The vexing issue of managing football comments on a newspaper website

 by Martin Belam, 14 October 2012

On Friday I was involved in a Twitter conversation with Guardian journalists Sean Ingle and Sid Lowe about the value of comments underneath football stories on newspaper websites.

It was sparked by the news that the Daily Record has suspended comments on football articles, because, in their own words:

“We continue to see an unacceptable level of behaviour that breaches our terms and conditions. We are talking about personal abuse directed against other users and members of our staff, foul language, hate speech [and] sectarianism.”

The Scottish situation is slightly more complex for news outlets than in England. The law has reacted to fans of the two biggest clubs indulging in sectarian abuse with the Offensive Behaviour at Football and Threatening Communications Act 2012, which came into force earlier this year. This makes the role of moderation at the Daily Record more likely to need advice from the legal team.

Hosting football communities can be vexing and problematic.

When I worked at the BBC in the early 2000s, the 606 messageboards were one of the busiest areas of user-generated content on the site, and also one of the most problematic. The volume of comments and the level of moderation required made them a massive drain on technical and editorial resources, and a source of friction between the central new media department and BBC Sport when things weren’t working well. To complicate things, the audience weren’t shy of playing the “I pay my Licence Fee so how dare you censor me” card when moderated or banned, leading to lengthy exchanges of emails.

News organisations can host good sport and football community interactions. On Twitter Sean Ingle cited the Guardian’s county cricket blog, Joy of Six, Classic YouTube and ‘The forgotten story of...’ as areas where “it can be done. Not easy though.” In fact, just last week I’d had a good chat on the Guardian site under stories about Davide Gualtieri’s San Marino goal against England, and Kevin Keegan’s resignation as England manager.

But, with a period away from working in-house at a news organisation, my perception of user comments has shifted. I still miss them when I read a printed paper. However, when you work for a site running community areas, you tend to see the 2% of really good community interactions. You point at things like “Readers recommend”, where the community comes together to suggest songs for a playlist on a theme, and sometimes even gets to choose the end result.

Outside of the business though, you mostly see the 98%.

I read a newspaper story headline online, and immediately know that the comment section is going to be hateful, without even having to read it. It jars with me that I know what is going to be “below the line”, even if I don’t cast my eye there.

The 98% drowns out the 2%.

So why is it that news organisations have struggled to build good communities on their websites?

I think volume is the big problem here. Everything about the news media transitioning from print-only to “digital first” has been about volume. About producing an ever greater volume of content to attract an ever greater volume of eyeballs to generate fractions of pennies and cents on serving volumes of advertising. About prioritising writing articles that will generate 500 angry, troll-ish comments as a by-product of the page views it will get, rather than investing the time to only hand select ten really great responses to publish.

For most news organisations the balance of community activity is all wrong.

There are millions of users who can potentially behave in a disruptive manner, hundreds of journalists publishing articles, and a handful of staff dedicated to moderating or shaping the community interaction that goes on at the foot of the articles.

If I was starting a site where I wanted to grow a great community, I’d be hiring more people to manage, nurture and promote the community activity than I would be hiring people to generate content. News organisations are set up in exactly the opposite structure.

The trouble for news organisations isn’t that people are complete dickwads online - although that doesn’t help. The trouble is that community and comment threads are something added on to what they see as their primary purpose, not their primary purpose.

And that doesn’t scale.


Martin - really enjoyed reading this entry but feel like you've ended it too soon.

Simple answer - make all fora publish real names.

I agree that news orgs need to do much more to organise and bring to the top the best comments and there is real value in that. I don't think it should the 'primary' thing though - the primary value is the story itself. There wouldn't be a conversation without that.

must be a solution to this one tho with a mix of tech and people - no?

Good article but spoiled slightly by you saying "The Scottish situation is slightly more complex for news outlets than in England. The law has reacted to fans of the two biggest clubs indulging in sectarian abuse" The abuse comes from SOME fans of ALL clubs.

I feel the same as the poster above in terms of you finishing the piece prematurely. I used to be an active participant on the Guardian football blog but feel like the negative and often vindictive voices drown out those looking for real discussion, to the point where the main article is not even a consideration. Terrace humor is great for football blogs but where do moderators draw the line? The Guardian has Terms of Use that often they don't adhere to, perhaps because of resource issues, but mainly, I feel, because it left to moderators making a subjective judgement of what is acceptable and eha is not. The challenge is to create a thriving blogging community where debate is open but where those hiding behind avatars need to be accountable for their own voice.

Steve, I must have re-written that line about 100 times in the full knowledge that no way of phrasing it was going to avoid someone coming on to the comments to complain about what I said. It seems impossible to even mention sectarian abuse in Scottish football without somebody shooting the messenger.

“Simple answer - make all fora publish real names”

Paul, that is a simple answer, and it simply doesn’t work like that. With the Guardian Facebook app we had some comment threads on Facebook with people using real names and faces, and comment threads on the same article on the main site where more anonymous commenting was allowed. There was no appreciable difference between the quality of comments and the level of poor behaviour. And there are very good reasons why you sometimes do want to get anonymous comments from the audience.

I have a lot of experience with Rangers/Celtic free-access public boards and quite simply the only answer IMHO is to close them all down. I have no probs with heavy football banter but the Old Firm sites plumb the depths of human depravity as do some other non-football related ones.

No media organisation, in the current financial climate, can afford the 'policing' costs required to adequately control the boards or accept the emotional strain that staff are exposed to. The legal power under the Scottish act seem to be ignored as are sectarian and other offensive comments. The only answer is to restrict access to paying subscribers so that their identity can be established.

Away from media boards, genuine fans can register with their club site, especially if they have a season ticket, and it is not beyond current technology to allow fans to move across networks retaining their club ID.

And it is ID that is the key to all of this because 99.99% of the unacceptable only happens because the poster is anonymous.

Everyone wants to make money off internet traffic but no one really wants to pay to police it - so we will continue to see vulnerable people actually hounded to death by people who either find it hilarious or who have a very warped personality.

Tbh I despair when I see articles likes this discussing the loss of nice community threads. Before we can get there we have to deal with the tidal wave of poison and hatred that is drowning forums.

There doesn't appear to be an easy answer to this question but if 99.9% of the unaccepatable activity occurs because the poster is anonyomous then that has to be a starting point.

Yes you do occasionally want anonymous posters who might not post otherwise but on a football thread?

If 99.9% is an accurate figure then perhaps restrict anonymity where you want the trollish comments reduced and allow it where it won't be a problem - anonymity isn't a right it's a privilege.

I don't think anonymity is as much of a problem as some people seem to. I was a regular on several Usenet groups back in the early 90s, everyone used their real names and things still got ugly.

When I first started commenting on the Guardian's Comment is Free, things often got heated, but no more so than had soc.culture.british back in my Usenet days. And people become quite invested in their user names. It was fairly traumatic for several CiF regulars of my acquaintance when they were banned.

I am actually far freer to be honest about my views and experiences as Montana Wildhack than I ever could be under my real name. I live in a small town. I could lose my job if some of my online comments became associated with my real name -- not because they were obscene or abusive, but because I sometimes described work situations with both from my current job as a SEN assistant and my former position as a DV/sexual assault advocate. From MontanaWildhack, they're anonymous anecdotes. From (my real name) they're potential violations of confidentiality (despite the fact that, even under my real name, specific situations would have been hard to identify).

I don't post much on CiF these days and I no longer comment on threads where I'm likely to refer to work situations, but I keep this alias because it's what most of my online acquaintances know me by. I don't think I'm any more abusive as Montana Wildhack than I would be under my real name (we're both foul-mouthed harridans).

In 97, when we were young, I was asked to set up the BBC News Online forum/interaction section. After much thought it was clear that, as you point out above, the scaling issue and the ever-present editorial issues, were insurmountable. We couldn't get over those problems, but we could go round them. So "Talking Point" was born - a pre-moderated conversation forum which, at the time was unique. The pre-moderation meant we could keep the level of debate high but also that we could reduce duplication. The nature of the forum was that it wasn't available on every topic... but only on topics where we set up the debate. So we could control the scale.

Did we edit? Yes, we did. But our editorial guidance was, of course, never to change meaning. If a user wanted to make a point vociferously, and using bad language, then however good the point was, we would not edit it to remove specific words. If sentence 1 made a good point, but sentence 2 then went on to swear about it, then yes, we would remove sentence 2 and publish - but only if there was no-one else making the original point.

We did see early examples of citizen journalism or ugc or whateverwenowcallit - where professor So-and-so would put right a comment made in Know-it-all Correspondent's piece. And we did get good contacts which we could follow up on later.

As for scalability ... we were a team of 3 (me, Mick and Rebecca) and I believe Talking Point, which became "Have your say" grew to become a team of about 8. But even in a pre-moderated environment when running a maximum of 3 conversations at any one time, I still remember the inbox being overwhelming.

"I read a newspaper story headline online, and immediately know that the comment section is going to be hateful, without even having to read it. It jars with me that I know what is going to be “below the line”, even if I don’t cast my eye there."

This is the nub of it. The headline is the sub's precis of an article, which at best reflects the tone/subject matter accurately, but at worst takes its most eye-catching and contentious idea and uses it as a traffic-trap. That's fine if all you want is traffic (commenting or just reading) but it's rarely conducive to producing well-thought-out responses. Even if the better comments get there, they're buried beneath millions of bad comments.

How many articles and headlines are written specifically with the conversation below the line in mind? How many are written differently than they would be if they were in the printed paper? I have some sympathy with the journalist who writes a sensitive piece that is eviscerated by a troll-baiting headline, but there are plenty of articles that are deliberately contentious or not edited to remove unnecessary tangential barbs seemingly on the presumption that bad responses will require the reader to go and actually purchase green ink.

If you write, as I did, for a blog where the community discussion is far more important than the article above the line, and where you judge the article largely by the quality of the responses you get, then you get better responses. The only enemies of that position are (1) journalistic/editorial vice (namely laziness and/or a disregard for community conversations) and (2) the eternal drumbeat of "traffic uber alles". If a website's strategy is to get the maximum number of responses, and it tailors its content and headlines accordingly, it can't complain that high quality conversation is not ubiquitous.

There's no shortage of sites, and not-small sites, where the majority of the conversations are excellent-to-benign. What they have in common is a sensible idea before every article is commissioned and published, asking "what is the general conversation we hope will follow this, and how can we best tailor the timing/content/headline to make that happen?". If you try and combine that aim with "how can we get the most hits/biggest response?" you trade in quality of responses for traffic. If all you care about is the latter, I think you forfeit the right to complain about reader comments.

This is a 'reap what you sow' problem. You either see reader comments (which are becoming the majority of the words published on even the largest sites) as an editorial issue or a janitorial issue, and too many sites see it as the latter. Either through single-minded traffic-hunting (what we used to call trollbaiting) or through refusal to change modes of commissioning, editing and headline-writing to put reader responses at the centre of what is published, comments are bad because editorial decisions disregard them until after the comments have been made. That might well be valid if you know you're doing it to maximise traffic, but it diminishes the case for putting blame on "people on the internet being horrid". Too many sites are proving that rule isn't absolute.

Keep up to date on my new blog