition  on  the  current   session.
Repetition  dates  for  particular  items  are  automatically
established and maintained by the optimization algorithm. The
learner is always asked items which are  most  likely  to  be
forgotten, and has no influence on  selection  of  repetitory
dates and intervals,  except  by  assessing  the  quality  of
answers. Careful estimation of the answer  quality  is  vital
for the effectiveness of SuperMemo aided learning process (at
least in software versions up to 5).

   The recommended range of quality grades includes:
 5 - excellent  -  for  a  quick,  correct  answer  without
     hesitation,
 4 - good - for a correct answer given with hesitation,
 3 - pass -  for  a  correct  answer  given  after  a  longer
     consideration,
 2 - insufficient - for an erroneous answer when the  correct
     response presented by the computer seems familiar,
 1 - vestigial - for wrong answer or no answer  at  all  when
     the correct response seems hardly familiar,
 0 - blackout  -  for  the  impression  of  meeting  an  item
     for the first time.
In SuperMemo you are unlikely to use grades 0 and 1.

   In principle, one should work with the  SuperMemo  program
and use the test option  every  day.  Irregular  repetitions,
although allowed, decrease the effectiveness of the  learning
process.   Another   important   factor    influencing    the
effectiveness is the proper concentration while  learning  or
repeating items (see chapter  6  for  further  discussion  of
prerequisites of successful learning with SuperMemo).

When using options like test, find,  select  or  random  test
(see chapter 4)  the  program  automatically  enters the,  so
called, browsing mode. It  provides  a  number  of  functions
allowing to scan and edit the items of the  currently  opened
database.  The  moment  of  entering  the  browsing  mode  is
signalled by the appearance of a list of available  functions
at the  bottom  line  of  the  screen.  This  includes   the
following:
 - E - edit the  displayed  item  (in-built  item  editor  is
   called),
 - D - delete the displayed  item  (you  will  be  asked  for
   confirmation),
 - S - select an item of a given number,
 - Ctrl-P - print the displayed item (Do not use this  option
   when no printer is connected. If  chosen  unintentionally,
   press Esc after the PRINTER NOT READY message appears.
   WARNING! It  may  take  a  while  before  the  message  is
   displayed! Do not turn your machine off!!!),
 - Ctrl-R - reset the item,  i.e.  transfer  it to the set of
   intact (not yet memorized) items. Process parameters  like
   burden, intact, memorized are updated  automatically  (see
   further),
 - <-/-> - display the previous/next item in the database,
 - PgDn/PgUp - scan the database  forward/backward  in  large
   steps,
 - Enter - continue the selected operation depending  on  the
   option in which the browsing mode was entered,
 - Esc - exit.

Item editing - the SuperMemo program provides  a  specialized
integrated editor  for  items.  The  editor  is  called  when
selecting the append option or by  pressing  E  whenever  the
program is in the browsing mode (accessible  while  in  test,
find, select and random test options).
   The question and the answer of an item are edited  in  two
separate panes. Both the question and the answer must fit  in
the field of 5 lines, each 48 characters  long. In  practice,
this  size  proves  to  be  absolutely  sufficient   (minimum
information principle!). See chapter 8 for  hints  concerning
the formulation of items.
   The following control commands are available while editing
an item:
 - Esc - exit to the answer pane (when in the question  pane)
   or exit from the editor (when in the answer pane),
 - PgUp,PgDn - move between question and answer panes,
 - Home/End - move the cursor to  the  beginning/end  of  the
   line,
 - cursor keys - move the cursor one character  left,  right,
   up or down,
 - Ctrl-Y - delete the whole line,
 - Backspace/Del - delete a character  before/at  the  cursor
   position,
 - Tab - swap the question and the answer,
 - Alt <code> - enter a special character,
 - Insert - change the mode between Insert and Overwrite,
 - Enter - move the cursor to the beginning of the next  line
   or add an empty line if the cursor is at the beginning  of
   the line.
Note that the dollar sign is reserved and you cannot  use  it
in formulation of items.

3. SCREEN LAYOUT

The SuperMemo screen is divided into rectangular areas called
panes. The top line contains the  main  menu  options.  Other
panes are used to present data about:
 - items (ITEM),
 - optimization (TEST),
 - database and the learning process (PROCESS).

 SUPER-MEMO 5.5 ͻ
   Open   Test  Append   Find   Stats   Miscells   Params   Help   Quit  
͹
  QUESTION ͻ   PROCESS   
  What is the capital of Zambia?                                       
                                                 Database    GEOG      
                                                 Started     01.03.90  
                                                 Date        01.03.91  
                                                 Day         366       
  ANSWER ͹  Memorized   5000      
  Lusaka                                         Intact      0         
                                                 Total       5000      
                                                 Outstanding 13        
                                                 Remains     4         
                                                 Burden +/-  -0.117    
 ͼ  Burden      11.91     
   ITEM     TEST    Quality     4.300     
  Number       3232      Previous OF   4.8420      Time        0:21      
  Quality      4         Modified OF   4.7299                            
  Repetition   5:0       +/- OF        -2.32 %                           
  Factor       2.700     New factor    2.760                             
  Used factor  2.333     Opt interval  366                               
  Interval     60        New interval  365                               
  31.12.90  -  01.03.91  Next rep      01.03.92                          
ͼ

Fig 1. Exemplary screen image after a repetition.

QUESTION and ANSWER panes present the question and the answer
of the currently processed item.
The ITEM pane presents parameters relevant to  the  currently
displayed item:
 - Number - the number of the displayed item  (Upon  deleting
   an item the database is not renumbered. Renumbering  takes
   place only at compression),
 - Quality - the quality of the answer,
 - Repetition - repetition number separated by a  colon  from
   the number of memory lapses.
   The repetition number tells how many times  a  given  item
   has been repeated since it was introduced into the process
   (memorized)  or  since  it scored below 3 (was relearned).
   Memorization of an item  is  counted  here  as  the  first
   repetition  (despite  the  usual  meaning  of   the   word
   repetition).
   When an item has been  temporarily  considered  forgotten,
   its repetition number is again assumed to equal 1,
 - Factor - the current e-factor of the presented item.
   E-factor is an internal SuperMemo variable associated with
   each individual  item  and  used  by  the  algorithm  that
   computes optimal intervals between successive  repetitions
   (see other publications by SuperMemo World).
   It bears information about how difficult  it  is  for  the
   user to remember a specific item.
   Its value is updated automatically after  each  repetition
   according to the latest and all previously  scored  answer
   qualities.
   It usually ranges between 1.3 and 3.5.
   The greater the value of the e-factor for a given item the
   easier the item,
 - Used factor - quotient I(n)/I(n-1), where
     I(n) - n-th interval,
     n - current repetition number,
 - Interval - the current interval for the item (in days),
   it determines how many days  must  elapse  (have  elapsed)
   between  the  last   repetition   and   the   next/current
   repetition,
 - dates of repetitions - the  dates  of  the  last  and  the
   next/current repetition.

In our example:
 - the displayed item has the number 3232,
 - the quality of the answer in the just performed repetition
   has been estimated to 4,
 - the item has been repeated five times  so  far  (including
   the  just  performed  repetition)  and  it  has  not  been
   forgotten since being memorized,
 - the item's recent e-factor was 2.7 (which means  that  the
   item was and is easy to remember),
 - factor I(n)/I(n-1) used to calculate the previous interval
   equaled 2.333,
 - the previous interval (up to the current session when  the
   repetition was performed) equaled 60 days,
 - the item was previously repeated on 31 Dec. 1990  and  its
   next repetition has just took place on 1 March  1991.

The TEST pane is always presented after  inputing  the  aswer
quality while in the test  option,  and  presents  parameters
relevant to the current repetition:
 - Previous OF - the previous optimal factor of the  repeated
   item (optimum ratio  I(n)/I(n-1)  for  items  of  a  given
   difficulty and a given repetition number),
 - Modified OF - newly calculated optimal factor,
 - +/- OF - the change of the  optimal  factor  expressed  in
   percent,
 - New factor - the new value of the e-factor assigned to the
   item  after  the  repetition  (its value  depends  on  the
   previous value of the e-factor and the just  input  answer
   quality),
 - Opt interval - the optimal interval (in days) between  the
   current and the next repetition of the item  (computed  by
   multiplying the current interval and the relevant  optimal
   factor),
 - New interval - the actual  next  interrepetitory  interval
   between the current and the next repetition.
   Intervals are slightly dispersed round the  optimal  value
   for the sake of optimization procedure,
 - Next rep - the date of the next repetition of the item.

In our example the new interval has been  computed  to  equal
365 days which results in scheduling the next  repetition  of
the item on 1 March 1992.

In the PROCESS pane the  following  parameters can be found:
 - Database - the name of the database in use,
 - Started - the date of initializing the database,
 - Date - the current date,
 - Day - the number of days since initializing the database,
 - Memorized - the number of items  that  have  already  been
   memorized,
 - Intact - the number of items in the database that have not
   yet been memorized,
 - Total  -  the  total  number  of  items  in  the  database
   (Intact+Memorized),
 - Outstanding - the number of items to be  repeated  in  the
   current session,
 - Remains - the number of items that have been  repeated  or
   memorized in the current session, but the quality  of  the
   answer was lower than 4 (all these  items  will  be  asked
   again at the end of the session),
 - Burden  +/-  -  the  increment/decrement  of  the  burden
   parameter  in  the  current  SuperMemo  session  (see  the
   following parameter),
 - Burden - the average  number  of  items  that  are  to  be
   repeated daily (sum of inverse intervals),
 - Quality - the average quality of answers  in  the  current
   SuperMemo session,
 - Time - the repetition time of the session.

4. MENU TREE

Open opens a SuperMemo database for use.  Only  one  database
can be opened at a time. Thus, if any  database  has  already
been  used,  when   calling   Open,   it   will   be   closed
automatically. The PROCESS pane  displays  information  about
the currently used  (opened)  database  including  its  name,
number of items, number of items  scheduled  for  repetition,
etc. (see the previous chapter).
   Upon choosing option Open, a list of  SuperMemo  databases
and subdirectories in the  current  directory  is  presented.
Using cursor keys select one  of  the  databases  listed  and
press Enter to open it (or press Esc to exit from OPEN).

The Test option is used both to  repeat  items  assigned  for
repetition on a given day and to memorize new items.  If  you
have not tested yourself for more than one day,  the  program
will  successively  examine  you  on   items   assigned   for
repetition to all outstanding sessions. You can memorize  new
items only after repeating the outstanding ones.
   SuperMemo automatically  selects  items  which  should  be
repeated on a given day and asks you relevant questions.  Try
to answer each question and press  any  key  to  display  the
correct answer. Compare  your  response  with  the  displayed
answer and evaluate the response quality. It must be a number
between 0 and 5. The program  prompts  you  to  type  in  the
quality  number  by  displaying  a  blinking  question   mark
(opposite Quality in the ITEM pane). Input  the  quality  and
press Enter. After you do this the  program  computes  a  new
interval and factors relevant to the  optimization  algorithm
(see the TEST pane).

   Also at this stage, the program enters the  browsing  mode
with a number of options available (displayed in  the  bottom
line). Press Enter to continue the process of testing or  Esc
to exit to the main menu.
   Items with answer quality lower than 4 are asked again  at
the end of the Test procedure. The  current  number  of  such
items is displayed in the PROCESS pane as Remains.
   When the list of outstanding items  is  finally  exhausted
all items scheduled as Remains are  tested  again  and  again
until they all get quality grades of 4 or better.
   The process  of  memorizing  new  items  is  analogous  to
ordinary repetitions.  It  may  begin,  however,  only  after
repeating  all  outstanding  items  and   will   be   started
automatically if  the  database  contains  any  intact  (new)
items. New items must be earlier introduced to  the  database
using Append option. The number of not yet memorized items is
displayed as Intact in the PROCESS pane. You  don't  have  to
memorize all of them at once. Press Esc  after  entering  the
quality of the  last  new  item  you  want  to  memorize.  In
SuperMemo 5, new items are memorized in the  order  in  which
they were added to the database.

Append allows you to add new items to your database.
Newly appended items do not automatically enter  the  process
of learing but are stored in the queue of intact items.  This
queue is emptied only when you memorize new items.  This  can
be done using the Test option from the main menu.  Each  item
added to the database increments the  Total  and  the  Intact
number of items displayed in the PROCESS pane.
   Upon choosing Append, the SuperMemo's built-in item editor
is called. It allows you to formulate questions or answers in
the field of 5 lines 48 characters each. Use  Esc  to  signal
the end of editing in a given field.
   To add a  new item to your database:
 - select Append option,
 - type the question in the QUESTION pane,
 - press Esc to get to the ANSWER pane,
 - type the answer in the ANSWER pane,
 - press Esc once to edit the next item or twice to leave the
   Append option.

Find option allows you to search through  your  database  for
items containing a given string of characters. The string may
be up to 20 characters long.
   Use the following control keys while  editing  the  search
string:
 - Home/End - to move the cursor to the  beginning/end  of  the
   line,
 - <-/->  -  to  move  the  cursor  to  the   left/right,
 - Backspace/Del - to delete a character before/at the cursor
   position.
Press Enter to  start  searching.

You can stop the search at any moment by pressing  Esc.  Each
time an item containing the chosen string  is  found,  it  is
displayed and the program enters the browsing mode  with  its
options displayed in the bottom line (edit,  delete,  select,
print, etc.).

STATS  provides  the  following  five   statistical   options
allowing to supervise the learning process:

Factor distribution displays the distribution of e-factors.
E-factor determines how hard it is  to  remember  a  specific
item. Its lowest  value  is  1.3  and  it  usually  does  not
increase over 3.5. The greater the value of the e-factor  for
a given item the easier the item. Initially, all  items  have
their e-factors assumed to equal 2.5. This value is  modified
depending on response qualities given by the  user.  When  an
item is remembered well, its e-factor  increases  or  remains
unchanged, otherwise it decreases.

Interval distribution displays the  distribution  of  current
intervals in the database.
   This option presents a  histogram  illustrating  how  many
items have their intervals belonging to a  particular  range:
one day, two days, three to four days, etc. up to  over  2048
days.
Daily  burden  informs  how  many  items  are  scheduled  for
repetition on particular days.
   Use the cursor up and down, PgUp and PgDn to  scroll  data
displayed in the daily burden  window.  Pressing  Enter  will
result in displaying the numbers of items to be repeated on a
given day. Press Esc to leave Daily burden.
   The program itself determines when a given item is  to  be
repeated. Items to be repeated on a given day are  stored  on
the disk in the form of a list.  Thus,  it  takes  some  time
before the program can determine the number of items assigned
to a day.

Optimal factors allows to view the optimal  factors  used  in
calculation of optimal intervals.
   An optimal interval is the maximum interval after which  a
determined fraction of items are  still  remembered  (90%  in
case of SuperMemo 5).
   The  top  row  of  the  optimal  factor  window  specifies
e-factor values and  the  left  column  indicates  repetition
numbers. The first row  of  optimal  factors  corresponds  to
first intervals used during repetitions, while the numbers in
the following rows indicate how many times intervals increase
in successive repetitions. For example, given  the  following
fragment of the optimal factors table:

       OPTIMAL FACTORS (OFs)

      1.3  1.4   ...   1.9  2.0  2.1  2.2  2.3  2.4  2.5

   1  8.29 8.31  ...   9.01 9.31 9.40 9.52 9.16 9.18 10.77
   2  2.53 2.56  ...   3.12 3.24 3.31 3.56 3.39 3.43  3.74
   3  1.20 1.34  ...   1.86 1.89 2.00 2.22 2.27 2.33  2.71
                 ...                                                                                                                1
  20  1.20 1.34  ...   1.80 1.90 1.98 2.15 2.25 2.34  2.43

the program will establish that the  first  optimal  interval
for a newly memorized item will be 11 days  (because  OF  for
e-factor 2.5 and repetition number 1 equals 10.77). Using the
same table the program will, for example,  compute  that  the
optimal interval between the second and the third  repetition
of an item with e-factor equal 2.0 is 3.24 times longer  than
the interval between the first and the second  repetition  of
that item.
   The repetition number is reset to 1 each time you fail  to
recall an item at repetitions (answer quality lower than  3).
Such failure is called here  a  memory  lapse  (or  simply  a
lapse).
   The  matrix  of  optimal  factors   is   modified   during
repetitions to adjust it to  the  difficulty  of  the  tested
subject, as well as to the knowledge and  capacities  of  the
learner.

Prospects displays the current approximation of  the  optimal
intervals based on the optimal factor matrix.
   The top row of the diagram  corresponds  to  the  e-factor
values and the left column indicates repetition numbers.  The
first interval in a given category equals to the first factor
in the optimal factors matrix  and  the  following  intervals
equal  to  previous  intervals  multiplied  by  corresponding
factors of the considered category. Intervals over  80  years
long are not displayed.

Support of OF displays how  many  repetition  instances  have
been used to calculate a given entry of  the  optimal  factor
matrix displayed by  the  optimal  factor  option.  Only  ten
thousand first instances are used  in  calculations  and  the
significance of each next instance decreases logarithmically.

Miscells menu provides the following eight additional options
of SuperMemo 5 Semiprofessional.

Select allows to display and perform operations on  the  most
recently used item.
   After choosing this option, the recently accessed item  is
displayed and the program enters the browsing mode  with  its
options displayed at the bottom line.  One  of  them  is  the
select option. Activate it by  pressing  S  if  you  want  to
display an item of another number.  You  can  also  scan  the
database item by item using left and right cursor keys or  in
large steps using PgUp and PgDn. As  usual  in  the  browsing
mode you can edit,  delete,  print  or  reset  the  currently
presented item by activating the  appropriate  option.  Press
Esc to leave the Miscells menu.
   Use Select when you know the number of the item  you  want
to display, edit, delete, etc., or when you just want to scan
a database. When you have no idea of what the number  of  the
item is, but you remember its content, you could try  to  use
the Find option from the main menu.

Date allows to change the DOS date.
   The date, however, may only be changed if no  database  is
opened at the moment. Otherwise, you will first be  asked  if
you want to close the database in use.
   It is important for the optimization procedure  to  always
keep the correct date setting. In case your computer does not
provide a battery-backed RAM, be sure to set the current date
either  by  using  DOS  DATE  command  or  by  selecting  the
discussed Date option from the Miscells menu (see earlier).
   The SuperMemo program detects if the date is lower or much
greater than the last day the database has been used. In both
cases,  it  will  react  accordingly  and  display  a  proper
message. Anyway, be very  careful  and  make  sure  that  you
always work with the  correct  date  setting.  Otherwise  the
learning process may be disturbed.

Random test randomly selects  items  from  the  database  and
presents them by displaying the question and, after you press
a key, the answer. Press Esc to leave the Random test option.
   You can use Random test for:
 - testing your colleagues or students,
 - checking the percentage of items you remember (on  average
   you should score about 95%),
 - calculating the distribution of particular topics  in  the
   database or the distribution of  various  item  parameters
   like repetition number, number of memory lapses, etc.

Compress allows to perform the database  garbage  collection,
check and possibly correct database integrity errors.
   Compression is done in two stages:
 - compression of *.DAT and *.FTD files; in order to get  rid
   of outdated information about the process,
 - compression of *.ITM and *.ITI files.
Compression of the *.PRI file is done automatically  whenever
the queue of intact items  (priority  queue)  is  empty.  The
*.IFN file is never compressed.
   The first stage of compression is  faster  and  should  be
performed every month to avoid excessive growth of *.DAT  and
*.FTD files. The second stage  of  compression  is  necessary
only when a large  number  of  items  have  been  deleted  or
changed (edited). This stage lasts considerably  longer  even
if it is done on a hard disk or RAM-disk.
   It is advisable that you make back-up copies of your files
before compression. If you cannot compress your files because
SuperMemo stops reporting an error, your database  must  have
been damaged. In such a case try  compressing  your  database
using  RECOV5.EXE from  the  distribution  disk.  It  applies
extensive compression and  enables  elimination  of  a  large
class of errors.

Initialize allows to create a new empty SuperMemo database.
Use this option whenever your  existing  database  grows  too
large or when you want to start a database on a new, not  yet
covered subject.
   Each newly created database has its unique  name  and  its
particular files are identified  by  file  extensions.  After
selecting Initialize, you will be asked a name for  your  new
database. Type it in and press Enter. If any database is open
at the moment of initialization,  it  will  be  automatically
closed. Newly created files will be stored on the  drive  and
the directory specified by the  Path  option  in  the  Params
menu.

OS Shell allows to leave the  SuperMemo  program  temporarily
to  execute  commands  of  the  DOS  operating  system.
   The operating system command interpreter (COMMAND.COM)  is
loaded to the memory and executed. If the COMMAND.COM file is
not  present  in  the  directory  indicated  by  the  current
SuperMemo parameter setting (the path may be  different  from
the database directory path) you will be asked to input a new
COMMAND.COM path. Using  Write  parameters  from  the  Params
menu, save SuperMemo  parameters  to  make  the  new  setting
permanent.
   The database currently opened  by  the  SuperMemo  program
will automatically be closed when selecting OS Shell.
   The SuperMemo program remains in the  memory  and  can  be
reactivated by issuing EXIT command from the DOS level.

Message  option  displays  the  most  essential   information
concerning the SuperMemo World company and its products.

User presents user data including  the  registration  number,
name and address (if available at the moment of  purchasing),
date of issuing the copy and remarks concerning  its  special
properties (if any).

Params menu provides the following four options  allowing  to
define some configuration parameters and to save  or  restore
the configuration.

Sound switches on and off the generation  of  sounds  by  the
SuperMemo program.

Path defines the directory in which the  SuperMemo  databases
are stored.
   Use this option to specify both the drive and the path  of
the directory in which you want to open  your  database.  The
new specification can be  stored  and  restored  using  Write
parameters and Load parameters options.
   Changing  the  path  will  not  affect   the   access   to
directories  where  the  parameter  file,   help   file   and
COMMAND.COM are stored. By default all these directories  are
assumed to be the same as the current directory.

Load parameters loads the SuperMemo parameter set from a disk
file.
   The parameter set is stored on the  disk  in  the  SM5.PAR
file and contains the following:
 - a  boolean  variable  determining  whether  SuperMemo  is
   to generate sounds (the default value is true),
 - database directory,
 - help file directory,
 - COMMAND.COM directory,
 - parameter file directory,
 - the recently input search string (used by Find),

Write parameters stores the  current  parameter  set  on  the
disk.

Help displays selected pages of the help file.
The content of the help is stored in the SM5.HLP file. If you
want to access this file from a directory different than  the
current directory then:
 - remove the help file from the current directory,
 - select the Help option,
 - input the path to the help directory,
 - use Write parameters to make the setting permanent.

   Help has the form of a network of interrelated  groups  of
pages  presenting   information   about   particular   topics
concerning the SuperMemo method and program.
   Use PgUp and PgDn keys to change between successive  pages
describing a single topic. Highlighted fragments of the  text
are called selectors and indicate topics  on  which  you  can
find more detailed information in help. Use  cursor  keys  to
select one of the selectors and press Enter  to  display  the
linked page. Press Esc when you want to exit help. Next  time
you enter it, you will be presented the page displayed  prior
to  exiting.  Other  control  functions  available  in   help
include:
 - F1 - to get to the first page,
 - Alt-F1 - to return to the page recently displayed,
 - S - to search for a page  containing  a  given  string  of
   characters,
 - Ctrl -> - to get to the next page,
 - Ctrl <- - to get to the previous page.
   You  can  use  EXTRACT.EXE  program   available   on   the
distribution disk to extract the content of the help  into  a
printable *.DOC file.

Quit performs the following:
 - asks for the confirmation  that  you  want  to  leave  the
   program (press Y to confirm),
 - closes database files,
 - deletes all temporary files,
 - restores the screen image as it was  before  entering  the
   SuperMemo program.
   You should always  terminate  your  work  using  the  Quit
option to ensure that all  database  files  are  updated  and
closed.  Also   OS   Shell,   Initialize,   Date   and   Open
automatically close the database files.

5. SUPER-MEMO DATABASE

Each SuperMemo 5 database consists of  a  set  of  six  files
(with extensions IFN, ITM, ITI,  DAT,  FTD  and  PRI).  These
files are created  automatically  when  selecting  Initialize
option and updated each time the database is  being  modified
or closed after a SuperMemo session. The operation of closing
a currently opened database is performed  automatically  each
time you Quit, exit to DOS using OS  Shell  option,  Open  or
Initialize another database.
   Absence of any of the mentioned files will  make  openning
of the  database impossible. Therefore  you  are  advised  to
regularly make back-ups of all the files making up  all  your
databases. The content of database files is as follows:
 - <name>.IFN - contains basic information about the database
   (total number of items, number of memorized items, date of
   creation, date  of  the  last  reference,  burden,  factor
   distribution,  interval   distribution,   optimal   factor
   values, optimal factor support, etc). This file is  stored
   in a compressed form,
 - <name>.ITM - contains the text of items.
   Items are stored sequentially with questions  and  answers
   separated by the dollar signs.
   Deleted items are replaced with dollar signs.
   The record and the position where a given item starts  are
   stored in the *.ITI file,
   To view the structure of an *.ITM file use  the  DOS  TYPE
   command,
 - <name>.ITI - contains information about  individual  items
   (e-factor, last interval, number of repetitions,  pointers
   to the *.ITM file where the first character of the item is
   stored),
 - <name>.DAT - contains a list  of  pointers  to  the  *.FTD
   file,
 - <name>.FTD - contains a list  of  lists  containing  items
   which are to be repeated on a given day,
 - <name>.PRI - the priority queue.
   It consists of records  making  up  a  bidirectional  list
   defining the sequence of intact items.
   In SuperMemo 5, it  is  established  that  the  items  are
   stored in the priority queue in the  same  order  as  they
   have been appended to the database.  Unlike  SuperMemo  2,
   SuperMemo 5 does not provide the option for changing  this
   sequence.

   Making changes in SuperMemo database files  in  any  other
way than by means of the SuperMemo program  (or  other  tools
supplied by SuperMemo World) will almost certainly result  in
destroying the integrity of the database. Such a  damage  may
be irreversible. SuperMemo World takes no responsibility  for
damages  introduced  by  unauthorized  modifications  to  the
database files.

6. HINTS AND TIPS FOR SUPER-MEMO LEARNERS

Formulation of items.
The important part of the SuperMemo method is that  knowledge
should be divided into smallest pieces so that repetitions of
particular items interfere  with  each  other  as  little  as
possible. If an answer part of an item gets complex,  try  to
divide the problem covered by  it  into  several  subproblems
each expressed by a different item. Do not be afraid  of  the
number of items in your  database  growing.  The  better  the
knowledge structuring, the easier, quicker and more  pleasant
the repetitions.
   The most popular field  of  using  SuperMemo  is  learning
foreign languages.  The  word-to-word  or  definition-to-word
combinations, characteristic  for  this  kind  of  knowledge,
naturally conform to the minimum information principle.  This
makes the creation of vocabulary databases  relatively  easy.
However, beginners formulating their first items may find the
following two remarks instructive:
 - if you want to remember a foreign word actively, input  it
   as the answer and its definition or its equivalent in your
   native language as the question, for example if  you  want
   to learn Esperanto:

   QUESTION: an animal with a long trunk and a pair of tusks
   ANSWER:   la elefanto

  or

   QUESTION: an elephant
   ANSWER:   la elefanto

 - if want to remember a foreign word passively, input it  as
   the question and its earlier memorized synonym  remembered
   actively as the answers,  for  example,  if  you  have  an
   active item

   QUESTION: stimulus   (Eng)
   ANSWER:   stimulo    (Esper)

   you may input

   QUESTION: impulso    (Esper)
   ANSWER:   stimulo    (Esper)

Systematic work.
All calculations on  which  the  SuperMemo  method  is  based
assume that the learner allocates  time  for  learning  every
day. Lower frequency of testing,  although  acceptable,  will
result in descreasing the effectiveness of  the  optimization
procedure.
   Using SuperMemo, you can memorize a large number of  items
in a short time but remember that in the following  days  you
will have to repeat them. If you  work  irregularly  you  may
find yourself deluged with the stream of items scheduled  for
repetition. Adjust the amount of time spent with the  program
to your capacity.

Concentration.
SuperMemo forces your brain to work with high  intensity,  so
that you may even find it often hard  to  work  with  it  for
longer  than  an  hour.   Beware,   however,   of   negligent
repetitions. Concentrate and try to answer each item the best
you can. It is better to  lose  some  time  while  trying  to
recall a given answer than to repeat the item  ad  infinitum.
If, during repetitions, you meet an item particularly  often,
which is the case in items with low e-factors (1.3), think  a
little why the item is so hard to remember. The experience of
many says that in most cases  such  items  are  simply  badly
formulated or interfere in meaning with other  items  in  the
database. Reformulation is a simple solution here.

Quality assessment.
You should never neglect proper  assessment  of  your  answer
qualities. The learning optimization algorithm  is  based  on
calculation of difficulties of items, so that judging answers
is tremendously important. You  are  even  allowed  to  input
fractional values to make estimations more accurate.  Try  to
obey  the  rules  of  quality  assessment   quoted   earlier,
otherwise you will never get the  most  of  the  optimization
algorithm.
   Do not abuse quality grades 0 and 1! You are  unlikely  to
suffer a complete blackout.

Safety principles.
SuperMemo databases are a precious possession of the  learner
and should be guarded carefully against loss or damage. Thus,
the following principles should always be observed:
 - one or more back-up databases should be held in store,
 - SuperMemo should always be terminated by Quit,
 - compression of files should be made once a month (at least
   the first stage),
 - disks should not be swapped while using a database,
 - all database files should not exceed 300kB in size,
 - on no condition should databases be modified manually,
 - printing option should not be  used  when  no  printer  is
   connected (if chosen inadvertently, Esc should be  pressed
   after the PRINTER NOT READY message appears),
 - date setting should be kept accurate,
 - back-ups should be  used  in  case  of  power  failure  or
   resetting the computer for any other reason.

7. PROTECTION MECHANISMS

All protection mechanisms have been scrapped as of March 1992.

8. DISTRIBUTION DISK FILES

The distribution disks contain the following files:
 - SM5.EXE - the SuperMemo 5 Semiprofessional program,
 - SM5.HLP - help file,
 - TRANSF5.EXE - program  for  conversion  of  databases  to
   intact form,
 - RECOV5.EXE - program for an fault-tolerant compression of
   databases,
 - UP2_5.EXE - program to convert SM2  files  to  the  SM5
   format,
 - EXTRACT.EXE - program to extract  text  from  the  SM5.HLP
   file into a printable *.DOC file,
 - README.TXT - readme file,
 - OFFER.TXT - SuperMemo World's sales offer and  information
   about the SuperMemo World's Database Bank,
 - ORDER.TXT - SuperMemo World's order form,
 - INQUIRY.TXT - SuperMemo World's inquiry coupon,
 - DEVELOP.TXT - review of the development of  the  SuperMemo
   method and software,
 - SMPAP.TXT - description of the paper-based variant of  the
   SuperMemo method,
 - GROSSMAN.TXT - declaration of the Grossman Club,

For using the SuperMemo method, all files except SM5.EXE  are
optional.

9. OUTLINE OF THE OPTIMIZATION PROCEDURE

Understanding of this section is not required for  work  with
SuperMemo.

In SuperMemo 5, the learned knowledge is split into  smallest
possible pieces  called  items.  Each  item  is  repeated  in
intervals  that  are  computed  to  satisfy   the   following
criteria:
 - each interval should be as long as  possible  to  minimize
   the frequency of repetitions,
 - each interval should be short enough to yield the expected
   90% retention of knowledge at  repetition,  or  about  95%
   overall.

Because of  different  difficulty,  items  are  divided  into
difficulty categories.
In  order  to  classify  items  as  belonging  to  particular
categories, each of them is associated with a  number  called
the e-factor. This number has the initial value  of  2.5  and
decreases each time the  item  appears  to  be  difficult  to
recall (the judgement is based on the response quality  input
by  the  learner).  By  analogy,  e-factors   increase   when
associated  items  fare  well  in   repetitions.   Difficulty
categories are  determined  by  e-factors  spanning  the  0.1
range, e.g. <2.0,2.1), <2.1,2.2) etc.
   A process variable called the matrix of optimal factors is
used  to compute optimal  intervals   along   the   following
formulas:

       (1)      I(EF,1)=OF(EF,1)
       (2)      I(EF,n)=I(EF,n-1)*OF(EF,n)

  where:
   I(EF,n) - n-th optimal interval for item
             whose e-factor is EF
   OF(EF,n) - optimal factor corresponding to  the
              e-factor EF and repetition number n
              (note that memorization is  counted as the
               first repetition).

The matrix of optimal factors that comprises optimal  factors
for e-factors from 1.3 to 3.2 and repetition numbers  from  1
to 20 may be inspected by means of the option Stats : Optimal
factors  (only  e-factors  in  the  range  1.3  -   2.5   are
presented).
   The  initial  value  of  the  matrix  of  optimal  factors
corresponds to the most likely set of optimal  intervals  for
items of  different  difficulty.  However,  as  the  learning
process  proceeds,  particular  entries  of  the  matrix  are
modified in order to comply with  the  optimization  criteria
mentioned earlier (i.e. 90% retention at repetitions).
   Partitular optimal factors increase if  the  retention  is
higher than 90% and decrease as it drops below 90% (retention
is  measured  for  different  combinations  of  two   values:
e-factor and repetition number).
   The current estimation of optimal interval  based  on  the
matrix  of  optimal  factors  can  be  viewed  with  Stats  :
Prospects.
   The longer the interval, the slower  the  modification  of
the relevant optimal factor and the slower its convergence to
the optimal value.

10. ABOUT SUPER-MEMO WORLD

SuperMemo World is the exclusive  owner  of  all  intelectual
property  rights  to  the  SuperMemo  method,  software   and
publications.

   At present the following versions  of  SuperMemo  software
are distributed:
 - SuperMemo 2 Public Domain
 - SuperMemo 5 Shareware
 - SuperMemo 6 Professional for DOS
 - SuperMemo 7 for Windows
See the OFFER.TXT file from the  distribution  disk  for  the
full list of currently available SuperMemo World products.

   SuperMemo  6  provides  the  following   improvements   as
compared to SuperMemo 5:
 - improved   optimization    technique    for    computing
   interrepetitory intervals based on lossless record of  the
   history of repetitions,
 - definable level of knowledge retention,
 - improved user interface,
 - increased fault tolerance,
 - substantially expanded and  improved  set  of  statistical
   options and parameters,
 - context-sensitive help, etc.
See DEVELOP.TXT for the full list of modifications introduced
to SuperMemo software since releasing SuperMemo 5.

11. GLOSSARY

answer -  (in  the  SuperMemo  method)  second  part  of  the
   SuperMemo item (compare response)
answer pane - part of the SuperMemo  main  screen  where  the
   answer is displayed
answer quality - see response quality
append - option of SuperMemo 5  that  allows  to  add  a  new
   question-answer pair to the SuperMemo database
to append an item - to add  a  question-answer  pair  to  the
   SuperMemo database
browsing mode - state of SuperMemo 5  in  which  an  item  is
   displayed in the Question-Answer panes and  the  following
   commands are available:  edit  item,  delete  item,  print
   item, reset item, select another item, etc.
burden - SuperMemo database  parameter  which  estimates  the
   average number of items that have to be repeated per  day.
   The formula is:
              1      1             1
      burden=---- + ---- +    + --------
             I(1)   I(2)        I(Total)
      where:
         I(n) is the current interval of the n-th item
         Total is the total number of items in the database
compression - (in SuperMemo 5) two-stage  garbage  collection
   procedure that  removes  superfluous  data  from  database
   files
daily burden - option that allows to see how many  items  are
   to be repeated on particular days of the learning process
database - (1) collection of files that store  all  important
   data about SuperMemo items and about the learning process,
   (2) collection of SuperMemo items stored in one SuperMemo
   database[1]
database files - 6 files that make up the SuperMemo  database
   (their extensions are IFN, ITM, ITI, DAT, FTD and PRI)
difficulty (of an item) - degree to which a given item causes
   recall  problems   in  the  process   of   learning.   The
   estimated value difficulty of a given item is expressed by
   means of its e-factor
current interval - interval that is currently in force for  a
   given item.
   It indicates the number of days that must pass between two
   points in time:
    - last repetition of the item,
    - next repetition of the item.
e-factor - number associated with  each  of  SuperMemo  items
   which estimates its difficulty. The  e-factor  belongs  to
   the range 1.3 - 3.5 and the greater it is, the easier  the
   item
factor - see e-factor
factor used -  quotient  of  two  subsequent  interrepetitory
   intervals:
                   I(n)
   factor_used = -------
                  I(n-1)
   where: I(n) - n-th interval used in repetitions of a given
   item.
   Note that if the new interval  differs  from  the  optimal
   interval then the factor used will differ from the optimal
   factor. Even if new interval equals optimal interval,  the
   factor used may be higher than optimal factor in case  the
   repetition has been delayed (e.g. by the  learner  failing
   to pass the test option on one of the days)
initialization - procedure in which a  new,  empty  SuperMemo
   database is created
intact - SuperMemo database  parameter  which  indicates  how
   many items have been appended but not yet memorized
intact item - item that has not yet been memorized  (opposite
   to memorized item)
interval  -  period  of   time   separating   to   successive
   repetitions of an item
item - (in SuperMemo method) question and answer pair used in
the process of learning
item editor - SuperMemo software facility that allows to edit
   questions and answers of individual items
item pane - part  of  the  SuperMemo  main  screen  in  which
   information  about  the  currently   processed   item   is
   displayed
lapse - see memory lapse
matrix of optimal factors - collection of optimal factors for
   varying e-factor values and different repetition numbers
to memorize an item - to commit an item to memory by means of
   the test option
memorized - SuperMemo database parameter that  indicates  how
   many items have already been memorized
memorized item - item  that  has  already  been  committed to
   memory by means of the test  option  (opposite  to  intact
   item)
memory lapse - instance of  forgetting  an  individual  item,
   i.e.  scoring  less  that  3  in  the   response   quality
   assessment
memory lapses - item parameter which indicates how many times
   the considered item has been forgotten. The memory  lapses
   parameter  is  displayed  in  the  item  pane  aside   the
   repetition number (separated by a colon)
minimum  information  principle  -  (in   SuperMemo   method)
   principle of formulating  items  that  are  as  simple  as
   possible, i.e. items that contain minimum information
modified OF - see modified optimal factor
modified optimal factor - new value  of  the  optimal  factor
   that has just been computed on the base  of  the  recently
   performed repetition
new factor - new value of the e-factor  that  has  just  been
   computed on the base of the recently performed repetition
new interval - new current interval computed for an  item  on
   the base of the recently performed repetition
next repetition - date at which  the  next  repetition  of  a
   given item is scheduled
optimal  factor  -  SuperMemo  database  parameter  that   is
   characteristic  for  items  of  a   given   e-factor   and
   repetition number.
   For the repetition number equal 1, the optimal factor
   equals the first optimal interval in days.
   For repetition numbers greater  than  1,  optimal  factors
   indicates how many time the  optimal  interval  is  longer
   than the recently used current interval
optimal interval - (in SuperMemo 5) interval that yields  90%
   retention at repetitions, i.e. interval which  makes  that
   90% of items are remembered at the moment  of  repetition
   (i.e. score at least 3 in response quality assessment)
optimization - (in SuperMemo method) process by which the set
   of e-factors of individual items, as well as the matrix of
   optimal  factors  are  modified  in  order  to  yield  90%
   retention  at  repetitions  (equivalent  to   optimization
   algorithm, optimization procedure etc.)
optimum interval - see optimal interval
outstanding - SuperMemo database parameter which  equals  the
   number of items that are scheduled  for  repetition  on  a
   given day. If the learner has not been repeating items for
   few  days,  the  outstanding  parameter  values  acumulate
   additively
overall retention - see retention
previous OF - see previous optimal factor
previous optimal factor - value of the  optimal  factor  from
   before the recently performed repetition
priority  queue  -  sequence  of  intact  items  waiting  for
   memorization. In  SuperMemo  5,  the  sequence  cannot  be
   changed by the learner, and items are scheduled  in  order
   of appending
process pane - part of the main  SuperMemo  screen  in  which
   information  about  most  important   SuperMemo   database
   parameters  is  presented  (e.g.  memorized,  outstanding,
   total, etc)
quality - (1) SuperMemo database parameter  that  equals  the
   average response  quality  in  a  given  session  (2)  see
   response quality
question  -  (in  the  SuperMemo  method)  first  part  of  a
   SuperMemo item
question pane - part  of  the  SuperMemo  main  screen  where
   question is displayed
remains - SuperMemo database parameter  which  indicates  how
   many  items  scored  below  3,  and  were  scheduled   for
   additional drill in the same session
repetition -  process  in  which  the  learner  performs  the
   following operations in turn:
    - displaying the question
    - responding to the question (without typing it in)
    - displaying the answer
    - comparing the answer with the given response
    - inputting the response quality (number from 0 to 5)
   Note that memorization is considered the first  repetition
   despite the  usual meaning of the word repetition
repetition number - number  of  the  repetition  of  a  given
   item. Memorization corresponds to the repetition number 1,
   repetition after the first interval corresponds to 2, etc.
   In case of a memory lapse (response quality lower than 3),
   the repetition  counter  is  reset,  i.e.  the  subsequent
   repetitions are numbered again starting with 1
reset - one of the  operations  of  the  browsing  mode  that
   allows to convert a memorized item into an  intact  item.
   The e-factor assumes the value of 2.5,  repetition  number
   becomes 1 and the memory lapses parameter is set to 0
response - recalled form of the correct answer that  appeared
   in the learner's  mind  in  the  course  of  a  repetition
   (compare answer)
response quality - (also quality) number from the  range  0-5
   that is the  learner's  score  for  the  accuracy  of  his
   response, i.e. how well the response  matched  the  answer
   (see the grade scale p. 7)
retention - (also overall retention) proportion of  memorized
   items that are remembered by the learner at a given moment
   (in  an  established  SuperMemo  process   supervised   by
   SuperMemo 5, the overall retention ranges from  94-96%  in
   most cases) (compare retention at repetitions)
retention at repetitions - proportion of memorized items that
   are remembered at the moment  of  their  repetition.  Note
   that  retention  at  repetitions  is  lower  than  overall
   retention (e.g. in SuperMemo 5 it averages 90%)
SuperMemo database - see database
SuperMemo item - see item
SuperMemo learner - learner who applies the SuperMemo method
SuperMemo method - method of fast learning based  on  minimum
   information  principle  and  scheduling   repetitions   in
   optimal intervals. Optimal  intervals  are  determined  by
   means  of  the  SuperMemo  optimization  procedure   which
   differs in particular software lines.
   The  basic  principles  of  the  SuperMemo   method   were
   developed by a Polish biologist P.A.Wozniak in  the  years
   1982-1985  and  its  computer   applications   have   been
   investigated since 1987.
SuperMemo software - computer  programs  that  implement  the
   SuperMemo method and all the accompanying software tools
SuperMemo World - world's sole distributor of  the  SuperMemo
   software and publications.  The  exclusive  owner  of  all
   intelectual property rights to the SuperMemo method
test  -  option  of  SuperMemo  5  that  allows  to   perform
   repetitions of items  (including memorization)
test pane - part  of  the  SuperMemo  main  screen  in  which
   information pertaining to a given repetition is displayed
total -  SuperMemo  database  parameter  that  indicates  the
   number of items stored in the database

12. REFERENCES

Wozniak P.A. 1990, Optimization of learning. A new  approach
   and computer application.
Wozniak P.A. 1991, SuperMemo 5 Semiprofessional. Help file.
Wozniak P.A. 1991, SuperMemo 5 Semiprofessional. Introduction.

====================================================================

       SUPER-MEMO WORLD
       P.O. Box 1
       Poznan 48
       POLAND
       Tel/Fax (48) 61 203157
       e-mail: JAM@BRAHMS.UDEL.EDU
