Pavatar Support Forum

Hi
I have just read your Pavatar specification, and like the idea. However, I do not like how you have specified the use of the link element in HTML / XHTML to allow discovery of Pavatars.

In particular, not allowing any other attributes on the link element (e.g. type) and not allowing the attributes (rel and href) to be given in a different order seems an unnecessary restriction that goes against the spirit of XML. I realise that this makes parsing slightly easier, but it is not that hard to parse without this restriction, especially as easy-to-use XML parsers are available for most languages.

Also, the requirement for the href to be an absolute URL seems to contradict the HTML specifications, which explicitly allow relative URIs according to RFC 1808 in this context. What is the reasoning behind this?

Hi,

»» I realise that this makes parsing slightly easier, but it is
» not that hard to parse without this restriction, especially as easy-to-use
» XML parsers are available for most languages.
A XML-Parser will show you only a parse error on 98% of the websites which are outthere today, I guess. So it is not really helpful, isn't it?

As if there is no other use for this tag as for Pavatar autodiscovery there is no use of a type attribute or changing the attributes order I think. But it is much more difficult to parse if we allow that.

Actually I got the idea from the Pingback spec, see the green note: http://www.hixie.ch/specs/pingback/pingback#TOC2.2

» Also, the requirement for the href to be an absolute URL seems to
» contradict the HTML specifications, which explicitly allow relative URIs
» according to RFC 1808 in this context. What is the reasoning behind this?
The reason was easy parsing again, but after all the time I think it is a unnecessary requirement, I'll perhaps change it in the next version.

Thank you very much for your comment, I really appriciate it :-)

/Jeena