should Konqueror render invalid HTML?

[Peter Kelly]

Unfortunately, 99% of real-world websites use at least some form of
invalid html. In most cases the spec violations are only very minor (and
easy to deal with), but in some cases there are larger deviations of the  
spec as well.

As great as it would be if every site was 100% compliant with the specs,
this isn't the case and so in order to have a usable browser we need to
support such sites. In particular this means emulating quirks of Netscape
& IE in terms of how they deal with such code. Of course, we also try to
render compliant pages correctly as well.

On Wed, 19 Feb 2003, Michael Wardle wrote:

> Hi
> 
> (I am not currently subscribed to the Konqueror mailing list.  I would 
> appreciate it if you copied me on any response.)
> 
> I just performed a search for Konqueror bugs on <http://bugs.kde.org>, 
> and noticed there were a substantial number of bugs reported either 
> expecting Konqueror to render invalid HTML or to work "like Netscape 
> does" or "like Internet Explorer does" on badly-written sites.
> 
> Is it the Konqueror developers' desire to make Konqueror work on invalid 
> HTML, especially when doing so might render valid HTML pages 
> incorrectly?
> 
> As I would rather a fast, small, and standards-compliant browser, I 
> would enjoy going thru KDE bugs and closing ones which ask for broken 
> pages to be rendered "right", but perhaps you would rather I didn't.
> 
> Could somebody from the development team please comment on what the 
> objectives of Konqueror are and how you would like such bugs handled.
> 
> I searched the kfm-devel archives and couldn't find a relevant answer to 
> this question, however it is of course fine for you to point me there 
> if I have missed something.
> 
> Thanks
> 
> 

-- 
Peter Kelly
pmk at kde.org