Databases Reference
In-Depth Information
over ears in my flies.”
8
The problem was not only the reproductive rate of fruit flies,
but also their propensity to mutate in response to environmental change—the precise
feature that made Drosophila so valuable to the geneticist interested in hereditary fea-
tures and mutations across generations. Mendel's peas had been docile and well behaved
by comparison: they were smooth or shriveled, and followed comparatively clear pat-
terns of generational inheritance—a far cry from the seemingly endless variety of eye
colors, wing shapes, and body sizes that emerged in the Drosophila “breeder reactor.”
In the face of this nineteenth-century data deluge, geneticists “had no choice but to
adopt a fundamentally new system of naming and classifying factors.”
9
In the lab, Dro-
sophila became a new creature, one that could not exist outside that institution. But, it
also reconfigured the lab itself, giving rise to new kinds of scientific places and persons,
including “a new variety of experimental biologist, with distinctive repertoires of work
and a distinctive culture of production”
10
In Kohler's striking language, experimental
biologists became “lords of the fly,” and the flies returned the favor.
Data Demand Care
Like Drosophila and Zea Mays, contemporary ecological data may be thought of as an
awkward and improbable species that has nevertheless found its perfect ecological niche.
Scientific data once fit on a few sheets of paper, which could last centuries if properly
stored; now, we have cultivated strains of data so densely compacted they need us to
take intricate care of them. As Cory Doctorow describes in a cover article for
Nature
,
we have created immense industrial data centers to store and process all this scientific
information.
11
In
Welcome to the Petacenter
, Doctorow stands in awe of the hundred-
million-dollar computing centers that have been established to store the tens of thou-
sands of terabytes (a terabyte being a thousand gigabytes) of data flowing from dozens
of meteorological satellites, hundreds of genomic sequencers, thousands of ecological
field sites, and the millions of sensors at the Large Hadron Collider. Just as the
Zea Mays
species of corn would die out in a couple seasons without our assistance, these comput-
ing centers would quickly overheat if not for the multistory cooling centers that control
the massive quantity of heat they produce. If the primary, secondary, and tertiary cooling
systems fail, it would only take ten minutes for the disk drives to bring their environ-
ment to a hazardous 42 °C (108 °F)—any hotter and they would begin to crack and
break.
These hives of industrialized data storage are potent symbols and key infrastructures
for the current era of “big” and “data-driven” science. But, the data center is also just
