Consulte o glossário usando este índice

Especial | A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | TODOS

S

Snared in the Web of a Wikipedia Liar

(Última edição: domingo, 4 de dezembro de 2005 às 16:26)
Rewriting History

Snared in the Web of a Wikipedia Liar

04seelye.jpg
Carl Mydans/Time Life Pictures -- Getty Images

FALSE WITNESS How true are "facts" online?

Published: December 4, 2005, NYT
 

ACCORDING to Wikipedia, the online encyclopedia, John Seigenthaler Sr. is 78 years old and the former editor of The Tennessean in Nashville. But is that information, or anything else in Mr. Seigenthaler's biography, true?

The question arises because Mr. Seigenthaler recently read about himself on Wikipedia and was shocked to learn that he "was thought to have been directly involved in the Kennedy assassinations of both John and his brother Bobby."

"Nothing was ever proven," the biography added.

Mr. Seigenthaler discovered that the false information had been on the site for several months and that an unknown number of people had read it, and possibly posted it on or linked it to other sites.

If any assassination was going on, Mr. Seigenthaler (who is 78 and did edit The Tennessean) wrote last week in an op-ed article in USA Today, it was of his character.

The case triggered extensive debate on the Internet over the value and reliability of Wikipedia, and more broadly, over the nature of online information.

Wikipedia is a kind of collective brain, a repository of knowledge, maintained on servers in various countries and built by anyone in the world with a computer and an Internet connection who wants to share knowledge about a subject. Literally hundreds of thousands of people have written Wikipedia entries.

Mistakes are expected to be caught and corrected by later contributors and users.

The whole nonprofit enterprise began in January 2001, the brainchild of Jimmy Wales, 39, a former futures and options trader who lives in St. Petersburg, Fla. He said he had hoped to advance the promise of the Internet as a place for sharing information.

It has, by most measures, been a spectacular success. Wikipedia is now the biggest encyclopedia in the history of the world. As of Friday, it was receiving 2.5 billion page views a month, and offering at least 1,000 articles in 82 languages. The number of articles, already close to two million, is growing by 7 percent a month. And Mr. Wales said that traffic doubles every four months.

Still, the question of Wikipedia, as of so much of what you find online, is: Can you trust it?

And beyond reliability, there is the question of accountability. Mr. Seigenthaler, after discovering that he had been defamed, found that his "biographer" was anonymous. He learned that the writer was a customer of BellSouth Internet, but that federal privacy laws shield the identity of Internet customers, even if they disseminate defamatory material. And the laws protect online corporations from libel suits.

He could have filed a lawsuit against BellSouth, he wrote, but only a subpoena would compel BellSouth to reveal the name.

In the end, Mr. Seigenthaler decided against going to court, instead alerting the public, through his article, "that Wikipedia is a flawed and irresponsible research tool."

Mr. Wales said in an interview that he was troubled by the Seigenthaler episode, and noted that Wikipedia was essentially in the same boat. "We have constant problems where we have people who are trying to repeatedly abuse our sites," he said.

Still, he said, he was trying to make Wikipedia less vulnerable to tampering. He said he was starting a review mechanism by which readers and experts could rate the value of various articles. The reviews, which he said he expected to start in January, would show the site's strengths and weaknesses and perhaps reveal patterns to help them address the problems.

In addition, he said, Wikipedia may start blocking unregistered users from creating new pages, though they would still be able to edit them.

The real problem, he said, was the volume of new material coming in; it is so overwhelming that screeners cannot keep up with it.

All of this struck close to home for librarians and researchers. On an electronic mailing list for them, J. Stephen Bolhafner, a news researcher at The St. Louis Post-Dispatch, wrote, "The best defense of the Wikipedia, frankly, is to point out how much bad information is available from supposedly reliable sources."


(Page 2 of 2)

Jessica Baumgart, a news researcher at Harvard University, wrote that there were librarians voluntarily working behind the scenes to check information on Wikipedia. "But, honestly," she added, "in some ways, we're just as fallible as everyone else in some areas because our own knowledge is limited and we can't possibly fact-check everything."

In an interview, she said that her rule of thumb was to double-check everything and to consider Wikipedia as only one source.

"Instead of figuring out how to 'fix' Wikipedia - something that cannot be done to our satisfaction," wrote Derek Willis, a research database manager at The Washington Post, who was speaking for himself and not The Post, "we should focus our energies on educating the Wikipedia users among our colleagues."

Some cyberexperts said Wikipedia already had a good system of checks and balances. Lawrence Lessig, a law professor at Stanford and an expert in the laws of cyberspace, said that contrary to popular belief, true defamation was easily pursued through the courts because almost everything on the Internet was traceable and subpoenas were not that hard to obtain. (For real anonymity, he advised, use a pay phone.)

"People will be defamed," he said. "But that's the way free speech is. Think about the gossip world. It spreads. There's no way to correct it, period. Wikipedia is not immune from that kind of maliciousness, but it is, relative to other features of life, more easily corrected."

Indeed, Esther Dyson, editor of Release 1.0 and a longtime Internet analyst, said Wikipedia may, in that sense, be better than real life.

"The Internet has done a lot more for truth by making things easier to discuss," she said. "Transparency and sunlight are better than a single point of view that can't be questioned."

For Mr. Seigenthaler, whose biography on Wikipedia has since been corrected, the lesson is simple: "We live in a universe of new media with phenomenal opportunities for worldwide communications and research, but populated by volunteer vandals with poison-pen intellects."


Software 'cannot stop cheating'

(Última edição: quinta-feira, 24 de novembro de 2005 às 09:39)
Last Updated: Tuesday, 22 November 2005, 17:12 GMT o.gif
dot_629.gif

http://news.bbc.co.uk/2/hi/uk_news/education/4460702.stm

Software 'cannot stop cheating'
Keyboard
The exam watchdog has warned of plagiarism using the internet
Technological solutions alone will not be enough to prevent children using the internet to cheat in their coursework, a government adviser has said.

Professor Jean Underwood of Nottingham Trent University says it is up to teachers and parents to show that plagiarism is inappropriate.

The government has asked Professor Underwood to provide technical advice on how to detect internet cheating.

It has commissioned a review of GCSE coursework in each subject.

The move comes after report by the Qualifications and Curriculum Authority said downloading essays from the internet "could not be controlled".

Internet search

Professor Underwood, an expert in the impact of new technologies on teaching and learning, said less repetitive and more creative questioning would reduce the scope for cheating.

o.gif
COURSEWORK CHEATING
_41046794_revision203.jpg
start_quote_rb.gif The availability of the internet is a powerful aid to learning but carries a new generation of risks of plagiarism. end_quote_rb.gif
QCA report on coursework
inline_dashed_line.gif

She said the parameters of her study were still to be defined, but she wanted to help find solutions "so that everyone is reassured that coursework is valid, relevant and secure".

Rules should be made clearer, Professor Underwood said.

"We all reject websites which sell essays, but where does that leave us when there are so many help books to get pupils through their GCSEs? Where is the line?"

And parents need to understand that by doing work for their children, or telling them what to include, they are not allowing them to learn effective research - an important skill for later life.

"If a parent helps their child to carry out an efficient internet search, I personally do not see anything wrong in that," she said.

"But downloading five papers from the internet would be a borderline crossed."

She said the government has recognised there is concern and will put down guidelines around February next year.

But it was "all our jobs to collectively show that cheating should not happen".

Teachers have voiced concerns that there are inconsistent guidelines across exam boards regarding how much guidance teachers should give to pupils.

They are also concerned that providing templates and checklists for work leads to "cloned essays" which are difficult to tell apart.

"Templates are worrying, if they lead to the pupil not understanding the material," Professor Underwood said.

The NASUWT teachers' union said it was important to keep the issue of plagiarism in proportion.

But it welcomed the idea of clearer guidelines, but said policing every line of work for plagiarism would "place an impossible burden upon teachers".

Coursework is marked internally within schools, while exam boards call in samples of the work for external checks, known as "moderation".

o.gif
start_quote_rb.gif The internet is a wonderful thing with the power to change lives - but there will always be a downside end_quote_rb.gif
Professor Jean Underwood

At GCSE level, it varies from 20% of the overall qualification in double science, to 60% in art and design.

At A-level it can be from nothing to 30%, or 60% in the case of art and design.

Education Secretary Ruth Kelly said coursework should only be used where it is the most appropriate assessment method.

The QCA's report found coursework was a valuable tool to stimulate pupils' own learning, but that the value placed on it by teachers varied between subjects.

Technical solutions

Professor Underwood said technology could help ameliorate the problem but was "no quick fix".

She said software already existed to help schools ascertain whether work was the pupil's own.

"It can even be as simple as typing a phrase into Google."

"If a phrase has been plagiarised, sites will bring it up."

"Software is already out there that schools can use, from the Joint Information Systems Committee."

Exam board Edexcel and the Joint Council for Qualifications said they were working with the Plagiarism Advisory Service with a view to rolling out plagiarism detection software.

A JCQ spokesperson said it would reduce the potential to use or re-use work produced by other people.

Professor Underwood said some software could check as well as mark work. But she said some clever students would find ways round such programmes.

"One method used is to translate phrases in papers into a different language and then back into English with a translation tool," she said.

She said that tackling firms providing essays for sale or download would not guarantee children could not access essays, as hackers could still make them available for sale.

"We need to think smart on an academic and technological level," she said

"The internet is a wonderful thing with the power to change lives - but there will always be a downside."


Students Ace State Tests, but Earn D's From U.S.

(Última edição: domingo, 27 de novembro de 2005 às 22:38)

Students Ace State Tests, but Earn D's From U.S.

Published: NYT, November 26, 2005
 

After Tennessee tested its eighth-grade students in math this year, state officials at a jubilant news conference called the results a "cause for celebration." Eighty-seven percent of students performed at or above the proficiency level.

But when the federal government made public the findings of its own tests last month, the results were startlingly different: only 21 percent of Tennessee's eighth graders were considered proficient in math.

Such discrepancies have intensified the national debate over testing and accountability, with some educators saying that numerous states have created easy exams to avoid the sanctions that President Bush's centerpiece education law, No Child Left Behind, imposes on consistently low-scoring schools.

A comparison of state test results against the latest National Assessment of Educational Progress, a federal test mandated by the No Child Left Behind law, shows that wide discrepancies between the state and federal findings were commonplace.

In Mississippi, 89 percent of fourth graders performed at or above proficiency on state reading tests, while only 18 percent of fourth graders demonstrated proficiency on the federal test. Oklahoma, North Carolina, Alabama, Georgia, Alaska, Texas and more than a dozen other states all showed students doing far better on their own reading and math tests than on the federal one.

The chasm is significant because of the compromises behind the No Child Left Behind law. The law requires states to participate in the National Assessment - known to educators as NAEP (pronounced nape) - the most important federal measure of student proficiency.

But in a bow to states' rights, states are allowed to use their own tests in meeting the law's central mandate - that schools increase the percentage of students demonstrating proficiency each year. The law requires 100 percent of the nation's students to reach proficiency - as each state defines it - by 2014.

States set the stringency of their own tests as well as the number of questions students must answer correctly to be labeled proficient. And because states that fail to raise scores over time face serious sanctions, there is little incentive to make the exams difficult, some educators say.

"Under No Child Left Behind, the states get to set the proficiency bar wherever they like, and unfortunately most are setting it quite low," said Michael J. Petrilli, a vice president of the Thomas B. Fordham Foundation, which generally supports the federal law.

"They're telling the public in their states that huge numbers of students are proficient, but the NAEP results show that's not the case," Mr. Petrilli said.

Other educators and experts give different reasons for the discrepancy between state and federal test results. A Standard & Poor's report this fall listed many reasons for such differences, among them that the National Assessment is a no-stakes test, while low scores on state tests lead to sanctions against schools.

The report noted that the National Assessment is given to a sampling of students, whereas schools administer state tests to nearly all students. The tests serve different purposes, with the federal one giving policy makers a snapshot of student performance across the nation, while state tests provide data about individual performance. Because of these differences, some state officials say it is unfair to compare the test results.

But the report by Standard & Poor's, which has a division that analyzes educational data, also noted some states' tests are just easier.

G. Gage Kingsbury, director of research at the Northwest Evaluation Association, a nonprofit group that administers tests in 1,500 districts nationwide, said states that set their proficiency standards before No Child Left Behind became law had tended to set them high.

"The idea back then was that we needed to be competitive with nations like Hong Kong and Singapore," he said. "But our research shows that since N.C.L.B. took effect, states have set lower standards."

Not all have a low bar. In South Carolina, Missouri, Wyoming and Maine, state results tracked closely with the federal exam.

South Carolina is a state that set world-class standards, Mr. Kingsbury said. The math tests there are so difficult that only 23 percent of eighth graders scored at or above the proficiency level this year, compared with 30 percent on the federal math test. South Carolina officials now fear that such rigor is coming back to haunt them.

"We set very high standards for our tests, and unfortunately it's put us at a great disadvantage," said Inez M. Tenenbaum, the state superintendent of education. "We thought other states would be high-minded too, but we were mistaken."

South Carolina's tough exams make it harder for schools there to show the annual testing gains demanded by the federal law.

This year less than half of the state's 1,109 schools met the federal law's benchmark for the percentage of students showing proficiency, a challenge that will get tougher each year. As a result, legislators are pushing to lower the state's proficiency standard, Ms. Tenenbaum said, an idea she opposes.

Because of the discrepancies, several prominent educators are now calling for a system of national testing that counts, like those at the heart of educational systems in England, France and Japan.

"We need national standards and national tests," said Diane Ravitch, a professor at New York University who is a former member of the National Assessment's board. "I conclude that states are just looking to make everybody feel good."

Ms. Tenenbaum too says the differences among states have convinced her of the need for a national test. "I think we should all just take the NAEP," she said. "Get it out of the states' hands."

But Representative John A. Boehner, chairman of the House Committee on Education and the Work Force, defended states' rights to define proficiency as they see fit and said that over time comparisons with the federal test would force them to draw up better tests.

"The bright lights of accountability are going to shine on the states who are kidding themselves," said Mr. Boehner, Republican of Ohio.

The battle lines have long been sharp in the testing debate. Most corporate leaders favor national testing, said Susan Traiman, a director at the Business Roundtable, a group that represents corporate executives.

Opponents include liberal groups that dislike all standardized testing; the testing industry itself, which has found lucrative profits in writing new exams for all 50 states; and political conservatives who fiercely resist any intrusion on states' rights to control curricula and tests.

Margaret Spellings, the secretary of education, says that the comparison of state and federal tests provides useful information. "It allows us to shine a light," she said. "This is a truth-in-advertising type deal."

But Ms. Spellings has declined to criticize states whose tests appear to overstate the percentage of their students who are proficient. The law leaves it to states to calibrate their accountability systems, including how difficult they make their exams, she said. "We're not going to sit up in Washington and look at all those moving parts," Ms. Spellings said.

The National Assessment uses three performance levels to classify student results: advanced, which denotes superior performance; proficient, which indicates that students have "demonstrated competency" and basic, which indicates students have attained only "partial mastery of prerequisite knowledge and skills." Many students also score below basic, which the National Assessment's governing board does not classify as an achievement level.

On Oct. 19, the day the federal results were released, Ms. Spellings urged reporters to compare the percentage of students performing at the proficiency level on state tests with the percentage of students performing at the basic level on the federal test.

Many state officials said they also preferred that comparison, which greatly softens the discrepancies. In Tennessee, for instance, the 66-point gap between the federal and state results in eighth-grade math shrinks to just 26 points if the state results are compared with the federal measure of basic skills.

"NAEP's basic is comparable to our proficient," said Kim Karesh, a spokeswoman for the Tennessee Department of Education. "Now whether Tennessee's test is stringent enough is something that we're reviewing constantly. Nobody here would say we have a perfect test."

Officials in many other states whose scores differed sharply from those of the National Assessment cried foul over the very idea of comparing the results.

"The comparison to NAEP is not fair," said Mitch Edwards, a spokesman for the Department of Education in Alabama, where 83 percent of fourth-grade students scored at or above proficient on the state's reading test while only 22 percent demonstrated proficiency on the federal reading test. "Making comparisons to the NAEP becomes very difficult without giving the impression that some states are not measuring up to others or to the nation."

In Georgia, 83 percent of eighth graders scored at or above proficient on state reading tests, compared with just 24 percent on the federal test. "Kids know the federal test doesn't really count," said Dana Tofig, a spokesman for the State Department of Education. "So it's not a fair comparison; it's not apples to apples."