From: Ian Hickson (ian@hixie.ch)
Date: Thu Sep 19 2002 - 17:30:48 BST
On Wed, 18 Sep 2002, Paul Freeman wrote:
> Ian wrote:
>> I would consider pingback spam be any of the following:
>>
>> a pingback from a page which doesn't include the appropriate link
>>
>> a pingback which has already been sent (a duplicate)
>>
>> a pingback from an automatically generated page (i.e. one with no
>> further comments)
>
> Whilst I can't see a way of filtering the last one, the first two
> could be tackled on the server-side. I say "could" but would I really
> be a good idea to parse the referencing url to check the link is
> actually there?
The spec recommends it, and at least two implementations do it. There's
even a fault code for that. :-)
> Duplicates could be checked for though.
There's a fault code for that too. My implementation used to internally
crash when it got a duplicate. That's fixed now. ;-)
>> Why would anyone ever _remove_ a link? That seems very dubious to me. What
>> if other people were linking to that post because of the link?
>
> Unless you pingback the wrong url by mistake?
Fair point. Is that rare enough to not require a pingback method for it?
Because a method for _removing_ pingbacks is going to be a whole lot more
complicated to pull off, because it will require authentication and all
that. Not exactly trivial.
I think a better solution would be for people to report incorrect
pingbacks by e-mail and have the other person manually remove it...
-- Ian Hickson )\._.,--....,'``. fL "meow" /, _.. \ _\ ;`._ ,. http://index.hixie.ch/ `._.-(,_..'--(,_..'`-.;.' Message sent over the Blogite mailing list. Archives: http://www.aquarionics.com/misc/archives/blogite/ Instructions: http://www.aquarionics.com/misc/blogite/
This archive was generated by hypermail 2.1.5 : Thu Sep 19 2002 - 18:05:00 BST