By Charles Henry
Part 1 of a 3-part series
A recent front-page article in the New York Times grabbed my attention. Titled “Brainlike Computers, Learning from Experience,” the piece described a new kind of computer processor: one that is designed to function as the human brain’s neuronal network functions. Large corporations such as Google and academic computer science programs at universities are working on these “neuromorphic processors,” including Stanford (Brains in Silicon program) and MIT, which was recently awarded in partnership with Harvard and Cornell a major NSF grant to inaugurate the new Center for Brains, Minds, and Machines. Larry Smarr, a well known astrophysicist, declares that this represents a major shift in computer design away “from engineering computing systems to something that has many of the characteristics of biological computing” with the processors’ wiring mimicking brain synapses.
The implications are astonishing. Rather than relying on typically voluminous lines of code and programming effort, these processors respond to data based on the accumulation of past experience. The Times article notes, “The connections between the circuits are ‘weighted’ according to correlations in data that the processor has already ‘learned.’ Those weights are then altered as data flows in to the chip, causing them to change their values and to ‘spike.’” The neural network is thus reset according to the new weighted values, and subsequent computation is adjusted to accommodate this experience.
These neuromorphic processors are a logical iteration in the development of artificial intelligence, a discipline in computer science that can be traced to the period following World War II. While it is well beyond the purview of this blog to trace that complex history and its often impassioned proponents and challengers [1], it is worthwhile to revisit, as often is the case for reflective essays, Vannevar Bush’s seminal 1945 article “As We May Think”to establish a context for our present academic digital environment in which biologically designed machines may soon appear. Among many prescient observations, Bush noted the vast disconnect between contemporary methods of knowledge organization and the supple associative power of the human mind.
On then-current library practices:
“The real heart of the matter of selection, however, goes deeper than a lag in the adoption of mechanisms by libraries, or a lack of development of devices for their use. Our ineptitude in getting at the record is largely caused by the artificiality of systems of indexing. When data of any sort are placed in storage, they are filed alphabetically or numerically, and information is found (when it is) by tracing it down from subclass to subclass. It can be in only one place, unless duplicates are used; one has to have rules as to which path will locate it, and the rules are cumbersome. Having found one item, moreover, one has to emerge from the system and re-enter on a new path.
On our cognitive capacity:
“The human mind does not work that way. It operates by association. With one item in its grasp, it snaps instantly to the next that is suggested by the association of thoughts, in accordance with some intricate web of trails carried by the cells of the brain. It has other characteristics, of course; trails that are not frequently followed are prone to fade, items are not fully permanent, memory is transitory. Yet the speed of action, the intricacy of trails, the detail of mental pictures, is awe-inspiring beyond all else in nature.” Bush proposes a machine called a “Memex” that more closely approximates the workings of the human mind.
In these paragraphs Bush cogently identifies the correlation between what we would call the methods of knowledge organization—the keys by which we access content—and the determining aspect of those organizational methods vis a vis the success of our “seeking,” or search. He describes a welter of silos, and the significant work it takes to find one item or a few items pertinent to our interests.
In 2014, we persist in cataloging information via indexes, key words, or other “linear” methods of accessibility. Our databases, repositories, and digital libraries are awash in metadata schema that do not easily communicate; information dwells in the umbra of disconnected silos, further compounded by ownership of large quantities of information that are rigidly fenced and require proprietary methods of access. In a recent posting for its 2014 conference, Metadata Intersections: Bridging the Archipelago of Cultural Memory, the DCMI described the contemporary information landscape for galleries, libraries, archives, and museums: “traditions in documentation and organization lead to significant differences in both their languages of description and domain practices. And yet, the push is on for ‘radically open cultural heritage data’* that bridges these differences as well as those across the humanities and the natural sciences. DC-2014 will explore the role of metadata in spanning the archipelago of siloed cultural memory in an emerging context of linked access to data repositories as well as repositories of cultural artifacts.”
So today the frustrations encountered by Vannevar Bush and his “seeking” over 60 years ago are sadly similar, even though we have migrated from an analog world to a digital one—or effectively have we? The old method, put another way, has captured the new medium in its traditional grip, rather unimaginatively. But on the horizon a new type of computer processor may emerge with the prowess of Bush’s memex device, aligning in function and aspect much more closely with our mental acuity.
What are the implications? For brevity’s sake, take as an example the ubiquitous MARC records. These are metadata conforming to machine readable cataloging standards developed in the 1960s. Readable, however, by what kind of machine? In light of potentially transformational biomorphic computing, what should we, as practitioners working in and relying upon the complex world of academic information, be thinking?
* John Voss, “Radically Open Cultural Heritage Data on the Web”
[1] See, for example, “Noam Chomsky onWhere Artificial Intelligence Went Wrong,”The Atlantic, November 1, 2012.
