In Russian: Latin letters DOS Windows Unix KOI-8 Unix ISO-5
In English:         ASCII
Institute for High-Performance Computer Systems RAS

Leonid Ovanesbekov [94b-96]

Hypertext Classification 2Q

Chapter from the candidate dissertation, revised version
Alternate headings for the given article:

2) Concept "a kind of hypertext"
3) Text Classification 2Q

Contents list

1. Concept "a kind of hypertext"
2. Classification 2Q
  2.1. List of kinds of texts "2Q-list"
  2.2. Catalogue of kinds of texts "2Q-catalogue"
3. Postanalysis for classification 2Q
  List of entities "2Q-entities"

1. The concepts "kind of text" and "kind of hypertext"

1.1. the main idea

The texts (and hypertexts) can be different by their internal logic structure. For example, the dictionary has one structure, textbook - other, and hypertexting catalogue, for example, spare details for car, third. Even such fluent and brief enumerating - dictionary, textbook, catalogue, put an idea, that the texts can be different by "kinds". But concept a "kind" is very dim, it don't informs us. For the concept "a kind of texts" has become meaningful, sharp and useful, it is necessary to find some idea, some certain point of view, by which we could see different kinds of texts distinguished from each other.

This idea is applicable to the texts in general, but is especially brightly revealed in relation to the hypertexts. Therefore we shall simultaneously define two concepts: "a kind of texts" and "a kind of the hypertext" on the basis of this idea.

The following point of view is offered. Some texts are naturally broken into separate topics (sections, the chapters, themes) and there is a some logic "basis", some semantic structure, connecting separate topics in whole text.

The hypertext, contrary to the texts in general, are already broken on separate topics by definition and there is something, enabling to speak about these subjects as about the whole, as about the hypertext.

For looking at the texts and hypertext from the offered point of view, it is necessary to set the questions:

a) Which sort the information is contained in separate topics?
b) What is the "logic basis"?

Or, in other form:

a) Which type of a topic make the majority?
b) Which structure is reflected in the text and connects separate topics in one whole?
Each particular original pair of the answers on these questions are particular "a kind of texts" or "a kind of the hypertext".


1.2. the experiment

We shall conduct microexperiment: let try to represent and to describe two parts of our viewpoint - "main type of topics" and "the logic basis" in the three set forth above kinds of texts (dictionary, textbook, catalogue).

Dictionary
Main type of topics: Descriptions of words.
The logic basis: Set of words, requiring descriptions..

Textbook
Main type of topics: Sections of the textbook, stating an items of information. The logic basis: Structure of a exposition of knowledge, from simple to complex.

Catalogue (for example, for films)
Main type of topics: Descriptions of instances any category (description of particular films ).
The logic basis: A collection of these instances (set of films ).

Our microexperiment has appeared successful: what we tries to describe as a "Main type of topics" and "The logic basis" was relatively easily to formulate. The resulting description reasonably well described the essence of the dictionary, textbook and catalogue just as kinds of texts, or as kinds of hypertexts, if the dictionary, textbook, catalogue is presented in the hypertext form.


1.3. not any two question can become the basis of classification

We shall make one methodological remark.

The basis of our classification is above-stated two question (the name "2Q" is "2 questions").

Certainly it easily to imagine, any 2, 3, 5 and up questions to the texts. But if we want that these questions will be the real basis for speaking about "kinds" of texts and about classification, it is necessary that these questions were "essential", i.e. the different answers to these questions would give reasonably different and pithy kinds of texts (hypertexts).

Besides, it is important, that these questions helps to predict new kinds of texts (hypertexts), selecting new nontrivial of answer-values for these questions-parameters.

The author considers, that the classification 2Q satisfies to these requirements. It is a main hypothesis of this small research. And the quality of the hypothesis can be evaluated on that, how many different and interesting particular kinds of texts it is possible to offer within the framework of the given main idea.


1.4. classification as a collection of the prominent answers on the two questions

The classification 2Q consists not only of two questions, but also of the list of the possible answers, i.e. the list of various kinds of hypertexts.

The author will use a viewpoint' method of a collection: naming the list of possible kinds of hypertexts as a "collection", thus implicitly it is supposed:

  • a) The author tried to select the best instances for the collection (i.e. the most prominent kinds);
  • b) At a opportunity, the author placed there and some "marginal", (i.e. not absolutly usual) instances;
  • с) This collection is extensible;
  • d) The author does not pretend for the objective explanation why the collection has just such structure (especially in connection with item (c)). I.e. this collection is made empirically and allows further structuring (and it is possible, a few different structuring).

    Except the list of the names of kinds of hypertexts (in section 2.1.) is given more extensive description of each kind. The list with the description of an each element the author names as the "catalogue".

    Thus, 2Q hypertext classification offered to you is the system from 4 objects:

  • small research around of this subject in the form of the article (given article),
  • two central questions,
  • list of kinds of hypertexts,
  • catalogue of kinds of hypertexts.

    For convenience of the references we have given separate names for the list, catalogue and pair of questions.

    List of kinds гипертекстов we have named "2Q-list", catalogue - "2Q-catalogue", pair of questions - "2Q-questions". The name "2Q" we have left for the classification as a whole and for the given subject of research.

    2. Classification 2Q

    2.1. list of kinds of hypertext "2Q-list"


    2.2. catalogue of kinds of hypertext "2Q-catalogue"

    Textbook
    Main type of topics: The chapters, stating items of information on a some subject.
    The logic basis: Structure of a teaching, from simple to complex.

    Reference book
    Main type of topics: The chapters, stating items of information on a some subject.
    The logic basis: Structure of the subject area.

    Scientific paper
    Main type of topics: The chapters and subchapters of the article usually presents various aspects of the stated subject.
    The logic basis: Structure of a exposition and structure of the stated subject.

    Paper with annotations
    Main type of topics: Annotations and comments to various places of the article.
    The logic basis: Structure of the paper.

    Explanatory dictionary
    Main type of topics: Topics of dictionary explained terms.
    The logic basis: The set of special terms in the area of knowledge.

    Translator dictionary
    Main type of topics: Topics of dictionary with translation of the words or phrases.
    The logic basis: The set of words in the language selected by the some criteria.

    Documentation
    Main type of topics: Topics and chapters describing an aspects of the complex object.
    The logic basis: The structure of the describing object.

    User guide
    Main type of topics: The sections, describing actions, which can be done by user in the complex program (or by the complex device).
    The logic basis: (1) Structure of the user interface of the program: windows, menu, buttons. For physical devices - keys, switches, handles of management and etc.
    (2) Structure of the program (or complex device ).

    Administration guide
    Main type of topics: Sections, describing actions on support of given program (or complex device) and main set up structures.
    The logic basis: (1) Structure of work of the person on support of the program (device) in working order.
    (2) A structure of especially important parts of the program (device), from which work of the program depends.
    (3) A structure of cause-and-effect links, connected with functioning of the program (device).

    Encyclopaedia
    Main type of topics: The articles, stating item of information.
    The logic basis: (1) Structure of the system of the subject areas,
    (2) Set of key concepts in each subject area.

    System of tables
    Main type of topics: Tables
    The logic basis: A structure of the subject area from the viewpoint: Which sets of objects, lists, tables are in the subject area?

    Data base of Pictures
    Main type of topics: Drawings, graphic images.
    The logic basis: The subject and set of objects, which is possible to represent as a drawings.

    Spatial maps
    Main type of topics: Maps or planes of various sites of the space.
    The logic basis: The structure of the described space.

    Abstract model
    Main type of topics: Description of a some elements of the model.
    The logic basis: The structure of the simulated object.

    Catalogue
    Main type of topics: Descriptions of the instances from the catalogues.
    The logic basis: The set of instances assembled in the collection.

    Chronology
    Main type of topics: Descriptions of the events.
    The logic basis: Sequence of the events in a time.

    Hypertexting knowledge base
    Main type of topics: Heuristics and rules.
    The logic basis: Structure of the complex work, executed by a person.

    Program
    Main type of topics: The texts of separate procedures, functions, types of data, descriptions for variables, comment to fragments of the program.
    The logic basis: Structure of the program.

    3. Postanalysis for the classification 2Q


    3.1. the "postanalysis" concept

    Above in the given article we have offered the idea of the classification for texts and hypertexts, have presented the information object embodied the idea, have given for the information object the name "Classification 2Q".

    Now the classification 2Q "exists".

    At the beginning of the paper the author tried do not deflect attention to adjacent aspects for to present given idea compactly and clearly. Now, when the classification 2Q is announced, it is possible to try find interesting ideas at the wider field: "How we can to look at the classification 2Q from other points of view", "How the classification 2Q communicates with other close subjects and etc.. It is other kind of the analysis, it differs from the analysis in the beginning of the given article in chapter 1. As in the beginning of the article the author tried to justify occurrence of the new information entity, as now it is possible to offer ideas, which lean on the existence of this entity.

    The kind of scientific research, in which:

  • be aimed to look at already announced ideas, or information objects, or physical objects, from the different points of view;
  • be aimed to connect the given subject with different other relevent aspects;
  • be aim to collect a set of different aspects-links of the given subject with other aspects, subjects and points of view,
    is offered to name "postanalysis".

    The Term "postanalysis" emphasizes, that this is analysis of some idea after ("post-") its announcement, as opposed to other kind of the analysis, ("pre-") when an author aims suit to announce this idea.

    Postanalysis is the text, which on its essence, generally speaking, has not the end. Set of interesting connections of the one finite idea with an infinite set of the possible relevent points of view and other relevant subjects is the collection, which can be only expanded. (And occasionally - be narrowed, if some aspects will become to seem to the author non interesting.) For the reason given chapter 3 has not the logic end and can to be changed and to grow from the one version to the next.


    3.2. list "entities underlying for foundation" for the list "kinds of texts"

    So, we have a following logic chain:

    The texts (hypertext) happen be very different.
    It is possible to pick up something, that it is possible to call as "kinds of texts".
    These kinds are defined by the different answers on (1) question about what are the majority of the text and on (2) question about what the structure makes the logic basis of the text.

    This logic chain it is possible to try to continue further - to make out the list of "what can be the logic basis of a text".

    Let we shall make out such list.

    Format of the list:
    - kind of text: the structure of which of entity underlies for a given kind of text?

    - Textbook: a structure of the subject area
    - Reference book: a structure of the subject area
    - Scientific paper: some subject
    - Paper with annotations: the article as information object
    - Explanatory dictionary: set of concepts
    - Translator dictionary: set of words
    - Documentation: a structure of some object
    - User guide: an interface of the program as complex object, its structure
    - Administration guide: a structure of the work of a person on support, structure of the object, structure cause-and-effect links in functioning of object - Encyclopaedia: structures of the subject areas, set of the subject areas, structure of a system of the subject areas
    - System of tables: a structure of the subject area, set of the various tables (lists)
    - Data base of pictures: some subject, set of objects, which is possible to draw
    - Spatial maps: a structure of the space
    - Abstract model: a structure of some object
    - Catalogue: set of some objects
    - Chronology: set of events
    - Hypertexting knowledge base: a structure of the work of the person, structure of the subject area
    - Program: a structure of complex information object of a particular type - program


    3.3. list of entities "2Q-entities"

    It is possible to make following conclusions from the previous experiment:

    The large part of kinds of texts from the list "2Q-list" is devoted to:

    - subject area,
    - subject,
    - set of an objects,
    - a complex object,
    - a work of a man.

    Furthermore, we can find more rare, "marginal" entities:

    - cause-and-effect links,
    - space.

    Given list (let and further be broken on these two parts) plays very important role for the classification 2Q, thus we shall give the separate name for it: "2Q-entities".

    To the start of the file
    About the author
    About the IHPCS

    e-mail: Leonid Ovanesbekov <olg@www.ivvs.ru>

    1992 – 1998 Mar 09 mo 1992 - 1997 jan 05 su