have found it necessary to edit manually the entire tape -- centi- 

 nneter by centimeter. And mind you, these are data from only one 

 space "expedition" and one which had rather limited scientific 

 objectives. 



However, I don't want to give the impression that all things are 

 black as far as data centers are concerned; the "shoe is as frequent- 

 ly on the other foot" as not. I recall an incident of several years 

 ago when several individuals approached our organization for some 

 computing and tabulation assistance on a research project. Since 

 there were only three individuals assigned to the project, they want- 

 ed their work facilitated by having data from a large number of 

 weather stations collated in a particular nnanner and the results to 

 appear as a series of individual machine listings. We raised some 

 questions as to the magnitude of the proposed job but their reaction 

 was one of impatience more than anything else. They were not 

 interested in our excuses, they simply wanted to know "could we or 

 could we not undertake the project? " We allowed that we could do 

 the job but we felt we were obliged to point out that even if they 

 could read the tab sheets at the rate of one line per minute and they 

 would do this 24 hours per day, seven days per week, 365 days per 

 year, the three of them couldn't possibly live long enough to complete 

 the reading task, let alone analyze the contents or do anything else 

 with them. 



I have already had some experience with the impact that some 

 of the new and uncoordinated observational programs can have on an 

 unsuspecting data center. I was for a time associated with one data 

 center whose relative efficiency over the years appeared to decrease 

 almost in geometric proportion to the amount of data it had acquired. 

 This occurred in spite of the addition of the most modern analytical 

 techniques and computing equipment that served to increase total 

 production tremendously. 



There are many concrete examples that can be cited as illus- 

 trations of developments that have had staggering impact on data^ cen- 

 ters and research institutions. One example is the meteorological 

 satellite developed by NASA. A single Tiros- or Nimbus-type 

 satellite can acquire, in a few orbital passes, more atmospheric 

 data than can be analyzed by all of the U. S. meteorologists working 

 together as a team for a year. Then add to this, the constant supply 

 of data from some 100, 000 surface weather stations, some 3, 500 

 or nnore ships at sea, the pilot reports from covintless aircraft, the 



60 



