Requesting the root favicon.ico automatically I will make configurable through a pref. The <link> solution is still coming and will be implemented also, and it will support multiple image formats, and it will be used in lieu of favicon.ico if specified by a Web page.
A special icon cache won't be required with some clever RDF work. I'll be sure to write something up on that when I get into implementing the feature for actual bookmarks. The intent here is not to emulate IE. It is to hijack the feature and do it better than IE. That means supporting <link> and supporting multiple image formats. Let's not get hung up over the /favicon.ico part. Heck, I had to support that just to test the darn feature. :) /favicon.ico is designed just to be the fallback when no <link> is specified, and I'll be sure to make it pref configurable so we can flip it on and off as we determine whether or not it's going to be a necessity or not. dave ([EMAIL PROTECTED]) Michael Nahrath wrote: > Colin Thefleau <[EMAIL PROTECTED]> wrote: > > >>Like you pointed out, Konqueror is >>requesting the file anyway, if it exists or not. I was aware of it only >>when I saw all those errors in my log files, and I agree, this is not the >>right way of doing it. I hope the Konqueror team will correct it. >> > > I hope the Mozilla team will not repeat it. > > Giving a better example and evangelizing it to the webmasters might even > be a help for the Konqueror team. > > Greeting, Michi >
