[Guest post by DRJ]
Note: An earlier post on this subject is here.
a one-post now a two-post blog called Investigate the Media claimed that the San Francisco Chronicle “activated a devious system by which it deceives commenters on its website, SFGate.com.” It also claimed commenters (apparently referring to conservative commenters) are “silenced for ideological reasons” or “because they strike too close to home — pointing out flaws in the article’s reporting or writing itself, or ethical or moral misdeeds on the part of the Chronicle editors or management.”
These are serious claims and, to the Chronicle’s credit, an online editor for the Chronicle responded promptly. Scroll down and click *More* for the full text of the response from Eve Batey, Deputy Manager-Online for the San Francisco Chronicle. (Ms. Batey was also interviewed here.) My summary of the main points from Ms. Batey’s response are set forth in the indented and bolded sections below, with additional thoughts following each summary:
The Chronicle’s online editors do not moderate every comment. Instead, they rely on commenters to report abusive comments.
It makes sense that the editors can’t moderate every comment – the website would lose its interactive quality if that happened – but this may be part of the problem. San Franciscans voted 85% for Kerry and 15% for Bush in the 2004 election. If most SFGate users are liberal, there is a greater likelihood that the users will object to conservative comments and flag them as offensive.
For instance, if there are 8 liberal commenters for every 2 conservative commenters, theoretically there are 4 times as many chances that someone will flag the conservative comment as offensive. Of course, the liberal commenters may flag other liberal comments or the 2 conservative commenters may flag all 8 liberal comments as offensive, but the odds don’t support either option happening on a regular basis.
Thus, the pool of flagged comments the moderators actually look at are more likely to be conservative and, as a result, conservative comments are more likely to be subject to deletion. Think of it as the online equivalent of profiling.
The online editors do not discourage diverse views and do not want to, either from a philosophical or from a business perspective.
I agree it doesn’t make sense to eliminate controversy if you’re in the newspaper business. In fact, the editors probably want diverse views because conflict drives up interest. However, even though it’s good business to encourage controversy, there is a flip side. The Chronicle may not want to irritate its mostly-liberal readers with too many conservative comments. However, until we can see the deleted comments, I accept the Chronicle’s statement that it handles comments in a fair manner.
The commenter who blogs at Investigate the Media was banned and/or had comments deleted because he “ran afoul” of the Chronicle’s Terms and Conditions.
To follow up on this, we requested copies of the deleted comments from the Chronicle and the blogger who posted on this at Investigate the Media. The blogger has not responded at his/her website. And despite the SFGate website’s Terms and Conditions that grant exclusive ownership of the contents to the Chronicle, the Chronicle has refused to provide copies of the deleted comments without permission from the commenter. However, Ms. Batey has reached out to this blogger at his/her website and asked permission to share the comments. The blogger maintains the “Chronicle admits to deceptive comment-deletion policy, offers bizarre excuse, then lies again.”
So what do we know? The Chronicle acknowledges that the SFGate commenter (jimjams) who posted Investigate the Media is a registered commenter. At his/her blog, jimjams provided one example where a comment was deleted by SFGate.
That comment was apparently deleted because jimjams called the Chronicle staff writer an “idiot.” Does every commenter who calls someone an “idiot” (or similar terms) get banned at the Chronicle or does this only apply to comments about a Chronicle staff writer? Was the deletion related to the critical tone of the comment or to a pattern of inappropriate comments?
It’s hard to tell without access to the deleted comments. I hope the parties will provide copies of the comments … and, in the meantime, I have to wonder why they won’t.
Update 11/30/2007 @ 5:20 PM PST: Ms. Batey corrects my statement regarding the single comment posted by jimjams: “[T]he comment you see on ITM referring to a Chronicle reporter as an idiot wasn’t a deleted comment. Jimjams had actually been blocked for some time before that day, and the “idiot” comment was one of his/her non-Terms violating comments that was also hidden from view by the use of the tool. That, right there, is why using the “block user” function was the wrong decision.” Thanks for bringing this correction to my attention and I’ve noted the correction by striking through two sentences, above.
The Chronicle requested but the software vendor was unable to provide a feature that specifically noted when and why comments were deleted.
Ms. Batey provided a lengthy explanation of the technology issues, and I encourage you to click *More* and read her complete response. In short, the Chronicle previously used a “block user” function that “blocks all comments made by a user from view by anyone but themselves (upon login), and replaces those comments with a deletion message, in a way that simply deleting a single comment would not.” She asserts that this function has been used sparingly and only with commenters who, in the moderators’ view, repeatedly violated the rules. The “block user” function was chosen in an effort to promote transparency because it gave some indication that a comment had been deleted or blocked. In practice, however, this function did not seem transparent. Ms. Batey apologized for that and noted the Chronicle editors have stopped using the “block user” function.
I appreciate Ms. Batey’s discussion of things that work and don’t work in moderating comments. There is a learning curve in blogging – at least there has been for me – and I’m impressed with her willingness to acknowledge a lesson learned and to make changes as a result.
Click *More* for the full text of the email from Eve Batey, the San Francisco Chronicle’s Deputy Managing Editor – Online.
Thanks so much for giving me the opportunity to respond. Out of all the bloggers who posted on this I reached out to, you’re the only one who has responded. I really appreciate that.
I want to make sure you have a complete understanding of how The Chronicle handles and moderates article comments, so this email is going to be pretty long. If you choose to do a follow-up post, please do feel free to excerpt or fairly paraphrase me, as I know that this explanation is lengthy.
When a reader registers to comment on SFGate, they agree to the Terms and Conditions of the site:
Knowing full well that this was a pretty dense read, we also put together a Comment Policy
Both of these policies are linked to right by the comment box, so readers can refresh their memories on what they can and cannot post on our site pretty easily.
We don’t have folks moderating the comment sections of every article — that would be impossible. Instead, we have a “report abuse” button on every comment. If any reader sees a comment they consider to be a violation of the TOC or Comment Policy, they’re encouraged to hit that button. At that point, the comment is flagged for (brought to the attention of) a set or Chronicle editors, one of whom will look at the comment, evaluate it for compliance with the TOC and Comment Policy, and either delete it or allow it to remain. It’s not always easy to decide, so we try to err on the side of caution and delete very few comments, overall.
Given the questions raised recently, I’ll spell it out for you: neither The Chronicle nor SFGate would delete a comment based on an ideological stance — if you don’t believe me, take a look at any of our articles, and you’ll see that viewpoints of all stripes are expressed there. Seriously, why would we bother to have comments on the site if we didn’t want varying viewpoints to be expressed? And how successful would any comments section be if the only comments that were there were in agreement with any agenda? Even if, philosophically, we didn’t believe that squelching any TOC-abiding viewpoint was wrong, businesswise it would be really stupid to kill every conversation by only allowing one viewpoint.
It’s important to me to respect the privacy of the commenter who created the Investigate The Media blog to raise this issue, so I won’t go into why this user ran afoul of our Terms and Conditions. But I will say that we delete comments that are brought to our attention and do things like use racial slurs or threaten or suggest violence against others. We would never, ever delete a comment just because we don’t agree with what someone says.
The software we use for article comments isn’t an SFGate creation — it’s provided by an outside company with which SFGate has contracted. When we gave them our requirements for article comments, we made it very clear that we really, really needed a way to indicate that a comment had been deleted — something as simple as having the text of the comment replaced by “This comment has been deleted due to violations of SFGate’s Terms and Conditions” would have done the trick nicely.
However, this wasn’t something the company was able to provide to us immediately. We at the Chronicle and the folks at SFGate weighed this problem, and decided not to let this keep us from moving forward on article comments. This nagged at me and at my colleagues, that deleted comments would just “be disappeared,” but we felt such a sense of urgency to add article comments to the site that we pushed this worry away, and hoped that this transparent deletion function would be added soon.
Unfortunately, the commenting company hasn’t been able to provide us with this tool yet, and suggested that we use their “block user” function as a stopgap measure. (This “block user” function is what you and your colleagues in the blogosphere have called us out about.) The “block user” function blocks all comments made by a user from view by anyone but themselves (upon login), and replaces those comments with a deletion message, in a way that simply deleting a single comment would not.
So, what we’ve been doing is deleting TOC-violating comments from folks who only occasionally violate our policy, but in cases where mass disappearances of comments would make the article comment conversation completely incomprehensible, we opted to use the “block user” function, so it’s clear to users that comments have, indeed, been removed. This function has been used very, very sparingly (and only a few of us have access to this function), and only for those few folks who have repeatedly violated the TOC.
Clearly, however, even though this only has impact on a few users, it was the wrong thing to do — and that, in our eagerness to have discussion and conversation on the site, we failed to take into consideration those users who would feel hurt and deceived by having their comments blocked from view. In addition, while the use of the “block user” function was actually expected to provide some small bit of transparency to our comment management policies, by illustrating that comments had, indeed, been removed, it has actually done the opposite. For this, I apologize.
I’m glad that this issue has been raised because I think that this will help make our commenting software providers understand the importance of having a function that makes it clear that comments have been deleted. We’ve stopped using the “block user” function as of yesterday, even at the risk of having comments “disappear” and at having some article comment section conversations suffer as a result.
DRJ, I hope that this addresses your questions and concerns. If I can be of any more help, now or in the future, please don’t hesitate to contact me.
Deputy Managing Editor — Online
San Francisco Chronicle