nofollow question

[avacado]

I read on another forum that if dmoz put nofollow on all of it's links, then the nofollow would not be included on all of the sites that use dmoz data.

 

I'm not interested in whether or not this will be or should be done, I am just curious as to whether this is true:

 

Would it be possible to have or require that the nofollow be included on any of the sites that use dmoz data?

[arubin]

It's not in our license agreement. I'm not sure it would be possible, even if it were in the license agreement, because of the possibility of generating intentionally defective NOFOLLOW tags.

[chaos127]

I don't see why it wouldn't be possible to create a data use licence that requires users to add a nofollow attribute whenever links appear on the web.

 

However, the current licence agreement does no do this, and I'm not sure what reasons there would be for AOL to consider changing it in that direction.

 

The data comes in the form a separate XML file which is then processed by downstream users, just adding nofollow attributes on dmoz.org wouldn't affect those data users. (It might affect people who grab the data directly from dmoz.org by screen-scraping the category pages, but it would depend on how they precessed the results, and that method is discouraged anyway.)

 

However, I really don't see why we would want to add nofollow attributes anyway -- either to our own links, or on those of data users.

[brmehlman]

The intended purpose of the nofollow attribute is to inform search spiders that a link isn't relevant. We happen to think our links are relevant.

 

A couple of unintended uses of nofollow also don't apply to us.

 

A page taking sides on an issue may well want to provide a link as a horrible example of their opponents' views but not give those views any search benefit. Nofollow serves their purpose, but we aren't supposed to take sides on any issue.

 

A page where the general public can post becomes an instant spam magnet. Nofollow prevents (or at least reduces) the spam. Our links are posted by our editors and (with a few mistakes which we try to correct) the spam doesn't get listed.

 

So, no, I don't think we have any reason to use the nofollow attribute.

[avacado]

I see.

 

Could the XML file be modfied so that all the links are automatically nofollowed by the downstream users?

 

Just curious I'm not saying this should be done.

[chaos127]

I doubt it: the parsing logic used by downstream users is likely to be to extract just the URLs, titles and descriptions from the XML file, and then to construct HTML code using that data. They'd likely ignore any other attributes that we decided to add (or worse such extra attributes would break their parsing).

[avacado]

Okay, I almost understand. So the XML file is just URLs, titles and descriptions, then the html code is constructed in some way?

 

How is it constructed? Because categories are bolded, and lists are bulleted...if that type of formatting can be done then it seems like other types of formatting could be done as well.

[brmehlman]

XML file is just URLs, titles and descriptions

Very close. We also supply information about the categories and their interrelationships but the main point is correct: It's just data, no presentation. Users of the data may present it any way they want, subject only to the provision that we be appropriately credited for it.

 

The bolding and bulleting and such are just our presentation, not present in our data dump, and nobody else need follow it.