Should There Be a Red Flag Link?
posted by Frank Pasquale
Links are the lifeblood of the web. You can almost think of a link as a light shining on a page, illuminating territory that might otherwise be hidden in a cloud of information overload. As Ray Cha has explained, “Google ranks pages by the number of links other sites point to a page.” So whenever someone writing online links to a page, they increase its prominence relative to other pages.
But what if you want to comment on something you disagree with? Or find utterly inane? If you link to it, you just increase its salience. If the site tracks back to you, you might be able to alert readers to your critique. But if it doesn’t, you just end up promoting the site even as you try to fight or mock it. Cha notes that there are some proposals to change this situation:
There have been suggestions to create a newer kind of syntax and link taxonomy which would add to the current binary options of link or no link. The simplest system would be to have three choices, positive link, negative link and no link. This system would actually be very easy for users. All you need to do is add a tag to the link.
This led me to think about the red and yellow flags on Westlaw that come up when a case is no longer good authority, or has been contradicted. I often think about the tension between accuracy and usefulness in signals like these. Part of me wants more gradations of meaning–I have definitely seen some “yellow flagged” cases that were far more questionable than other “yellow flagged” ones. But if we went to some “rainbow scheme” of vitality of authority, the system would likely get unwieldly. Even discussion forums that permit negative or positive rep points (or karma) tend to keep things very simple.
However “tiered linking” might emerge, it’s important to note how vital search engines will be to the process. Even if a platform like Blogger or MoveableType puts in multiple flavors of links, they’ll do little to alter web discourse if search engines don’t recognize them. Given the “chicken and egg” problem here, it’s likely that tiered linking is going to remain the province of limited and proprietary databases for the time being.
Photo Credit: Jen Waller.
December 11, 2007 at 12:54 pm
Posted in: Cyberlaw
Print This Post
Responses (9)
Logical Extremes - December 11, 2007 at 3:07 pm
I wouldn’t be surprised if Google and others adopt something like this. It’s not dissimilar to what Digg and its ilk attempt to do. But even a negative ranking will draw attention - like a car crash or a shock meme.
greglas - December 11, 2007 at 4:30 pm
Frank,
There is something like this. See:
http://en.wikipedia.org/wiki/Nofollow
Jon Garfunkel - December 11, 2007 at 7:23 pm
From the wikipedia page for nofollow, it was introduced in 2005, and many blog tools quickly adopted it.
View Source on this very web page. You’ll see that this blog is already using the nofollow tag!
The main aim of nofollow was to accomodate user comments, which were increasingly seeing spam in an attempt to boost SEO. But the nofollow simply works against ordinary users as well: If I link to articles I’ve written in the comments here, I am unfairly boosting my own website’s SEO.
(Conversely: if you as a blogger/publisher find something interesting in the comments, and you want to boost its Google rank, you should write a new post to truly signify endorsement. See NetProps for some thoughts about linking dynamics.
Bear in mind that del.icio.us was explicitly designed a shared intelligence service, and yet, when I last asked him, Josh Schachter was uninterested in why a “neg” bookmark would be useful. (While digg is has only pos/neg, and no other adjectives.)
Grant that HTML editing tools don’t all make it apparent to the user. If you want to hack a nofollow (in such a way to make it explicit to the user that you are not giving any PageRank), you can do so by constructing a link like this (which goes to a page explaining the “Google Redirect Notice”.)
Also, I might check with the SEO community whether Google uses URL references from non-HTML sources in PageRank calculations. Maybe journals should reconsider having PDF-only output! … but then again, if a journal article is referencing a URL, it’s probably well known.
Andrew - December 11, 2007 at 10:21 pm
Not sure this is necessary. I take the point that sometimes people link because they disagree with a post and are arguing against it. (And there are already 2 defences here: the link’s context (for humans) and nofollow (for bots).)
But if I don’t like a site, as a rule I won’t link to it. There’s a virtually unlimited amount of stuff out there on the web; I’ll read (and link to) the bits of it I like. This is (or course) essentially why Google works and is tied to the point you made earlier:
This may be the difference between Digg and del.icio.us. Items in Digg are added to a linear list - there is no context or means to weight their relative importance initially, so voting buttons must be supplied. In contrast, a del.icio.us link forms part of a web; the more links it has, the greater its value.
Of course this argument doesn’t hold in cases where the amount of available information is limited (such as Westlaw?), and it isn’t always possible to substitute a bad source for a better one.
Seem remotely plausible?
PS When SEOmoz discussed bad practices a while ago, they linked to an offending site thus: bad site (link condom applied). Made me smile.
Drew - December 12, 2007 at 8:48 am
Do we really need a technological way to register our dissent with something? Isn’t just disagreeing with it by arguing against it in writing enough? I fear also that this could lead to blacklisting: hey, let’s gather a lot of red flag links to National Review! That’ll show ‘em how much we disapprove of their message.
Also, if you cared about something enough to link to it and write about it, then it deserves its popularity. People can decide for themselves whether they like it or not, without technologically-imposed opinion enforcement. This strikes me as similar to building copyright into music via DRM. It’s opinion DRM.
Jon Garfunkel - December 12, 2007 at 6:16 pm
Drew,
You missed the premise here. Google’s PageRank algorithm assumes that each hyperlink is a positive vote. But these are vastly different purposes. Web writers had reason to link (enabling the reader to go directly to the resource) long before PageRank was invented.
It’s as if you have made the decision to reside in a particular city, and a local planner were to include the statistic of your residency (and many others) to justify taking public dollars for a baseball stadium.
The problem that Frank wishes to solve has partially been solved through the nofollow tag.
Jon
Drew - December 13, 2007 at 10:06 am
Jon - I get the premise, I just don’t think it’s necessary. PageRank doesn’t care about “positive” or “negative” votes, it cares about links. In my view, if something is worth being discussed negatively, it’s worth whatever page ranking results.
Eric Turkewitz - December 13, 2007 at 10:38 am
I believe that you can use a redirect site such as TinyUrl to mask the offending site that you want to link so, thereby both linking to them and not giving them any Google pagerank. I did that today in this post:
Personal Injury Lawyer, Ryan Bradley, Using Blog for Blatant Solicitation
Jon Garfunkel - December 13, 2007 at 7:10 pm
Drew– Again, you are mistaken. PageRank *does* distinguish between nofollow and regular links. Thanks to nofollow, publishers are aim to control whether user-posted links affect PageRank.
Negative links are a different manner. They probably will not be recognized by PageRank anytime soon. But a dissenting link could be valuable for conversation aggregators.
Leave a Reply