W3C home > Mailing lists > Public > www-html@w3.org > January 2005

Re: rel="nofollow" attribute (PR#7676)

From: Beth Epperson <beppers2@cox.net>
Date: Fri, 21 Jan 2005 16:34:48 +0000
Message-ID: <41F126E0.8030302@cox.net>
To: Mark Birbeck <mark.birbeck@x-port.net>
Cc: w3c-html-wg@w3.org, www-html@w3.org




Hi Mark,

Again, I have to disagree - what may seem daft to you, may be brilliant 
to someone else, and if this option resolves a problem, then it isn't 
daft at all.

If you have a more elegant solution for them, then clearly define it, 
and provide examples.

//beth

Mark Birbeck wrote:

>Hi Beth,
>
>[Note that I'm cross-posting to the public list since there is quite a large
>thread on this there.]
>
>If something is an advert, you only need to say it's an advert -- you don't
>need to say "don't follow this link", since whether the user follows the
>link is up to them.
>
>And whether Google follows the link is also up to them -- they could choose
>not to follow adverts, for example. But I don't see why we should 'dumb
>down' metadata so that Google can work out what to index. If the problem
>really is about page rankings, then why not go the other way, and give more
>weight to those links that have an explicit @rel value?
>
>Anyway, there are plenty of solutions better than "nofollow", and ultimately
>it's just daft to put a qualification on a link (which is what @rel is) that
>says the type of this connection is that there is no connection.
>
>Regards,
>
>Mark
>
>
>Mark Birbeck
>CEO
>x-port.net Ltd.
>
>e: Mark.Birbeck@x-port.net
>t: +44 (0) 20 7689 9232
>w: http://www.formsPlayer.com/
>b: http://internet-apps.blogspot.com/
>
>Download our XForms processor from
>http://www.formsPlayer.com/
>
>  
>
>>-----Original Message-----
>>From: Beth Epperson [mailto:beppers2@cox.net] 
>>Sent: 21 January 2005 14:30
>>To: Mark Birbeck
>>Cc: dmh@dmh.org.uk; xhtml2-issues@mn.aptest.com; w3c-html-wg@w3.org
>>Subject: Re: rel="nofollow" attribute (PR#7676)
>>
>>
>>Hello,
>>
>>I disagree Mark, saying we are not related is a valid option. 
>>They may 
>>be connected via a link for a variety of reasons - i.e. an 
>>advert, but 
>>following that link would clearly not enhance the users 
>>knowledge of the 
>>object at hand. For example, many hosting sites force you to place a 
>>link on the bottom of the page directing your audience to 
>>read about the 
>>hosting company. If I am posting data about biotechnology and 
>>studies on 
>>MS, the link to ABC Web Company" is an illogical relationship - so to 
>>say "don't follow that one" is correct.
>>
>>//beth
>>
>>Mark Birbeck wrote:
>>
>>    
>>
>>>Dave,
>>>
>>> 
>>>
>>>      
>>>
>>>>Google and others are introducing a 'rel' attribute value
>>>>that refers to 
>>>>a document that shouldn't be indexed by search engines.  (See 
>>>>http://www.google.com/googleblog/2005/01/preventing-comment-
>>>>        
>>>>
>>spam.html 
>>    
>>
>>>>for more information on why they're doing this.)
>>>>
>>>>Should this, or some alternative mechanism that achieves the
>>>>same ends, 
>>>>be included in XHTML 2.0?
>>>>   
>>>>
>>>>        
>>>>
>>>Speaking for myself, I'd say no, for two reasons:
>>>
>>>* @rel implies a relationship between two documents. It's a bit
>>>  weird to say that the relationship between two documents is that
>>>  a search engine should ignore the relationship between two
>>>  documents.
>>>
>>>* whether you like it or not, there *is* a relationship between the
>>>  two documents -- there's a link.
>>>
>>>Google wants this feature so that they can tell the 
>>>      
>>>
>>difference between 
>>    
>>
>>>real links and those placed there to improve someone's rating in the 
>>>search engine. It's a commendable goal, but it's got nothing 
>>>      
>>>
>>to do with 
>>    
>>
>>>HTML. Indexing should really be transparent. Will the next 
>>>      
>>>
>>move be to 
>>    
>>
>>>say which paragraphs of text should be indexed!
>>>
>>>So, to achieve this goal I'd prefer to see something 
>>>      
>>>
>>'positive' rather 
>>    
>>
>>>than something 'negative'. By that I mean, some mark-up that 
>>>      
>>>
>>indicates 
>>    
>>
>>>that the link is for UI only, rather than this link is not to be 
>>>indexed. Another alternative would be to state the type of the 
>>>document, and for Google to work it out for itself whether links in 
>>>that document should be followed or not.
>>>
>>>For the first solution, the new @role attribute would probably be a 
>>>good place to put it:
>>>
>>> <a href="http://pills-and-potions.com/" role="ui-only">spam</a>
>>>
>>>and for the second solution, the new QName @rel values could work:
>>>
>>> <head>
>>>   <title>A comment</title>
>>>   <link rel="rdf:type" 
>>>      
>>>
>>href="http://www.google.com/blog-comment" />  
>>    
>>
>>></head>
>>>
>>>I'm not saying either is the answer -- but I certainly don't like 
>>>"nofollow"!
>>>
>>>Regards,
>>>
>>>Mark
>>>
>>>
>>>Mark Birbeck
>>>CEO
>>>x-port.net Ltd.
>>>
>>>e: Mark.Birbeck@x-port.net
>>>t: +44 (0) 20 7689 9232
>>>w: http://www.formsPlayer.com/
>>>b: http://internet-apps.blogspot.com/
>>>
>>>Download our XForms processor from
>>>http://www.formsPlayer.com/
>>>
>>>
>>>
>>>
>>>
>>>
>>> 
>>>
>>>      
>>>
>>
>>    
>>
>
>
>
>
>
>  
>
Received on Friday, 21 January 2005 18:50:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:01 GMT