perm filename LIBRAR[E82,JMC] blob
sn#676184 filedate 1982-09-02 generic text, type C, neo UTF8
COMMENT ā VALID 00002 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00002 00002 For a number of years, the technology has been available
C00006 ENDMK
Cā;
For a number of years, the technology has been available
for a computer-based library of all the world's literature. Such
a library can make available in everyone's home and place of work
everything that has been written - technical, cultural or commercial.
The costs of disk file have long made this feasible when we
take into account the fact that one computer based copy can take the place
of many copies on shelves. More recently costs have come down to
where disk storage is competitive with a single copy on a library
shelf.
The main problems are:
1. organizational. Hundreds of commercial data bases have developed,
but none have the ambition of providing at reasonable cost access to
all the world's literature.
2. time-sharing. Existing large computers, e.g. the IBM 3081, do not
provide efficient access to databases for the number of users their
speed should support. This limits the performance of some quite modest
databases. We are not aware of any real effort to study what features of
operating systems will permit really large numbers of users.
3. character reading. While there are many character reading machines,
their performance is effective only for special tasks - reading
special fonts or reading to the blind who can force a repeat when the output
is confusing. As the matter stands, there is no company that will take
fixed price contracts to read books. Most keyboarding of bulk material
is done in the low wage countries of the Far East.
We propose:
1. To build an experimental library and make it generally available
to anyone with a terminal. The operating system will be optimized
for very large numbers of users.
2. To develop, using mainly existing technology, an OCR system cost-effective
for reading books - at first non-technical books consisting of text
only. It will be capable of handling essentially all the fonts
prevalent in libraries.
3. To develop a plan for public computer available library. We don't
expect this to happen by itself as a development of the commercial and
Government supported databases, and public libraries developed in the
first place with foundation support and later with Government support.
The role of Andrew Carnegie in supporting the development of public
libraries may need repetition in the current era.