ment sectors and data exchange also has been 
established with certain foreign countries. Concern 
over proper data processing, archiving, and re- 
trieval also should be applied to its collection— 
both to instrumentation and methods employed to 
gather data. 
Most ocean programs have been limited in both 
staff and budget. As a result, specific program 
objectives are often compromised and only limited 
instrumentation is procured. Unlike conditions in 
many other non-oceanographic programs, such as 
the space program, ocean instrument specifications 
are often minimized, meaningful quality assurance 
programs are largely nonexistent, and service and 
maintenance manuals and other documentation 
are often inadequate to meet basic user needs. In 
addition, statistical information defining condi- 
tions of use, maintenance and repair cycles, and 
modes of failure are seldom documented and made 
available to the manufacturer. This, in turn, slows 
down the correction of problem areas and prevents 
the upgrading of performance and reliability in a 
logical manner. 
Past experience shows that user demand for a 
particular type of ocean instrument is generally for 
a limited quantity of highly complex instruments, 
often requiring custom design. In such cases, 
manufacturing does not lend itself to mass produc- 
tion, one factor that has allowed the small, 
technically oriented firm to compete favorably 
with large corporations. Although large capital 
facilities are not always essential to produce 
marine instruments, expensive facilities are often 
necessary for development and qualification test- 
ing. 
B. Specific Problem Areas 
The most valid complaints about oceanographic 
instruments are their lack of reliability and lack of 
user confidence in the data gathered. Many articles 
have been written and symposia sponsored to 
examine the diverse sources of unreliability. One 
recent meeting identified two primary factors:** 
33 Government-Industry-University Symposium on In- 
strument Reliability, May 6-7, 1968, Miami, Florida, 
sponsored by the National Security Industrial 
Association-Ocean Science and Technology Committee 
(OSTAC) Ocean Platform and Instrumentation Subcom- 
mittee. 
—Lack of a common instrument performance “lan- 
guage” and satisfactory communciations between 
instrument producers, procurement agencies, data 
collectors, and data users. 
—Present ocean instrument procurement policies. 
1. Language and Communications—Need for 
Specification Guidelines 
Producers, procurement agencies, and users 
require standards and specification guidelines en- 
compassing the following: 
(1) Performance requirements, (2) environ- 
mental conditions, (3) test procedures, (4) quality 
assurance requirements, (5) design requirements, 
(6) interfacing and/or installation requirements, 
(7) terminology, (8) formats for specification and 
data, and (9) documentation. 
Performance requirements should indicate 
types of functions an instrument must perform 
and how well these functions must be carried out. 
Thus, tests will have to measure such items as 
repeatability, stability, data rate, accuracy, and 
precision. 
Additional specifications should be related to 
an instrument’s interaction with various environ- 
mental conditions during operation, storage, and 
shipping. Therefore, testing would have to ascer- 
tain the instrument’s ability to withstand such 
environmental aspects as temperature, shock and 
vibration, pressure, noise, salt spray, and humidity. 
Case histories have been compiled showing the 
seriousness of time lost through equipment failures 
from such sources as corrosion and shock and 
vibration. Developing laboratory tests to simulate 
environmental conditions is difficult, quite com- 
plex, and expensive. 
For a test procedure specification, for example, 
it is necessary to define precisely what constitutes 
an acceptance test to determine if each perform- 
ance and environmental specification is met satis- 
factorily by the manufacturer. 
Furthermore, oceanographic instruments and 
important components should be classified by 
type, and standard specifications should be devel- 
oped for each classification. 
The foregoing indicates types of specifications 
that could be standardized by Federal agencies 
V-47 
