Home / Technology / Facebook Shovels Out Some of the Filth

Facebook Shovels Out Some of the Filth

facebook is enmeshed in yet another brouhaha over its advertising policies, but this time it’s not the users making a fuss — it’s the .

Major clients such as Nissan and Nationwide recently pulled the plug on their after their ads showed up next to objectionable, sometimes downright hateful, content.

It’s not that users have been silent about the offensive posts. The advertisers were prompted to act following an outpouring of ire for allowing their ads to be placed next to content glorifying sexual violence or gay bashing.

Advocacy groups had also been lobbying Facebook to remove some of these posts in recent months but apparently to no avail.

Got Their Attention

Facebook reacted almost immediately to the loss of advertising, however, publishing a statement on Tuesday clarifying its stance on controversial posts and hate speech, and explaining why it took so long to remove offending content.

Facebook wants the site to be a place where thoughts and exchanges are freely expressed, it said, even if that means some of user posts may be offensive or controversial.

However, Facebook will not tolerate speech that is directly harmful, and it prohibits “hate speech.”

Facebook acknowledged that it has been lax in identifying and removing hate speech, particularly around issues of gender-based hate.

“In some cases, content is not being removed as quickly as we want,” said Marne Levine, VP of global public policy. “In other cases, content that should be removed has not been or has been evaluated using outdated criteria.”

Facebook has set out a series of steps to ensure that its safeguards work better, including updating training for the teams that evaluate reports of hateful speech, and establishing more formal lines of communications with representatives of women’s groups and other such associations.

Tighter Targeting?

From a practical standpoint, there are other measures Facebook can take to ensure that all parties — or at least the advertisers — are kept happy.

For one thing, it could deploy tighter targeting, suggested Rob Enderle, principal of the Enderle Group.

“If you tightly target ads and content to users, then you are far less likely to put content and ads together that don’t belong together,” he told us.

While the tighter targeting could be automated to some degree, removing content requires human interaction at some point in the process, noted Charles Palmer, associate professor of new media at Harrisburg University of Science and Technology.

“I sympathize with Facebook, because identifying and removing hateful speech and images is a manual process and it can’t be done quick enough for the general public,” he told us. “Facebook is relying on the public to report inappropriate images, but that is a slow and tedious process. ”

Not a Fight

One thing this episode is not is really a First Amendment fight.

The First Amendment only limits what of government, Martin Margulies, professor emeritus at the Quinnipiac University School involving Law, told us.

“Far via curtailing First Amendment rightt under the law, the advertisers have economic crisis Amendment right of their own — to ask Facebook or myspace to reject or range them from speech that they deem offensive, ” this individual continued.

“And Facebook, an individual actor, likewise has economic crisis Amendment right to decide tips on how to respond — by rejecting the actual advertisers’ pressure or by yielding to it. They do not violate the very first Amendment if they elect to do the latter, ” Margulies talked about.

Further, there are no laws against so-called loathe speech, he added. “Such legislation, when enacted, have uniformly been pronounced unconstitutional beneath the First Amendment. “.

Source: TechNewsWorld

Leave a Reply

Your email address will not be published. Required fields are marked *

*

Scroll To Top