The Datafile PD-CD 5/DATAFILE_PDCD5.iso/demos/ko2/Tutorial

Text File | 1993-04-02 | 21.6 KB | 455 lines

KO II Read Only Version
=======================

COPYRIGHT NOTICE: This Read Only version of KO II is NOT public domain. The
copyright of the program resides with Clares Micro Supplies. However, you
may distribute the program with your own files for non commercial
applications. If you wish to sell your files and use the read only version
then you must obtain a licence from Clares in order to do this. You must
also supply this file with the program together with the KO II file
"Quotes" supplied on the master disc.

Getting Started
===============

These instructions take the form of a simple search tutorial using the
file "Quotes' supplied with KO II. We simply show you the program, and how
it works. At this point you should be looking at the directory display
holding the KO II application.

Load the KO II icon onto the icon bar in the usual RISC OS manner - a double
click of the Select button on the KO II application.

Next display and drag the file icon "Quotes" onto the KO II icon on the icon
bar. A window should open, headed "Selection of Humorous Quotations". Quotes
is a KO II Subject or file. It is an unusually short file, designed simply
to demonstrate the program. It contains about eighty quotations, with
referencing data which allows you to find these quickly.

The word Subject will be used throughout this tutorial to refer to a body of
related data filed together and accessed using KO II. A Subject could be
anything from "My last voyage around the world" to "Correspondence from
America".

LINKER WINDOW
=============

The Linker window is the central KO II window and appears with every KO II
Subject. The main KO II menu will appear if you place the mouse pointer
anywhere on it, and then click the Menu button. Remember the term Linker
window as it is very important - this window links all the data together.

The two white rectangles on the Linker window are fields - areas where data
is typed and displayed. These two Linker fields - Context and Source - are
standard with all KO II Subjects, but a varying number of optional fields
may be added when setting up a Subject. To show you what these two standard
fields do, we merely need to look at one of the quotations filed in Quotes.
Click on the "Goto" button on the Linker window. A small dialogue box
appears headed Goto text item. Below, you will see the figure 1. Change this
to 10, and then click on OK. Immediately, the KO II Text window will open,
displaying Joseph Heller's celebrated explanation of Catch-22.

CONTEXT CODES
=============

KO II is designed specifically to file economically, and then quickly find,
text items such as this. Read this item if you wish. From now on we use the
term text item or item to refer to these pieces of text.

Now look at the Context field in the Linker window. You will see a string of
codes, separated by the slash character: C22/WAR/MI/USA. These are Context
codes, each of them stating one of the main ideas under which the Catch-22
item has been filed. Items are most quickly found in KO II by searching this
field for a code representing the area of interest that you want to recall.

What do these four codes represent? Three are perhaps fairly obvious - C22,
WAR, USA. But what does MI mean?

To find out, bring up the main menu by clicking the Menu button with the
pointer anywhere in the Linker window. Then click on the menu option
Contexts. Immediately another window will appear, the Contexts window.


This window displays all the key ideas or Contexts under which the text
items in a Subject are filed. A separate menu - the Contexts menu - will
appear on this  window if you press Menu with the pointer in the Window.

We want to find out what MI means. As the codes are arranged in alphabetical
order, we could simply scroll the Contexts window. However, let's instead
get KO II to find the code for us. Bring up the Contexts menu and choose
"Search" by moving the mouse pointer onto the small arrow to its right.
Immediately the Contexts Search dialogue box opens. It allows you to search
in any of four fields.


As we know the code (MI) and want to find out what it means, simply type MI
into the Code field and click on the Go button.

Immediately the Contexts window will scroll to display the code MI at the
top - and you will see that it means "mental instability". The Catch- 22
item has been filed with this code in the Linker window, because the
definition of Catch-22 involves the concept of 'craziness'. Many humorous
quotations refer to this idea, and use of this code in the Context field
means that all such quotations can be quickly collated.

What we have learned so far:

1       KO II is designed to file and then find text items;

2       Each item filed is accompanied by a string of Context codes which
        appears in the Contexts field of the Linker window;

3       This field can be quickly searched to locate items;

4       Every Context code is explained in the scrolling Contexts
        window, and can be found by doing a search from the Contexts menu.

Why not simply reference the text item using words selected from the item
itself ? Hopefully, the code WAR will be a sufficient explanation. Although
Heller's novel is about the irrationality of war, and Catch-22 has
essentially to do with war, the word war does not occur in this quite
central excerpt from the book. This will be true of almost any important
piece of text: language is so rich, so full of synonyms, so capable of
implicit meaning, that the facility to reference text by external codes or
keywords is essential if we are to collate all related items. However, KO II
keeps an index as well, which can be used to identify an item's explicit
content during the filing process.

Why use codes instead of full words or phrases ? To economise. All other
quotations in this Subject which relate to mental instability are also coded
MI, but the explanation of MI occurs only once - in the Contexts list. No
other code will occur which has the same meaning. So, a search for the code
MI will collate all of these items.

Longer text items are likely to contain a large number of key points which
require storing as Context codes. The use of codes rather than words makes
it possible to file these longer items economically.

Context codes, then, relate to the contexts, or circumstances, in which you
may later be looking for the item. USA was used, for example, to indicate
the possibility of recovering all quotes originating in the United States of
America. A Context can be any connection, however personal, between the text
item and your need to recall it for any reason.

SOURCE FIELD
============

Look at the Source field in the Linker window, which contains the code JH.
To confirm that this stands for Joseph Heller, bring up the main KO II menu
once more and click on Sources. Immediately the Sources window will appear.

This window does for the Sources what the Contexts window does for the
Contexts - it explains or describes all source codes, displayed in
alphabetical order. When this code has been explained once, it can be used
subsequently any number of times in the Linker, depending on how many Heller
items are filed. The Sources window also has a menu attached, and this too
includes a search option.

Searching for related items
===========================

Let us suppose you now wish to find all quotes
relating to "marriage". Bring up the Contexts window and menu and choose
Search. Type marriage on the Explanation line and click on Go. Immediately
the Contexts window should scroll, to show the code MAR at the top.

Look more closely at the data accompanying this code. The explanation
marriage is followed by two other lines:

        Hierarchy:      >BMY/MMY
        Index words:    marri*, marry*, wed

Index words is fairly clear: any of these three words might occur in a text
item relating to marriage. Typing them onto this line enters them in the
index for this Subject, and this index can be used to search a text item in
preparation for filing.

The Hierarchy line tells us that there are two codes relating to marriage
(BMY - bigamy; MMY - monogamy). It also tells us that MAR is more generic
than these codes.

KO II is designed to allow subdivision of codes as data grows, to allow more
precise location of data. However, the more generic relationship is still
retained by linking these codes in the Hierarchy field for MAR.

To see this working, let's do a search using MAR. To tell KO II that we are
interested in this code, click on it so that it is highlighted. Next, we
need to clear the visible text item: it is actually a subset of one item.
Bring up the main KO II menu by pressing the mouse Menu button with the
pointer over the Linker window. Enter the Subset menu, and then select
Clear. Confirm that you wish to clear the subset. We can now do our new
search for MAR.

Click on the Search button on the top right of the Linker window.
Immediately, the main Text Search window appears. This is used to locate
text items - using Context or Source codes, or data appropriate to a
user-created field, or a word in the Index, or even by searching the text
items directly with any word or phrase.

We want to search for items using the Context code MAR. As most searches are
done by using the Context field, it is given as the default search field on
the top left of the window. Click in the field below Search Value(s). Now
type ? in this field, and press Return. A small menu will appear, holding
the code already selected - MAR. Click on MAR to move it into the search
field. Type the operator <* to the left of the code, so that it becomes
<*MAR. The prefix <* means find all more specific codes and search for these
also, such as BMY and MMY. We do not need to specify these - KO II locates
and uses them automatically in its search.

Look now at the bottom panel, which tells KO II where to search. The Search
Whole File (Add to subset) option should be selected. If not, select it.

Now click on the Go button at the bottom left of the Text Search window.
Immediately a small dialogue box will appear, and show you the progress and
success of the search. You will be told that seven items have been matched.
In other words, MAR or BMY or MMY have been found in the string of Context
codes, referencing seven different items. The first of these is immediately
displayed. Notice that although both BMY and MMY occur in the Linker's
Context field for this item, the code MAR does not.

Now, click on the button Forward in the Linker window. The second item found
will be displayed. Together, the buttons Forward and Back, allow you to move
through the items in sequence in either direction. These seven text items
are a subset: i.e. a number of items selected from the whole file of items.

Let us suppose that we wish to compare one of these items with another,
simultaneously. Click on the Hold button. The current item will be displayed
in a second Hold window. Now click on Forward or Back. The adjacent item in
the subset will replace the held item in the main Text window. The two items
may be compared on screen. Notice that the item in the Hold window cannot be
edited - it is simply there for comparison. Close this Hold window now to
tidy the screen.

Now let us suppose we are particularly interested in two of these seven
items, and wish to select them for export to another program, dropping the
other five. Click on the Tag button. This selects the current item as a
marked or tagged item. Now go Forward or Back and Tag another item.

Now bring up the main KO II menu on the Linker window. Move the pointer down
to Subset and then onto the arrow to bring up the Subset menu. One of the
options is "= Tags". Click on this. Immediately the subset is reduced to the
tagged items, dropping the other five.

Suppose you have forgotten how you selected these two items. Bring up the
Subset menu again and click on "History". A window will open, showing the
series of steps taken to arrive at these two items.

We will perform one last task with this subset of two items. Let us suppose
that you wish to move them out of KO II into your favourite word processor
or DTP package. If you have sufficient RAM, both programs may reside
simultaneously in memory. If not, you can simply move your subset onto a
directory display. In either case the procedure is much the same. Once you
have your destination display visible (DTP or WP window, or directory
display), select the menu option Subset=>Export, type a filename, and drag
the file icon as normal into your destination display. Answer "No" to the
question "Code text items for auto-import?" and then click on Export. The
two quotes will be exported as a text file together with relevant
information on the file. Again, let us summarise what we have learned:

1       Text items are referenced by a number of Context codes, and by  one Source code;

2       These codes are explained in separate Contexts and Sources Windows;

3       The data in these windows is searchable;

4       Context codes can be continuously subdivided for greater precision;

5       KO II can 'remember' particular codes if you highlight them in  the
        Context window;

6       The Search button on the Linker window will display the main text
        search window;

7       ? followed by Return in the Context search field will reveal the
        highlighted Context code or all codes if none is highlighted;

8       Text items are quickly found by clicking (or typing) the appropriate
        code into the search field;

9       The operators <* can be used to recover all more specific data,
        even when more specific codes are not explicitly used in the search;

10      The buttons Forward and Back on the Linker window allow you to move
        through the items in a subset;

11      The Hold button in the Linker window allows you to hold the
        visible item in the subset for comparison on screen with any other
        item in the subset;

12      The Tag button in the Linker window marks or tags the visible text
        item;

13      Untagged items in a subset may be dropped from it by selecting
        the Subset = Tags menu option.

14      At any stage you can review the steps taken to produce the
        current subset by selecting the Subset menu option, History.

15      A subset is any number of items selected from the total number in
        the Subject.

16      Subset=>Export allows you to move a subset of text items out of  KO
        II as a text file, for immediate or later use in another
        application. The KO II Index The last important set of data in KO II
        is the Index. It is an alphabetical list of all keywords which might
        occur in the text items to be filed in a given Subject. It is linked
        to the Context codes by means of the Index line in the Contexts
        window, and can be used to find these codes. However, its most
        important function is to search text items for filing. When KO II
        matches any word in the text item with any Index word, it displays
        both the word and the Context code associated with it, allowing the
        code to be clicked and transferred into the Contexts field in the
        Linker window.

INDEX WINDOW
============

Let's take a look at the Index window, by bringing up the main menu and
clicking on Index.

You will see straight away that the index allows use of the 'wildcard'
character, *. This means that the index word in question will be matched
with any word beginning with the typed characters, whatever its ending. So,
account* will be matched with accountant, accounting, account, accounts.
This facility saves repetitive typing of words with a common root, and
reduces the length of the Index.

The Index window too has its own associated menu. Words may be added to the
Index directly using Index=>Add. They may also be added when defining a
Context code, simply by extracting keywords from the Context code
explanation.

For speed when searching text using Index item, KO II treats a number of
punctuation characters - including hyphens - as equivalent to spaces. So, if
it is intended to match text occurrences of, say, 'left-wing', leave out the
hyphen when adding the word to the index. The index entry 'left wing' will
match text occurrences of both 'left wing' and 'left-wing', whereas the
index entry 'left-wing' will match neither. Let's now see how the Index can
be used to search for text items. First, clear any current subset, using
Subset=>Clear from the main menu. Next highlight the index word advertis* in
the Index window by clicking on it.

Now bring up the Text search window by clicking on the Search button in the
Linker window. Go to the left "Index Words" field on the text search window,
and type ? Return. A menu will appear, showing the word advertis*. A single
click on this will move it into the Index Words field.

Checking that the "Search Whole File (Add to subset)" option is set on the
Text search window, click on Go. KO II will find three items and display the
first. The key things we have learned about the Index are:

1       It is an index of words which may occur in the text items to be filed;

2       It is linked with the Context codes by means of the Index Words line
        in the Contexts window;

3       It is used to search text when filing, providing the Context codes
        associated with any index words matched in the item;

4       It has its own window and associated menu;

5       Index words may be highlighted and then moved into the KO II Text
        Search window, to locate text items filed using the Context codes
        associated with the chosen index words.

Finally, we will undertake a slightly more complicated search, to find items
which relate to two different search values.

Let us suppose we want to locate a dimly remembered quote from W.C.Fields in
which he uses the word 'midget', which is not in the index. This means we
will need to search the text items directly for the word 'midget'. This is
by necessity a slower process than searching for occurrences of Context
codes in the Linker fields, so we do not want to search the text in all
items for 'midget' if we can help it. We proceed by first selecting all of
the W.C. Fields items, as follows:

1       Search for the word 'Fields' in the Source window, using
        Sources=>Search, and then typing 'Fields' into the Description
        field;

2       Highlight the code WCF by clicking on it;

3       Bring up the main Text Search window by clicking on the Search
        button on the Linker window;

4       In the Linker Field(s) panel, click on the arrow icon to the
        right of the word Context. This will display a menu of two field
        names, Context and Source. Click on Source to select the Source
        field for search;

5       Type ? Return in this field, and then click on the provided
        code WCF to move it into the Source field;

6       Check that the Search Whole File (Abandon subset) button is set in
       the Text Search Window, and click on Go. Note that this time we used
       the option to Abandon the current subset before searching thus
       eliminating the requirement to use Subset=>Clear first.

You will find that there are six W.C.Fields quotes. We want to search these
items only for the word 'midget', so we return to the Text Search window.
This time we go to the Text Body Direct panel and type 'midget' in the first
of the three fields underneath the words Search Value(s). You will notice
that by default, the search will be correctly conducted in the whole of the
text items, although we could, if we wished, search just the first n words
of each item.

One last adjustment needs to be made to the Text Search window. Click on the
Search Subset button, as we want to search only the subset of W.C. Fields
items, not the whole file. Then click on Go again. The item required will be
found very rapidly.

From this exercise we have learned:

1       We can select the Linker's Source field, or any Linker field, for a
        search by selecting the appropriate field name from the menu
        provided for each Linker field in the Text Search window;

2       To search for items which relate to two or more ideas (in this case
        W.C.Fields AND 'midget') we can make successive searches, returning
        to the Text Search window each time;

3       Direct searches of the text items themselves are slower than
        searches of the fields in the Linker window, so should be left till
        last;

4       When searching a subset of items, we must select the Search Subset
        button in the Text Search window;

5       We can opt to search through the first n words in text items when
        doing a direct text search (this is appropriate when searching long
        items which refer to their key subjects in the first paragraph or so
        - e.g. news data).

This completes your introduction to KO II. Hopefully you will have learned
that the entire program is designed for easy filing and fast, powerful
access to text oddments of any kind. And you should now also understand the
main menu structure.

Three of the options (Contexts, Index, Sources) provide windows allowing you
to access these separate lists. Create is for setting up your own files. Add
text opens a window into which you can type or import new items for filing.
Finally, Subset allows you to perform various operations upon a subset of
items selected from the whole file.


THE SALES MESSAGE
=================

If you have found this Read Only version of KO II useful and interesting
then you should consider purchasing the full version so that you can edit
files and also create your own.

Knowledge Organiser II can be purchased from:

Clares Micro Supplies
98 Middlewich Road
Northwich
Cheshire
CW9 7DA
UK

Tel. [phone]
Fax. [phone]


Price: £99.95 inc VAT @ 17.5%