[pmwiki-users] Google local site search

[Patrick R. Michaud]

On Thu, Dec 29, 2005 at 11:52:48AM -0700, H. Fox wrote:
> On 12/29/05, Patrick R. Michaud <pmichaud at pobox.com> wrote:
> > On Wed, Dec 28, 2005 at 03:24:27PM -0700, H. Fox wrote:
> > > So, for example, a page's Edit and History links become self-referring
> > > links, correct?
> >
> > Yes.
> 
> Then it seems like it will mislead the search index.
> 
> Every page will be indexed as a link that says "Edit" for instance. 
> Nearly every page on pmwiki.org will have eight extra self-referring
> links that lack rel='nofollow' attributes?

Ummm... search indexes normally index the contents of the page,
not the contents of links to the page.  Or am I mistaken here?

> (I still don't understand why the default skin doesn't use nofollow
> attributes in the PageAction links.)

Oh, we can probably do that -- I forgot that we can just add
rel='nofollow' to the wikistyle before each action not to be
followed.

> In that case, why cloak for googlebot?  Why not keep the ?action=
> parameters intact and use the rel='nofollow' attribute for bots that
> understand it?

Because there may be links where an author forgets the nofollow.
For example...

> > And even if the Skins include rel='nofollow' in the templates,
> > what about markup...?
> >
> >     [[OtherPage]]
> >     [[OtherPage?action=edit]]
> >     [[OtherPage?action=dc]]
> 
> It's a separate issue.  The bottom two should get rel='nofollow'
> attributes by default.

Er, wrong.  ?action=dc (like ?action=rss) should be followable 
by robots.  And I don't entirely agree that it's a separate issue,
since sidebars and other items may contain markup with edit links
or other actions.  Even if there are three links to ?action=edit
with rel='nofollow' and one that omits it, then the link is likely to
get followed.

> > So, the advantage of cloaking "?action=" is that it will work
> > even for robots that don't understand rel="nofollow".
> 
> Then it should be used for those, and not for robots that are known to
> understand rel='nofollow'.
> 
> I'm not entirely convinced it "works" for robots that understand
> rel='nofollow', since you're showing the robot a lot of misleading
> links (i.e. "transformed pages") for no reason.

Can you give me an example of this -- i.e.,  how the links are
misleading, and how it won't "work" for robots that understand 
rel='nofollow'?  

Also, can anyone tell me which robots implement rel='nofollow' such
that they don't follow the link?  I know that Google doesn't (since
July 2005), but what about the others?

Pm