Hunting & Gathering

Three Ways to Optimize

In order to optimize, an actor’s goal is to alter the total search cost by gaining many calories in turn for processing an item. In a presentation for the Institute of Human Origins, Steve Churchill explains three means of optimization, or lowering search cost for high-ticket items: wild harvesting, creating information systems, and planning depth.

Wild harvesting involves cooperative hunting methods in order to turn a field of large animals into meals for several months. He claims Neanderthal people (early Homo sapiens mono-strategic hunter-gatherers) would have seen the field-full and thought Dinner!—not five months of secured meals. Creating information systems involves some planning. If the size and location of the herd can be narrowed down, then the right labor group can be prepared and sent to hunt. Other people are integral to working information systems. Planning depth of the information network optimizes foraging further. The more that is known before the hunt, the better prepared the hunters will be, and the fewer calories will be expended. Cro-Magnon reduced search cost by planning, and they made low-cost bulk caloric inputs possible.

Optimal foraging is not functional only for explaining survival success and extinction in early hominids. If one equates calories to information (Campbell, 1982) (which is not a far leap if one recalls the First Law of Thermodynamics where energy can be neither created nor destroyed, just converted), there are interesting implications for modern hunter-gatherers and the ways we search, or our attempt to gain the most information with the lowest search cost.

Foraging Tactics

Tactics move a search forward. In Information Search Tactics (1979b), Marcia Bates articulates six foraging tactics, as seen in the following chart (figure 6.2). Bates describes how these (and other) tactics can be applied both to bibliographic searches in which one seeks bibliographic citations to documents that will contain information one desires and to reference searches in which one searches to find information to answer a specific question.

One can observe that these same tactics are, often unconsciously, applied to foraging in common, everyday situations. We use the same tactics, for example, while foraging in the grocery store, as posed in figure 6.3.

In addition to the tactics Bates described, we observed another tactical strategy. Occasionally, one encounters information that, though it does not answer any pressing questions, seems valuable. One tucks it away because it may come in handy later. This may be what the scholar does when she spots an intriguing book in the new acquisitions shelf at the library. It could help one formulate what may turn out to be the answer one seeks. This save-it-for-later tactic assumes planning ahead, or at least knowing that one may require some planning at another time. Seeds, an article, or Y2K supply of instant rice: the effect is the same. This save-it-for later tactic assumes that one may find a piece of information useful, though not immediately pertinent, and saves it for later, anticipating that this tidbit might move a yet unknown search forward. Hunter-gatherers, by nature, store information for use, understanding that there may be a time when information is scarce.

specify: to search on terms that are as specific as the information
exhaust: to include most or all elements of the query in the initial search formulation; to add one or more of the query elements to an already prepared search formulation
reduce: to minimize the number of elements of the query in the initial search formulation; to subtract one or more of the query elements from an already prepared search formulation
parallel: to make the search formulation broad by including synonyms or otherwise conceptually parallel terms
pinpoint: to make the search formulation precise by minimizing the number of parallel terms, retaining the more perfectly descriptive terms
block: to reject, in the search formulation, items containing certain terms, even if it means losing some sections of relevance
Adapted from Bates (1979)



Figure 6.2. Foraging Tactics

Foraging Strategies

Tactics move the search forward, and strategies are the search methods employed to execute the hunt. One can observe foraging strategies in action in everyday situations. Several foraging strategies are defined in the following descriptions. Any of the strategies can be used to forage for whatever relevance is being sought. The strategy employed depends on one’s depth within the search process and the specificity of information that is required.

specify: I want Land-o-Lakes, Morning Blend margarine.
exhaust: I want cottage cheese, large curds, low fat.
reduce: Since there is no low-fat, large-curd cottage cheese, I’ll take any large-curd cottage cheese.
parallel: I need acetaminophen; Tylenol or generic brands will do.
pinpoint: I need Tylenol.
block: I want potato chips, not BBQ, salt and vinegar, or rippled.



Figure 6.3. Foraging Tactics Applied to Grocery Shopping

Hunting and Picking
Hunting is predicting the likeliest location to find a useful item (P. Wilson, 1968). Imprecision in representations of document sets still plagues information seekers of disabling item specificity. The hunter has narrowed the search by predicting the likeliest location. If representations were specific, single bibliographic records would be at least as lengthy as the document itself. Each word, idea, and dog-ear would be necessary to provide definite identification of each item, rendering the catalog redundant and useless. An accompanying strategy is necessary to identify relevant documents. Picking is selecting items based on a description when the description is imprecise (Wilson, 1968), narrowing the possible solutions set further. Presumably, the purpose of bibliographic tools (indexes, abstracts, bibliographies) is to minimize the amount of hunting and picking a forager needs in order to find relevant information. In the library, professionals have tried to create systems of representations to reduce hunting and picking. Essentially, these professionals have attempted to reduce search cost for patrons. Hunting on the World Wide Web may often prove to be a futile enterprise, as likely locations are frequently impossible to predict. The search engine Google.com provides an I Feel Lucky search option that, allowing one to feel like the possibilities are not infinite, takes the searcher to what the algorithm decides is the best fit for the search. In typical Web searches, one must pick through hits to discover what one wishes to have hunted in the beginning.


Coupling
Coupling is an extension of Wilson’s ideas on hunting and picking (1968). It involves generating links between what one hunted and picked. Coupling supposes that the forager has focused the likeliest location of the resources (hunted) and has gathered the seemingly relevant documents (picked), but has not generated any links between the question state and what has been hunted and picked. We consider this process a foraging strategy in how one might use coupling to search tidbits that have been stored in a save-it-for-later tactic for what may fill a knowledge gap. In this way, it is possible for one to couple one’s own thoughts. One might assimilate and accommodate new knowledge, or draw connections from current knowledge stores that have not yet been linked.

Indexing
Indexing points to what has a reasonable likelihood of being relevant (O’Connor, 1996). Bibliographies point to resources that may also pertain to the topic of the bibliography, for example, and maps point to where one might need to be. A forager uses an index, of whatever sorts, to direct the search or to point the search in the direction of what could be the relevant target. Back-of-the-book indexes reduce search time and cost by directing the search in what could be the right direction. An index, no matter the intent of the indexer, can be useful only insofar as it points to what the forager needs (Wilson, 1968). The optimal index, then, is only situationally relevant to individual foragers.

Browsing
Browsing is hunting for sustenance in an area where finding and evaluating resources are issues. It is foraging for an unknown among uncertainties (O’Connor, 1993); it tends to have a high cost-to-gain ratio. If the library was filled with volumes of unorganized documents, then this could be a browsable collection. Just as an animal might browse to satisfy a need for nourishment without knowing what shape the anticipated nourishment might take, browsing can be a response to subject indeterminacy (O’Connor, 1996) when a basic unarticulated need for information exists. Even in an organized collection, if the collection is not organized to the specific requirements of a particular user, browsing may be a necessary hunting strategy. The browser anticipates, or is open to, discovering an answer somewhere in the wilderness through which he forages.