Yahoo meddles with “no-content” class
I have had a few ideas in the past about how search engines should contribute to the spread of web development best practice in the past.
I have also changed my ideas on this a bit, after I had a few discussions with Google employees in the past. The point coming out of those conversations being that it shouldn’t be a search engines role to dictate accessibility or best practice web development across the web.
This role should be up to individual businesses and the web development community to decide what minimum level is acceptable and how to further educate and reach out to the masses on these topics. When it comes to Google, they are simply trying to index all of the pages on the internet as efficiently and thoroughly as possible.
Well, today Yahoo has started creating microformats on their own accord. This Yahoo blog posting explains how to block content from being indexed, not at an external or robots level, actually within your HTML markup.
Here are a few examples of how to apply this attribute for various uses and different syntax options:
<div class=”robots-nocontent”> This is the navigational menu of the site and is common on all pages. It contains many terms and keywords not related to this site</div>
Is it just me? Or is this just wrong?
Yahoo have now decided to meddle and create protocols on how this should be coded, with actual examples of how to mark up <div>s, <span>s and other block level elements that a search engine should not regard as important content.
Now of course, I believe that no right-minded developer is actually going to bother with this new tag. Besides Yahoo being just a tiny blip on the Google horizon, why would any developer choose to litter such markup within their HTML pages? With concepts such as microformats and formalising the practice of semantic HTML with POSH, why would I even think about adding extraneous inline class names to my markup for the sake of a minority search engine?
My concern here is the search marketers or SEO gurus out there, that don’t really know code, but get wind of this new tag and begin to start changing markup on all of their sites without knowing what they are actually doing. My other concern, although not a huge one as Google strikes me as being quite switched on, is that other search engines start supporting this practice.
My solution? An external, search engine only protocol of some sort. Why can’t you reference these elements externally, just as we do with everything else in modern web development?
div#navigation {rel=”no-content”}
Could something like this (mine is a non-working sample of course) not be incorporated into the ever growing development and support for robots.txt? This inline ’solution’ is just plain ignorant and is not contributing to the web, as Google’s support for accessibility could have done, but only watering down the tag soup that is already out there.
Time to rethink your strategy Yahoo, and this time think about the bigger picture. It’s not your role as a search engine to dictate how I code my websites, just concentrate on finding the damn thing.
Comments
- kwijibo says: May 3, 2007 @ 6:41 pm
Yep, that’s a weird one alright. Semantic mark-up by negation:
p.not-a-headerIt does seem pretty poorly thought out, and I’m surprised that Yahoo have seen a need for it. One might have thought that they had the computational power to be able to figure out which page sections are ‘boiler-plate’ and which are unique content.
I suppose what it all comes down to is whether or not it helps them return a better quality of search results. If they start ranking less relevant pages that have used .no-content higher than more relevant pages that haven’t, then the whole scheme is no longer in their interests. However if it genuinely does solve a problem faced by search engines, Google et al might adopt it too, and force us all down the path of markup by negation.
I think you’re right; if this is something that really would help search engines to index our content better, it ought to be done in the robots.txt file, using CSS or xpath selectors
nb: there is a µf draft that does the same thing: http://microformats.org/wiki/robots-exclusion
I don’t like the idea of that either.
- Ciaran McNulty says: May 3, 2007 @ 7:23 pm
I agree, a preferred solution for me would be to be marking up sections of the page as ‘navigation’, ‘header’ and so on, and have the search engine work out what is and isn’t important.
On a sidenote, there’s a bit of a problem at the moment with people using the phrase ‘microformats’ in a sense that doesn’t refer to the microformats.org community. We’d prefer it if people didn’t use the phrase as a generic term for semantic HTML if they can avoid it. Thanks!
- Sebastian says: May 3, 2007 @ 9:40 pm
I second that, see my yesterdays rant at http://sebastianx.blogspot.com/2007/05/yahoo-search-going-to-torture.html
Not that surprisingly both of us came to the solution that this thingy is for robots.txt, not the markup. Yahoo obviously launched a not well thought out “tag”. Just boycott it.
- Standardzilla says: May 4, 2007 @ 11:18 am
@kwibjibo - That’s just what scares me about this stuff… the actual potential for Google to also pick up on it and watch it snowball out of control. It would be terrible after all the work done in the past by the passionate web community.
@Ciaran - the “no-content” class was fairly similar to the elemental microformat rel=”no-follow”. I realise it’s not exact but I wanted to make the point that they are negating any kind standards being set by groups like Microformats and web standards communities. It’s like a proprietry microformat in a sense.
@Sebastion - boycott it I will :-)
- Adam Messinger says: May 13, 2007 @ 10:40 am
Not only is this a stupid idea to begin with, I find their example of using it on navigation troubling. 98% of the people making websites have no clue about XML sitemaps and similar tools. If search engines ignore navigation, how will sites get thoroughly indexed?
