Power-Programmierung/CD2.mdf/doc/mir/newsrel

Text File | 1992-06-29 | 7.9 KB | 157 lines

From:    Marpex Inc.
         5334 Yonge Street, #1102
         North York, Ontario
         Canada  M2N 6M2
         [phone]


             Advanced Retrieval Strategies Unveiled
               By Leading CD-ROM Software Designer

Toronto, July 1992... Free source code, and the secrets of indexing
and retrieval for massive databases are offered in a unique
tutorial package by a leading developer of retrieval engines.  The
Mass Indexing and Retrieval (MIR) Tutorials - an interactive
publishing series - have been developed by Douglas Lowry, former
chief scientist of Reteaco Inc.  He has produced some of the
world's most innovative and successful CD-ROM products.

In the seven years since challenging the industry with his trend-
setting FindIt retrieval software, Dr. Lowry has continued to
develop advanced techniques for handling large data indexing and
retrieval problems.  His first high speed, full-text retrieval
system for personal computers had to break many of the supposed
barriers in file size, indices, and ease of use of Boolean logic
and he continues to change the rules.  His current research is on
automating the identification of key words and concepts in any
text; with the addition of several trillion words to computer
databases every year, old methods of having a person read and
describe every piece of text is far too costly and inefficient.  In
the MIR Tutorials, Dr. Lowry shares his knowledge of these and
other advanced techniques.

The tutorials are available as shareware.  The series aims at "open
architecture" sharing and development of software technology.
Written for programmers and interested information industry
participants, the tutorials include free source code under the Free
Software Foundation GNU Public License.

Dr. Lowry is determined to change the ground rules in information
software development.  "I believe that society is better served by
an economy of sharing rather than one based on secrecy and greed.
We must recognize that the demand today is for a more open
approach," he explains.  "We must enable professionals to
concentrate on the needs of information users.  To do this we have
to change the rules.

"Too much of the development of indexing and retrieval systems is
being done behind closed doors.  If we develop a strong cadre of
people in the industry who reject that secrecy, we can have far
better software.  We can have people sharing ideas, accelerating
improvements.  The alternative is to have the information industry
controlled by a few billion-dollar corporations.  The MIR Tutorials
are simply going to level the playing field," he said.

The Mass Indexing and Retrieval Tutorials consist of five
interactive tutorial sections and a final wrap-up, complete with a
CD-ROM, that will be published at the end of the interactive
process.  Each tutorial includes learning materials on eight or
more topics.  The software tools and source code are available in
DOS and UNIX versions.

Participants are encouraged to communicate with Dr. Lowry as
tutorial instructor through electronic or normal mail and by
facsimile.  "This is a process of sharing.  One of the first things
I learned in teaching at the university level was that more is
learned in open discussions than from one-sided lectures. The final
result of this process of MIR Tutorials will reflect the interests
and insights of many of those who want to further the ability of
people to access and use information," he said.

The MIR Tutorials have been partially funded by the Canadian
government to promote ease of information distribution.  This is in
recognition of the growing awareness that effective distribution
of, and access to, information are keys to economic growth.  There
is a recognized need to break through data glut, and to turn raw
data into corporate intelligence.

Dr. Lowry earned his doctoral degree at the Massachusetts Institute
of Technology.  In 1985 he introduced a high-speed retrieval engine
for full text and statistical data that broke many of the supposed
limitations of personal computers.  One of his first CD-ROM series
was originally 4.5 gigabytes of data which he put on a single disc.
His highly successful visible indexing interface for CD-ROM is now
used widely in the industry.

Dr. Lowry has been instrumental in the design of leading CD-ROM
series for statistical research, chemical data, journals, books,
directories and multimedia programs.  Many continue to be among the
most sought-after CD-ROM products in the world and discs based on
his designs are sold in the thousands every month.  He also
collaborated in the original design for the award-winning CD-ROM
books from Discis Knowledge Research.

A 65 page introduction to the project was released at the end of
May.  An updated version has been placed on major electronic
bulletin boards (CompuServe, Canada Remote) at the end of June.

MIR Tutorial ONE has been released as shareware ($29 registration)
as of July 1, 1992.  It deals with the problems of database
analysis.  The tutorial provides automated analysis programs and a
discussion of the many problems facing the information provider and
some of the possible solutions.  25 data analysis programs,
complete with source code, were released with Tutorial ONE; the
programs may be used without royalty or payment.  The source code
may be copied, distributed, tinkered with, improved... almost
anything except copyrighted or patented.

The remaining four tutorials follow the same guidelines... low cost
shareware registration for the tutorial text and "copyleft"
programs.  Each successive tutorial will be published when the
there have been 1,000 shareware registrations for the preceding
tutorial.  For example, Tutorial TWO will be released after the
number of registrations for Tutorial ONE passes 1,000.

MIR Tutorial TWO gets the participant involved in the secrets of
how to prepare data for indexing and retrieval.  The various
software programs provide highly useful tools for not only cleaning
up some of the common database problems but greatly accelerating
the process.

MIR Tutorial THREE offers the programs and background on automated
indexing.  Programs such as these used to take many hours to
process on mainframe computers and can today be performed as
routine on high-powered personal computers.

MIR Tutorial FOUR shows in detail how to get split second response
when searching for combinations of words and phrases across massive
sets of data.  The software includes a powerful retrieval engine.
This tutorial also delves into the concerns and opportunities for
standardizing user interfaces.

Tutorial FIVE examines a wide range of related applications...
records management, archiving, encryption, correlation tools, etc.
It projects the future in which high powered search is a basic tool
at all levels of society.

The five interactive tutorials will be followed by a compilation of
all the previous releases of text and software, together with new
material jointly developed by Dr. Lowry and participants during the
process.  The accompanying CD-ROM will include software that has
been adapted to work on other computers, further worked examples,
and a compendium of insights and program enhancements from
participants.

Each of the five MIR Tutorials may be registered individually at
$29 per release.  For those who access the files from electronic
bulletin boards, there is no further cost.  Software diskettes,
UNIX system tapes, and paper copies are available by mail order at
moderate prices.

Orders or registrations for the MIR tutorials can be made through
electronic mail via Compuserve 71431,1337, Internet
doug.lowry%canrem.com, or UUCP canrem!doug.lowry.  The FAX number
is 416 963-5677.  Orders may also be placed through Marpex Inc.,
5334 Yonge Street, #1102, North York, Ontario, Canada, M2N 6M2.

                              -30-