Online attacks on individuals and organizations can have far-reaching and serious effects. Not merely incidences of questionable behavior, at their worst attacks can cause stress disorders in their victims and result in reputational and economic harm that’s difficult to undo.

But for some targets, simply being able to identify and publicly reveal their attacker is enough.

That’s what happened in the case of model Liskula Cohen. In 2009 Cohen was labeled  “skanky” and a “ho” by an anonymous blogger using Google’s Blogger platform. She sued, determined to find out who made the disparaging remarks, and a judge ordered Google to hand over the blogger’s IP address, leading to the identification of the individual. (Cohen subsequently dropped her lawsuit, which led some members of the legal community to accuse her of having abused the legal process by filing suit simply to determine the identity of her attacker, while others supported her action as a practical realization that she had been vindicated in the public arena).

Sports writer Jeff Pearlman, meanwhile, decided to track down and confront his Twitter detractors on his own. He wrote about his experience in Tracking down my online haters, noting that “when journalists take the time to respond personally to venomous notes, proving that they are made of flesh and blood, the reaction is strikingly — and puzzlingly — positive.” (Read the response from one of the Twitter detractors, as well as further analysis of the entire situation, here.)

And perhaps therein lies a part of the solution to the problem: putting a human face on the targets and requiring accountability from their attackers.

Online hostility takes on many different forms – from the relatively mild to the extreme and illegal.

How can you tell if online speech is hostile? First examine the context the statement is in. Facts and circumstances often impact whether something is intentionally aggressive or just a sharp, off-the-cuff comment. Keep in mind that something can fall within the realm of hostility without being illegal.

Some questions to ask yourself:  (1) Was it  a one-time or ongoing attack?  (2)  Was it perpetrated anonymously, by an online pseudonym, or under someone’s real name?  (3) Was there one attacker or are several involved, i.e.  mob attack,  (4) Was the attack unprovoked or a defensive response? (5) Was it made privately to the target or publicly online? (6) Do the attacker and the target know each other, and if so, what is their relationship? (7) Was the attack made exclusively online or has it migrated offline as well? (8) Was it about the target’s appearance, character or behavior? (9) Was a specific or general threat made against the target? (10) Could someone interpret that attack as humor or satire?

It bears repeating – online hostility covers a broad spectrum of actions which may or may not be illegal depending on content and the law. Actions include but aren’t limited to:

• Rude comments aimed at target
• Online screaming at a target
• Belittling target
• Teasing or mocking target
• Online cursing at target
• Insulting and name-calling
• Negative insinuations about target that have no basis of truth or are knowingly false
• Negative gossip about target that is intended to harm target, which may or may not have a basis of truth
• Ad hominem attacks against target
• Criticism of target’s appearance, age, gender, race, intellect, and so forth
• Revealing personal or private information about target
• Revealing embarrassing or damaging information about target
• Privacy intrusions
• Sharing or publishing information the target thought was private between the attacker and him/her, including photographs
• Impersonating the target online
• Spreading half-truths of lies about target and his/her family
• Photographic or other visual manipulations placing target or his/her family in an embarrassing or negative light
• Accusations of fabricated wrongdoing by target
• Threats of attacks against target’s reputation
• Sexual or racial harassment or harassment due to sexual orientation
• Hate speech
• Defamatory statements about target or his/her family
• Google-bombing campaigns
• Denial-of-service attacks
• Hacking target’s site or online accounts
• Threats of physical attacks against target and his/her family
• Cyberstalking

Note: Download our Quick Overview: The Spectrum of Online Hostility here.

What does the media do when it discovers it has tweeted an error? That’s the question journalist Craig Silverman answered in Columbia Journal Review’s To Delete or Not to Delete?, and he found that it depends on the news organization.

When it was erroneously reported that Rep. Gabriella Giffords died in the recent Arizona shooting, some organizations, such as NPR, chose to leave an existing incorrect tweet online, arguing that once posted, people had already seen it and the organization didn’t want to be accused of “trying to cover our tracks on Twitter,” but instead decided to “be transparent about the mistake.” Others, such as PBS NewsHour, chose to delete a tweet to avoid incorrect information from being retweeted and thereby spreading the false information. Silverman’s own view is that “I’m not in favor of news orgs deleting incorrect tweets. Take time to push out correct info & contact RTers to alert to new info.”

While scrubbing (i.e. changing an error in an online story without including an official correction) is frowned upon in journalism, an important question is whether the same rules should apply to regular people.

One individual suggested that “maybe Twitter should just create a redact function that crosses out the tweet in stream and auto notifies RTers by email” – an intriguing idea. Barring the existence of that option, however, perhaps the answer of what individuals should do must be decided on a personal and  case-by-case basis.

Possible things to consider include: (1) What is the information posted – does the tweet contain information that could potentially harm an individual or entity? (2) What is the likelihood of the tweet having already been read or retweeted – how long was the tweet online before the error was discovered? (3) What is the likelihood of the corrected tweet receiving as much attention as the initial error? (4) What is the likelihood of success in being able to contact the individuals or sites that have reposted the incorrect information? (5) Is it even important to correct the initial error?

We don’t yet have all the answers to what’s best in today’s hyper-connected environment, so perhaps we should err on the side of caution…or at least be aware of the potential outcomes of our actions.