A story in The Boston Globe reports that a caustic online commenter appears to be posting “vicious Internet attacks” against Mayor Michelle Wu and others from a city email address and a shared subscription used by city councilors and staff members. The comments have been published under the handle “Interested Party.”
Globe reporter Emma Platoff writes that “the comments posted by the Interested Party account stand out because they appear to be authored by one or more of those public officials’ colleagues — members of the City Council or council employees — according to account details and people with knowledge of the subscription.”
It’s not the first time I’ve seen a news outlet do this. The one that stands out in my mind involved a Haverhill city councilor who was exposed years ago by the local newspaper. I’ll get to that. But, first, more on the Globe’s story.
Although the Globe doesn’t have enough information to identify the person (or persons) behind the comments, Platoff appears to have used internal information provided by every user (“account details”) in order to get as far as she did. If you look at comments on the Globe’s website (or just about any other website), you won’t find any email addresses — that’s information you provide when you sign up, but it doesn’t appear alongside your screen name. Likewise, the Globe almost certainly has access to each commenter’s IP (internet protocol) address, which can help the paper locate where a user is based.
Nor does the Globe warn users that it reserves the right to use information commenters provide in order to track them down. If you try to post a comment, you’ll find guidelines that speak to what’s allowed and what isn’t; but there’s nothing about the possibility of being outed. (The guidelines refer not to the Globe but to its free sister site, Boston.com. I’m not sure whether that means the rules haven’t been updated for years or if it’s simply an indication that the same rules are in effect at both places.) Another message advises you, “This comment may appear on your public profile.” But when you click on the “Public Profile FAQ,” you hit a 404.
The question is whether it’s ethical to use information that the Globe’s subscribers freely give the paper in order to track them down. Users have a right to expect that their information won’t be used to violate their privacy. Platoff wasn’t able to expose “Interested Party,” but she’d clearly like to. Is that fair?
All of that brings me to the Haverhill situation I mentioned up top. In 2008, The Eagle-Tribune of North Andover reported that a Haverhill city councilor named James Donahue had posted comments on the paper’s website under at least 38 different screen names, an accusation that Donahue partly confirmed, though he claimed that some of the comments had actually been posted by someone else. As I wrote at the time, Donahue’s activities included lambasting the mayor and some of his colleagues. The Tribune apparently used the IP address attached to Donahue’s multitudinous screen names in order to locate him and to figure out that all those screen names were coming from the same computer. The Tribune defended its actions in an editorial:
It is not the general practice of this newspaper to seek the identity of those who comment on stories, although there is no explicit guarantee of anonymity. Virtually all the management of the comments section of the online edition is aimed at removing posts that are profane, racist or personal attacks.
However, one of the forum moderators noticed a pattern of posts under dozens of different names, and then discovered that they had all come from the same computer address. When it became clear that they were coming from the computer of an elected public official, it became our obligation to let the public know.
The average citizen does not take an oath to serve the public. An elected official does. An attempt to deceive the public is clearly not serving it, and a public official who does so is not only undeserving of the protection of confidentiality, but deserves public criticism.
I was uncomfortable with the Tribune’s actions, and I said so, writing: “It seems to me that the paper has chosen to humiliate Donahue for doing something the paper itself implicitly invited him to do, and that it used information available to no one else.” But many of my commenters disagreed with me and sided with the Tribune, arguing — as the paper itself had — that Donahue’s status as a public official overrode any expectations of online privacy that he might have had.
Which brings me back to the Globe and “Interested Party.” Arguably, there is a public-interest reason to try to expose the commenter. The comments are being posted by a public official or officials, whether they are members of the city council or employees. It’s a story, and it will be interesting to see whether the Globe is able to take the next step and name names. It makes me a little queasy, as The Eagle-Tribune’s actions did 15 years ago. On balance, though, I think my commenters were right in 2008 and that the Globe is right now.
A final observation: The Globe’s comments are a toxic-waste pit. The paper shouldn’t have them unless it’s willing to screen all of them before they’re posted. If that’s impossible, then get rid of them. Plenty of news organizations have, and no one seems to miss them.
15 thoughts on “Should the Globe have used private data to try to expose a toxic commenter?”
I agree that this is a very difficult issue. We have public officials who post under fake names on our site in New Haven. When I have noticed that they are using multiple names, I have asked them to stop doing that (to avoid making it seem like multiple people are sharing that point of view for political purposes), and they have complied. And we have rejected posts where they praised themselves under a fake name. (They really try to do that!)
But we do let them post under a single consistent anonymous handle. I think that might be wrong. They come out against their opponents, more strongly that they would if they were using their real name. But lke Dan, I’m hesitant about stepping over that line, and would not feel comfortable reporting on something they post anonymously as having come from them. It feels like an implicit promise of anonymity. We do work hard to vet those comments before the appear so that they don’t violate our guidelines of civility. However, I’m not sure what the right answer is.
In my city, campaigning is not allowed on city property and especially not City Hall. Is it not the same in Boston? So, if someone is using a City Hall-based IP address to malign an elected official, that should be a violation since it is on city property. But if a public official used comments to try to notify other commenters of a city service, for example, that would be okay. (And of course, regardless of someone’s public service day-job, they could do whatever they want from their personal internet accounts.)
So, I wouldn’t block City Hall as a whole from commenting. And, I wouldn’t kill comments, though I’d do my best to try to maintain a “Real names only” rule.
That leaves one option – the Globe can either tell or not tell City Hall someone is violating the rule to whomever’s in charge of giving .. internet access at city hall? The legal department? Somewhere… Yeah, I’m thinking out loud, here. It’s still the Globe’s choice.
I just want to amend my post to address privacy concerns. It does seem like the Globe should update their user agreement for commenting. In the fine print no one reads, it should say that comments submitted to the Globe may be shared with 3rd parties. If anyone wants to be an anonymous source, they should contact a reporter and not use the public comments under an article.
“Real names, first and last, are required.” – That there would solve a lot of problems. IMO, the Globe should have the same id requirements for comments as they do for Letters to the Editor. That said, if you leave a comment on a website, you really should have no presumption of privacy.
And (shhh) I’ve stopped enforcing that rule. I let it go as long as the comment meets basic civility requirements.
I always worry that people who really want to game the system can create fake identities online and get approval with what seem like real names. And some people still try to publish god-awful stuff under their real names. I think it’s the end behavior we are concerned with: what kind of toxic comments result from anonymity. Our approach, which I do not know is ideal, is to look at the end result: if it’s toxic, or even outside civility, zap it. Whether or not it has a real name.
Disagree; often I learn far more about a subject from reading the comments, which include links to other articles and supporting facts, which are missing from underreported stories.
I posted the link to Dan’s post to Arlington Patch Neighbors. One respondent calls him “naive,” saying that one cannot connect an IP address to a name without a court order. Is that true?
I wrote that an IP address “can help the paper locate where a user is based.” There are IP location services online. Plug in the address, and you can find out roughly where the computer is located. A commenter here posted something recently that was antisemitic and anti-Catholic. I used his IP address to find out where he lives, and from there I found his LinkedIn account, so I know where he works. I hope he’s reading this, ha ha.
Thanks for your response, Dan.
The IP address, depending on your ISP, could be more or less accurate as an identifier. For years, everyone on AOL was identified as being in Ashburn, VA (a spot that still produces a disproportionate amount of web traffic). There are, of course many techniques one can use to hide one’s IP address (the most easily-accessible one being a VPN).
Interesting post. I’m on vaca, so I can read more than skim some recent posts online.
These stories have mystified me since there are so many other things to cover. There is also the wonder of motivation of trying to expose online commenters instead of just letting them do what they do. Why was the Globe even looking at the commenter’s email or IP address? Curiosity, sure. Desire to know, too? Maybe this person who is sharing inside stuff could be a source! Heh, heh, heh … But doing a deep dive to tout them, due to who they are being critical of? Is that “journalism” or something else?
When online commenting was like the “wild west” and quite active instead of silent, I had a situation in Belmont when I was the editor of the Citizen-Herald, where liberal readers were upset conservative readers were attacking their desire to raise property taxes for the school system. Many liberal readers used their own names and emails (although some didn’t). Most of the people against higher taxes did not use their own names and emails — because they didn’t want to be targeted by their neighbors for having opinions that some felt were “against children.” There was often political and personal retribution for challenging school budgets in town due to its changing demographics, something more complicated to get into than a blog post reply. The WickedLocalBelmont.com site was the first time readers could vent about their high taxes.
As editor, I kept track of the IP addresses and emails of the more active users. Racist comments or ones with swear words were deleted. But people were able to say and feel what they liked. Many readers on both sides often raised relevant arguments. It made the site lively and worth watching and reading.
This was also the first time that I started seeing something that has become quite common today: One side of the argument calling for censorship of the other, in this case, liberal readers, who wanted all commentary challenging their desired outcomes, higher property taxes, to be silenced, whether the commenters were anonymous or not.
But I gave everyone anonymity, even those dumb enough to use their email addresses. Why? First, engagement was good for business — it kept readers interacting with each other on the site. Second, it built up the WLB site into the most read sites in the company even though it was this tiny town, something regularly highlighted in corporate memos. How can the dinky Belmont site draw interactivity higher than some of the more significant suburban news sites? I let people say what they wanted and what they felt. Simple as that.
Why not just let people, whether employees or not, say what they think and feel?
I can’t edit the comment but that should be “out” not “tout,” LOL.
* yes, comment sections on public or social websites are often trash. read at your own risk.
* while it is certainly possible to obscure your identity on the internet, i personally never post with the expectation of anonymity. clearly, this is not a universally used approach.
* if they have reasonable evidence that an elected or public official is using work resources to comment, then i feel that falls in a different category than posting from their own email/computer.
Thanks for writing about this story. I have to admit I was surprised by it. The cesspit of the comments section seems like a bit of a honeypot now, too. But exposing one toxic commenter won’t fix the problem of the toxic environment news comment sections seem to create. I wonder what their cost/benefit analysis is like. Do readers stay for the comments? Maybe the trainwreck below the break is part of the reason some subscribers just can’t quit.
Comments are closed.