How does hypertext compare with relational databases?
                   =====================================================

Both hypertext and relational databases index information.  However, hypertext
makes far different assumptions about the capability of the users and the
purposes for indexing information.  <FILE17 DESIGN GOALS>

In looking at the expectations of users, relational databases make the
following assumptions:

    Users must know the language (or they can't access specific information).
    Users must know the synonyms (or they can't access all the information).
    Users must know set intersection (or they can't access the information).

In return, the relational databases provide the following:

    Information that matches the search criteria (ignores close fits)
    Information that matches the search criteria (words rather than ideas)
    Information that matches the search criteria (but has no sense of structure)
    Information that matches the search criteria (but has no browsing capability)

Given this, you might ask why are relational databases used?  <FILE69 DATABASES>

The answers to this question may center on:

    -  Not knowing of alternate ways to index information
    -  Saving initial expense while increasing access costs to each user
    -  Assuming data-field approaches work with free-field text and ideas
    -  Not understanding the needs and talents of the users of such systems
    -  Not understanding the structural defects of relational databases
    -  Not having the time and talents for building more useful index systems

In contrast, here's a quick review of the goals in designing hypertext
knowledgebases:
     ÚÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ¿
     ³   - index information by idea content                     ³
     ³                                                           ³
     ³   - use minimum-keystroke preconstructed paths to         ³
     ³     information                                           ³
     ³                                                           ³
     ³   - allow users to rapidly browse information (find what  ³
     ³     they didn't know they were looking for) <FILE56>      ³
     ³                                                           ³
     ³   - transmit knowledge by expanding understanding of the  ³
     ³     structure of the information within the system        ³
     ³     <FILE55 KNOWLEDGE>                                    ³
     ÀÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÙ

These are the key advantages of hypertext (and the standards by which to
measure various hypertext systems).

At the core, hypertext systems are philosophical alternatives to relational
databases <FILE69 DATABASES>.  That is because hypertext centers on two ideas
-- indexing information by idea content, and rapid access to such information
regardless of the user's level of understanding.

Relation approaches to information by indexing words simply does not index
ideas.  As a result, hypertext systems that index ideas are magnitudes
faster and more useful than database approaches to information that index
only words.

As you know, relational databases presume both a knowledge of the language
of the field and usage of set intersection techniques with the language. If
you are unfamiliar with either, you simply can't extract information from
the system.  That's the reason why hypertext exists -- it overcomes many
deficiencies in relational databases.

For that reason, I think hypertext systems that depend on database
methodologies (as most do <FILE69 DATABASES>) have missed the main advantages
of hypertext, which is putting information in formats already matching the
needs of users.

Neil Larson  1/16/88  FILE23
44 Rincon Rd., Kensington, CA  94707
Copyright MaxThink 1988 -- Call 415-428-0104 for permission to reprint