Power-Programmierung/CD2.mdf/doc/mir/02who

Text File | 1992-06-29 | 10KB | 213 lines

      ══════════════════════════════════════════════════

            2.  WHO SHOULD BE USING MIR TUTORIALS?

      ══════════════════════════════════════════════════


        ═══════════════════════════════════════
2.1           Table of users and benefits
        ═══════════════════════════════════════

            FOR WHOM?                        BENEFITS

        Directors of            »   Outstanding quality
        Information Services    »   Speeds up standards process
                                »   User learns a single interface
                                    for all data searches
                                »   Open architecture

        Custom software         »   Cuts customization time, costs
        providers                   through engine-independent
                                    techniques

        Information publishers  »   Reduces costs through applying
                                    own expertise to data

        Government information  »   Massive quantities of data can
        distributors                be searched in record time

        Educators and trainers  »   World class, "copyleft"
                                »   R & D contributions welcomed

        Programmers             »   Pride of participation
                                »   Expanded capabilities
                                »   Builds reputation through
                                    sharing improvements


        ═════════════════════════════════════════════
2.2           Directors of Information Services
        ═════════════════════════════════════════════

            In large organizations, MIR Tutorials and software are
directed in the first instance to the person with responsibility
for corporate direction of information management.  The job title
may be Director of Information Services, or Vice-President in
charge of Information Services, or an equivalent position.  The
reason for aiming toward the top of the organization is that high
speed search technology has management implications:

        »   Progressive management puts priority on service and
            people-oriented skills.  When business empowers its
            people to source information for themselves, individual
            creativity and job satisfaction increase, and there is
            a reduction in supervisory levels and costs.

        »   An increasing portion of staff are information workers.
            Time for these people to learn and relearn a wide range
            of search and retrieval methods are sapping time from
            the needs of the organization.


            All software embodies beliefs about people.  A tell-
tale sign is found in the error messages embedded within a program.
We are all familiar with techno-centered software which is written
by technical people to please other technical people, and the end
user is treated as a naive inconvenience (or a necessary evil).  If
there are to be truly user-driven or market-driven information
services, there must be a concerted push from the top to make it
happen.  Implementation of new retrieval and pattern-search systems
will be delegated to lower echelons.  But the user focus must have
support from the top.

            The table at the beginning of this topic lists a
variety of benefits to the large firm.  Performance quality is
important because the large organization is more likely to handle
massive databases, and is more prone to system degradation through
inefficient software.  Standards are relevant because too much of
data processing budgets are consumed by incompatibilities stemming
from lack of standards.  MIR technology ties in with the current
drive to make it possible for users to learn a single interface for
all data searches; the benefits in convenience and reduced
re-education will be substantial.  Closed architecture hardware and
proprietary software elevate costs; here we provide source code and
techniques that may be freely used and shared.  This kind of open
architecture approach reduces your dependence on suppliers.


        ═════════════════════════════════════
2.3           Custom software providers
        ═════════════════════════════════════

            Organizations that provide "custom" computer programs
and support to others make their profit through selling variations
of the same software as many times as possible.  Their competitive
advantage stems from high quality support and timely delivery more
than from the software itself.  Further on we will look at engine-
independent techniques.  They help by reducing substantially the
time and cost required to customize indexing and retrieval
software.  More resources can therefore be directed toward user
education and support.

            The firm selling off-the-shelf software is not as
likely to benefit from MIR software.  This is because of the
"copyleft" requirement that programs embodying MIR techniques must
be accompanied by source code without restriction on copying.
Proprietary secrecy is a common means of preserving a competitive
advantage.  We believe that in the long run the market is better
served by technical excellence coupled with access to source code
and high quality support services.  The software firm that shares
these commitments need not fear disclosure of source code, and can
benefit from the technology and method offered here.


        ══════════════════════════════════
2.4           Information publishers
        ══════════════════════════════════

            Here we address the needs of organizations that have
large quantities of data that they wish to make accessible to many
people.

            The challenge of in-house communication sometimes
overwhelms people in a two-person organization (marriage, for
example).  In the context of 2,000 or 20,000 employees, it's even
tougher.  Examples of these organizations include:

        »   a public accounting firm that produces audit standards
            and procedures for internal use;

        »   a bank using 30,000 pages of manuals that must keep
            several hundred branches up to date;

        »   a regional real estate board servicing thousands of
            agents with detail on current listings.

The more rapidly the data changes and the more geographically
dispersed the people, the more difficult the task.

            CD-ROM publishers have multiplied in recent years.  For
example, an aircraft manufacturer may wish to supply maintenance
manuals to every buyer and to every airport at which the particular
aircraft type may touch down.  CD-ROM publishers have diverse
interests ranging from financial trends or hazardous commodities to
bible study materials and telephone directories.

            The common factor in in-house communication and
external publishing on CD-ROM is expertise concerning, and access
to, large databases.  Each stands to gain by being equipped to
apply its expertise to its own data.  The publisher can format the
data; it often has particular knowledge of how people have made use
of the data in the past.  Even if a service bureau or consulting
firm is brought into the picture, the organization can reduce its
costs.  Some service bureaus (such as Innotech, a partner in the
MIR project) offer substantial discounts to customers that process
data into a standard form and/or set up uniform field indexing
files prior to sending the data to the service bureau for CD-ROM
publishing.


        ═══════════════════════════════════════════════
2.5           Government information distributors
        ═══════════════════════════════════════════════

            Government records can grow to massive proportions.
Consider the millions of tax returns filed each year, the task of
tracking immigration, military procurement programs dealing with
millions of different parts, medical care of veterans, and so
forth.  Automated indexing and high speed retrieval is essential,
whether the information is for government use or for distribution
to the public.  There is considerable software on the market that
can handle reasonably well databases with perhaps a hundred
thousand records.  But many government databases have millions of
records; for them, a better class of software is essential.

            Government also stands to gain if search and retrieval
methods move in the direction of shared standards.  Discussion of
standards receives strong impetus from open architecture software.


        ══════════════════════════════════
2.6           Educators and trainers
        ══════════════════════════════════

            Material in the public domain on indexing and retrieval
tends to be either obscure or effective only for limited size data
sets.  In the MIR project, we have set aside the proprietary
approach, to the direct benefit of teachers and students.  Courses
in management information systems and in computer science can use
the techniques and software here... discuss them, improve them,
extend their applications.  The interactive publishing technique is
a variation on the exchange of ideas through academic journals.
Your research and development inputs are most welcome.


        ═══════════════════════
2.7           Programmers
        ═══════════════════════

            You may be any, all, or none of the above.  But you
love computer programming... the ultimate in problem solving where
you can see the results of your creativity.  You didn't invent what
is here, but once you see the software, you may see ways in which
you can improve on it.  The best software emerges, not from tidy
development groups, but from public "skunkworks" in which people go
at the task for the sheer love of it.  The EMACS editor is a lovely
example.  There is pride in having taken something good and making
it better.  And reputations are built through hacking at a popular
project and getting one's name associated with innovative
improvements.  This project is built around "free" software.  So,
go at it.  And have fun!