[uf-discuss] hcard noobie question

[Benjamin Carlyle]

On Fri, 2006-04-07 at 02:29 +0100, Nick Swan wrote:
> just to let you know, posted to LC mailing list and they confirmed
> that it would only parse the hcard in, in the format you can copy out.
> They are working to improve the demo page so it parses all hcard
> formats.

I'm curious as to why parsing is required. Couldn't they just use the
literal html snippet (possibly normalising to xhtml, and stripping out
potential nasties)?

I would have thought that parsing of the html should be deferred to
whatever you paste it into. The intermediatary (the web browser
performing the cut and paste) may have reason to try and understand the
content itself, but by parsing the content as a hcard and pumping a new
"equivalent" hcard out the outher side there is a potential for
unfortunate data loss. If it wants to analyse the content it may, but it
should then paste as close to the verbatim input as it can achieve.

Benjamin.