The strange case of Google News and its “(blog)” label
Last month, Google News made a small and mysterious change to the way it displays some news sources: “We’re now visibly marking articles published on a news blog with a ‘(blog)’ label attached to the publication’s name,” Google explained, attributing the switch to user feedback. The label, a spokesman later told me, applies to any content “published through blogging software.”
This was all sorts of weird. On both technical and philosophical levels, there’s no meaningful difference between blogs that publish news and news sites that aren’t published as blogs. Many news organizations place material on both types of platforms without considering the content any different. Some use blogging software like WordPress to produce sites that look nothing like blogs.
Dividing content along these lines is like classifying brownies based on whether they were baked in aluminum or glass pans. There’s no difference, and it obscures what you really want know: if they contain chocolate chips.
Weirder still is the slipshod manner by which Google News has deployed the labels. There are dozens, if not hundreds, of inaccurately labeled news sources, and it’s inconsistent even within companies like Gawker Media: Deadspin is a blog, but Gizmodo isn’t. Also not labeled a blog, despite identifying as such: West Seattle Blog and Talking Points Memo’s Editors Blog.
The latter was actually labeled a blog until September 24, when Google switched TPM back to non-blog status, as you can see below. Asked to explain, Chris Gaither, the spokesman for Google, said the company doesn’t comment on individual publishers. TPM declined to comment as well.
The “(blog)” label is small and may not affect the user experience, but I actually think it’s pretty significant. Splitting up content by subject matter makes sense, but why would Google want to classify news by platform? The only other label it applies to content in Google News is “(satire).” [UPDATE, 3:15 p.m.: Gaither helpfully points out that Google News also labels press releases, subscription content, and videos, among other content.]
The company describes this new label as “highlighting the diversity of content in Google News.” It’s hard not to wonder if Google sees this as a gesture to newspaper publishers who have occasionally complained that their content isn’t privileged over blogs. (Of course, the “(blog)” label is supposed to apply to newspaper blogs, too. But that’s also wildly inconsistent: Blogs like The New York Times’ Media Decoder are sometimes labeled correctly and sometimes not.) Google has been unusually public of late in positioning itself as a friend of the news industry, and that could be the context here.
If so, and even if not, we should be worried. Google’s neutrality when indexing and displaying news sources is critical to allowing new systems of news to flourish. Already, Google News has run into trouble by occasionally excluding neighborhood news sites and popular blogs like Boing Boing.
The new “(blog)” label may not be about creating a hierarchy of content, but if there are no corresponding labels for “(newspaper article)” or “(front-of-the-book squib in a biweekly magazine),” and if they can’t even identify blogs correctly, why is Google even going there?
I asked Gaither, the spokesman for Google News, about all my concerns, and he provided this statement:
We added the “(blog)” label after receiving feedback from some Google News users who told us they’d like to know whether a listed story is a blog item before they click on it to visit the publisher’s website. Our goal is to give users as much information as possible so they can choose what kind of news content they’d like to read.
Zachary M. Seward | Oct. 6, 2009 | 9 a.m.
Tags: blogs, Chris Gaither, Deadspin, Gawker Media, Gizmodo, Google, Google News, Media Decoder, New York Times, Talking Points Memo, West Seattle Blog, WordPress
Yes, we’ve shown up as a blog at various times. We are a newspaper site and we don’t use blogging software ( for the articles labeled as blogs), so it does seem rather random.
Any possibility that this could be related to ongoing negotiations with the AP? It always seems like there’s something else to the story.
I’ve heard from some readers that they feel a “bait and switch” when they click on a link that takes them to a blog. Apparently Google has as well.
Yeah, Daniel, doesn’t it? I have no inside knowledge, but it’s hard not to read this in the context of Google’s negotiations with AP (such as they are) and their broader PR push with the news industry. —Zach
We’re using WordPress, and they aren’t labeling us. We publish in a specific niche, and we don’t have any other feed, just the WordPress blog.
I think the distinction may be not so much the technology, but the writing style. We do serious, not chatty news posts where the writer is not front and center. On many newspaper sites, the blog sections are emulating normal blogs, where the writing is personal and opinionated.
By the way, the blog ID has been there from years back, although they may have extended it to major media blogs just recently.
I think that its a good idea. There is a substantive difference between the majority of blogs and newspapers: one goes to blogs for opinions, and newspapers for facts. Even if this is not really true, or at least not always true, it is the popular perception. A blogger can read a news article and redigest it for public consumption. But they don’t produce the news, in most cases. They don’t do investigative reporting.
Okay. But what is a “blog”?
If the distinction that indicates a “real news” site is characterized by, as a commenter said, “serious, not chatty news posts”, then even a company as replete with advanced analysis as Google is seriously jumping the gun on what can be done with today’s computational linguistics offerings — to say nothing of scaling this computing to Google’s purview!
If Google really is responding to user feedback in doing this — and given their uncharacteristically half-assed approach at blog labeling here, I wouldn’t be quick to dismiss this conclusion — then what we observe with the (blog) labeling more reflects a societal confusion as to what information can be trusted as we continue down a path of shrinking traditional journalism and burgeoning informal blogging, with the lines between these items ever blurring.
I’ve abandoned trying to use Google’s “Search Blog” tool to find blogs about certain items because it brings up all sorts of not-blogs, including MSM sites. Their “advanced search” / “search by date” also frequently returns results outside the requested range. Maybe this is about SEO stuff gaming Google, or maybe its about inherent weaknesses. Either way, I wouldn’t expect Google News to perform any better.
Most problematic, I think, would be for web-pages appear further down the list of search returns because Google’s been trying to sort them based on type rather than keyword and content.
“… why would Google want to classify news by platform? …”
Editors writting in blogs are cheap information sources, which will probably compete against traditional news sources in the future. Integrating blogs will allow Google to become a publisher themselves as long as these blogs were not interrelated. So there is enough reason for Google to work on classifying the sources and their relevancy. – No doubt Google will proceed …