Accuracy of Internet academic (fwd)

[phil cash cash]

Accuracy of Internet academic
'Like trying to pull a library book without the Dewey Decimal system'
http://www.canada.com/national/nationalpost/news/artslife/story.html? 
id=35deb42a-93e2-4cbc-befe-59b5f8ce483

Siri Agrell

National Post

Saturday, February 14, 2004

Laura Buszard-Welcher thought she had found the perfect source to  
support her PhD research on endangered languages. The anthropologist  
was completing her doctorate at the University of California, Berkeley,  
when she discovered the Web site of someone who spoke the obscure and  
threatened dialect she was investigating.

She referred to the site in her research, which was later published as  
a book. Months later, when she revisited the site, she discovered her  
research now cited something quite different.

"It took me to some porno site," says Buszard-Welcher, now a professor  
at Wayne State University in Michigan and one of a growing population  
of anthropologists who have logged on to the Internet as a way to study  
human society.

Cyberanthropology is a growing area of study. Through the Web,  
anthropologists are discovering societies that reveal themselves online  
while remaining submerged in the world at large.

But, as Buszard-Welcher and others have found, cyberanthropology  
requires the navigation of many pitfalls -- issues of accuracy,  
credibility and professional esteem -- even as it unearths communities  
at the click of a button.

"I'm finding a lot more language information than I would otherwise  
have had access to," Buszard-Welcher says. "But you have to be  
critical. Just like picking up a book off the library shelf, you have  
to decide how important it's going to be."

As an undergraduate, Buszard-Welcher knew she wanted to explore  
endangered languages and the tools used to protect them.

At first, she studied a native dialect (a form of Ojibwa) which was  
local to the area of Michigan where she grew up. But she soon began to  
look farther afield and her curiosity led her online.

When her computer started speaking Potawatomi, a language spoken by  
only about 50 people, she had found her thesis.

"People have created Web sites to promote this language, to learn this  
language again, bring it back to widespread use," she says. "People  
from these far-flung communities are becoming integrated in ways that  
they couldn't have before because of physical distances."

Online language preservation is now her specialty. When not studying  
Potawatomi, she is collecting URLs for other language projects,  
monitoring the different techniques people are using to protect their  
traditional tongues.

But as any Web browser knows, surfing the Internet often becomes a  
struggle to stay afloat in a sea of spam. Looking for an academically  
useful online source is like finding a library book without the Dewey  
Decimal system. But anthropologists are patient.

"There are all these resources out there that people haven't learned to  
share," Buszard-Welcher says. "I think if people start making their  
information available on the Web, it's going to have a huge impact on  
our field."

She fantasizes about the world of information that will be at her  
fingertips as soon as someone thinks to upload it. Right now, someone  
could be building the site she has been looking for all her scholarly  
life.

When she does unearth a useful site, she is confronted with issues of  
credibility. "It's a problem of figuring out whether a site is  
authoritative," she says. "That's a problem in academia in general."

Though most universities have their own standards for citing the  
Internet, following The Chicago Manual of Style, there is still  
resistance to its use as a primary or even secondary source. The  
Internet is frowned upon because of the very traits that make it  
effective for the public: an utter lack of controls.

Academic journals are rigorously researched, cited and peer-reviewed, a  
level of oversight that is currently impossible for  
cyberanthropologists.

"If you do it anyway, it's going to impact you negatively in terms of  
tenure decisions and that kind of thing," Buszard-Welcher says of  
citing online material. "I would be very happy if people in my  
profession would change their attitude and accept it as more valuable  
than it's particularly seen to be."

David Zeitlyn, an anthropologist at the University of Kent in  
Canterbury, England, says there are valid reasons to distrust Internet  
research. "There's a lot of what I perceive as bad anthropology that  
uses the Internet," he says. "People say they're doing the ethnography  
of the Internet by sitting at their desk and playing around in [chat  
rooms].

"For me, the whole point of anthropology is that you're not looking at  
one little part of society, you're looking at how all the parts  
interconnect. So, you might start or end up with how people are using  
the Internet. But as an anthropologist, I want to look at a whole load  
of other things."

The Internet reveals little about human behaviour, he says, because it  
does not explain why people are doing what they are doing.

Resources such as the Google Zeitgeist -- a yearly compilation of  
search trends -- should be used as an index, not an encyclopedia, he  
says.

"It's a starting point. It raises questions," Zeitlyn says of the  
feature. "But taking those patterns by themselves doesn't do enough.  
Questions can only be answered by doing weird things like talking to  
people. And I don't mean sending them an e-mail."

Zeitlyn is a leading voice on the Internet's impact on society. It is  
an interesting jump from his anthropological area of expertise: the  
methods of divination in rural Cameroon -- how tribes interpret the  
signs and omens around them.

He sees the Web as a harbinger of things to come, but does not believe  
the patterns of its use can be read like chicken bones to divine the  
past or the future.

For him, the Internet is a way "to do classic anthropology better" --  
to access more research, monitor more communities and keep in constant  
touch with the world.

Already, he sees ways to make the Internet a more authoritative  
research tool. Through Google's cached pages -- photographs of every  
page created -- such researchers as Buszard-Welcher can find expired  
sources without being redirected to HOT TEEN GIRLS.com.

"There are ways of turning the clock back," he says. "There's loads of  
unreliable stuff printed on paper, too. The critical thing is being  
critical."

Zeitlyn doubts that the Internet can be sharpened into the primary tool  
for anthropologists, but his research partner and co-author, Michael  
Fischer, who also spoke to me from the University of Kent, says,  
"Things could change over the next decade. There could be pervasive  
network connections that will make our broadband seem like teletype.  
It'll look like we're just standing still right now."

He believes the Internet is already "pretty authoritative" as a source  
and is limited only by perception, not content.

"Just like in real life, some people misrepresent themselves," he says  
of online content. "But you have to know an awful lot to misrepresent  
yourself successfully.

"I suspect [most people are] being more themselves," he says of the  
online individual. "Part of what society and culture do is force you to  
present yourself in a way that is appropriate to the parties that are  
observing you, not to do what you want to do."

But relevant human voices are being drowned out by corporate or private  
interests on the Internet, he says. Such search engines as Google are  
quickly becoming redundant as people learn to manipulate the system.

Type the words "Miserable Failure" into a Google search and the first  
link delivered is the White House biography of President George W.  
Bush.

The Google algorithm rates sites by the number of other pages that link  
to them. The more sites that connect the phrase "miserable failure" to  
the White House, the higher its ranking becomes.

A search of "Santorum," the last name of U.S. Senator Rick Santorum,  
brings up his own Web site but also references to the sexual by-product  
of a homosexual act, named for the Senator by a syndicated columnist  
outraged by his anti-gay remarks.

This sort of "Google Bombing" is so pervasive that it is impossible to  
know if search results are clean.

"Google is only adaptive in terms of popularity," Fischer says. "And  
popularity doesn't always result in the highest quality."

While Google is busy developing commercial and pop-culture  
applications, the less advanced sites, such as Hot Bot and Lycos, are  
more useful for authoritative searches, he says. "The net was actually  
at its most useful for academics between '95 and '98, because the rest  
of you weren't on it."

In the early days, the Internet was the domain of dorks and doyennes  
only.

"At that time, most of it was actually people trying to use this as an  
alternative medium to get stuff out there," he says. "Now it's gotten  
much harder to find information. Google ... but it doesn't promote  
depth. On the other hand, most people don't want depth; they want to  
buy a TV and it's really good for that."

For academics, the Great Byte Hope is called the "Grid."

The U.K. government is funding Zeitlyn and Fischer to try out a new  
subscription-based information-sharing system.

The Grid is for research organizations, accessible by invitation only.

"It's the so-called eScience grid," Fischer says. "It's actually built  
around the notion of people having to have credentials to know things."

The Grid is privately controlled, expensive to get into and extremely  
interesting to academics, private research groups and governments.

Unlike conventional networks that accommodate communication between  
machines, grid computing harnesses the unused processing capacity of  
all computers in a network to solve problems too intensive for any one  
machine. It feeds on a network of communal resources, sucking its power  
like a matrix. "They need all the power to analyze all the material  
that's out there in the public," Fischer says.

With Zeitlyn, Fischer has long used the Internet to track human  
behaviour. Now he is considering writing a book about ways to avoid  
online monitoring, or "data mining" as he calls it.

As an anthropologist, he has learned to embrace the Internet as a tool  
to study, but as a person, he thinks applications such as the Grid have  
frightening Big Brother potential.

"Am I trying to sabotage the Internet's effectiveness in terms of  
monitoring human behaviour?" he asks. "Absolutely. Once it becomes  
effective, you're going to have to."

© National Post 2004
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 10943 bytes
Desc: not available
URL: <http://listserv.linguistlist.org/pipermail/ilat/attachments/20040214/fc3de794/attachment.bin>