6 key points from a Twitter conversation about comments on news sites

 by Martin Belam, 22 June 2011

Yesterday I got involved in a long Twitter conversation about anonymous and pseudo-anonymous comments on news websites involving Adam L. Penenberg, Mathew Ingram, Anna Tarkov, Amrita Mathur and Brad King. It was kicked off as people responded to this piece on the issue by Mathew at Gigaom.

Anon posts, fine. Anon comments? I disagree. “Anonymity has real value, both in comments and elsewhere,” by @mathewi: http://bit.ly/mChVln

I’ve tried to sum up the six main points I was making in bursts of slightly more than 140 characters, and I’ve tried to interweave some of the conversation.

1. There can be value in pseudo-anonymous comments

@mathewi I cud be swayed, tho, if u cud point to 10 really import but necessarily anon comments made on gigaom. http://bit.ly/mChVln

I’ve blogged before about gherkingirl’s powerful comments about being a rape victim - a blog post it transpired which she noticed. There are a host of situations where a news site can get insight from users where forcing them to provide a real visible identity would prevent them contributing. The Guardian’s coverage of public sector issues in the UK would be poorer if people couldn’t post in a semi-anonymous way about the things that happen, for example, in schools over taking biometric details of children.

2. Persistent, accountable, identity is important

@currybet @mathewi @thebradking Anon comments akin to Victorian masquerade ball. Wear a mask + nastiest, most outrageous behavior tolerated.

+1 RT @currybet: for me, the key is for people to have *persistent* identity within a community, even if anon/psuedo-anon.

If using real names solved online community problems, there would be no bullying on Facebook. What is important within a news environment is that people have persistent, recognisable identities, and that they can be held to account for their behaviour with community and moderation sanctions. There are plenty of people on guardian.co.uk who I recognise from their username and avatar as either helpful members of the community or nuisances. I don’t need to see their real name for that, just as many people only know me on the internet as currybet, not Martin.

3. The Slashdot model does not work for mainstream news

@currybet the slashdot system has human ed control. U can absolutely have a stance w it. What don't you think it does?

@currybet also: ui is easy to fix.akin to saying wiki software is useless cause of ui. It's open code.

The Slashdot method of ranking and hiding user contributions is great for Slashdot, but it does not work for general mainstream news. The user interface is complicated for non-geek users. More importantly, the “newbie” leaving a comment on a live blog about a revolution in Egypt might be eye-witness testimony. You don’t want to wait three months until they have left enough comments to gain enough karma to be visible to all before displaying it or starting to verify it. [See comments]

There is another wider issue, over editorial stance. People cause trouble on Slashdot because they are pro-Apple or anti-Microsoft or Google fanboys etc. Nobody trolls on Slashdot because they hate what Slashdot itself stands for editorially.

News brands are different.

People come to The Telegraph or The Guardian or the New York Times or Fox News specifically to pick a fight with the perceived editorial outlook of the news source. That makes the requirements of the system that you need to put into place for community management different.

4. Only open comments where you’ve understood what you are trying to achieve

News sites that insist on opening comments on every single story are missing the point. Only open comments where you believe they have the possibility of adding value to the site.

At The Guardian we don’t open comments on news stories with ongoing legal implications, for example, and we’ve learned from being publicly chastised by users for opening comments on a live blog about a developing news situation.

The comments you get will depend on the content and tone of the original article. As my colleague Meg Pickard observes, if you write provocative comment pieces, expect people to be provoked.

I have a line I use in presentations, saying that one of the problems with comments directly underneath news pieces is that generally what moves people to leave a comment is that they are angry about the news, or they are angry about the way that the news has been reported. Either way, they are angry, which isn’t condusive to good community interaction.

@mathewi @currybet Often what makes people angry is the fact that no one responds. Not the reporter, not someone else at the news org.

@mathewi @currybet I'm always telling people this at MSM news orgs where they tend to dismiss "crazy" commenters as not worth their time...

5. Active community management is vital

News sites that try to host communities, but which have the arrogance to think that community management is an unimportant role are doomed to have dreadful online communities. The ratio between “content creators” and “community people” in most news rooms is vastly at odds with what you’d expect to see if you were trying to start a website with a large engaged community element.

That ought to be a clue.

6. Community managers on a news site should work with journalists, not replace interaction with journalists

@karimkanji Question on that: does community mgmt operate in parallel to content mgmtt? Or are they one and the same? cc: @mathewi @currybet

@currybet Gotcha. So one does the 'talking' while the other does the writing/ content generation. In theory, they could be the same person.

The best type of community management I’ve seen on news sites is when a community person is alerting the principal journalist or editor of a piece or live blog to comments that need responding to, or which are interesting and move the story along. They don’t assume the role of speaking on behalf of the author, but they help facilitate that conversations, as well as being a presence themselves. And community management doesn’t just mean engaging with comments “below the line”. It means watching how the audience is reacting to stories on Facebook and Twitter and across the blogosphere, and alerting editorial staff and the rest of the audience to the best bits of that activity.

See also: The ongoing debate over anonymous comments on newspaper websites - March 2011
In praise of... newspaper website comments - August 2010
“Real name” comments on news websites - the up and the downside - May 2010
My notes from the BBC Social Media Summit - May 2011


I should add that via Twitter, Brad King says of my points about Slashdot: "yikes, I don't think you understand what the slashdot software does :( that description doesn't relate to its function at all", "your argument against the software isn't right bc slash dot software doesn't work at all how you say it does", "karma, for instance, doesn't do what you say. It's used to give individuals Ltd mod function, not delay a post."

I wanted to add something.

The problem we're still facing under the surface of all this is that many, many news organizations have still not seen the value of hiring a community manager. At least I assume that's why this job doesn't exist in every newsroom. So this work doesn't get done, uncivil comments pile up, resentment builds on both sides and journalists become ever more hardened in the notion that their audience isn't to be engaged with.

It's true that some journalists are doing the job themselves on their own beat or column, but it seems they are still in the minority and a community management person (or people) is still needed. The question is how to spread the idea that this is as necessary a position for the news org's long-term survival as the ad sales person or the circulation director. How can we do this??

@Anna - I think it has to be done by emphasising and demonstrating the business benefits of healthy communities. I have yet to see a decent numbers-based analysis of the value of constructive comment communities to the health of a business, but my experience suggests it pays off in page views, time on site, visitor loyalty - all delicious, money-making metrics that can be quite persuasive to the people who make the hiring decisions. Persuading them that it may also prevent legal issues, help source stories and improve the paper's standing in the community might also help.

On Facebook, Ian Betteridge has commented: "Hmm. I think your comment about the rape victim is a bit of a classic case of positing a general point from an edge case. I know you say there are "a host" of other examples, but that host still forms a minuscule proportion of general comment discourse. I still think the lesson of The WELL that "you own your own words" hasn't been adopted enough online. The WELL wasn't always not anonymous - forums where the subject matter was sensitive allowed anonymity - but the concept of never allowing yourself to forget that there were real people at the keyboard on both ends worked. Persistent anon identities don't cut it, I think: someone is just as likely to create persistent anti-social personae as ones which are more moderate - sometimes even as sock puppets to make their own point. Of course a real identity isn't the complete answer - but what it does do is demonstrate quickly that what you say online has real world consequences. The point you make about Facebook actually demonstrates this well - bullies do get outed, and face real world punishment for their bullying."

Keep up to date on my new blog