Re: SVG12: Script Encoding

Ian Hickson wrote:
> On Sat, 15 Jan 2005, Bjoern Hoehrmann wrote:
>>Various document formats either provide or propose means to specify the 
>>character encoding of external scripts if there is no reliable way to 
>>determine the character encoding e.g. when loading the scripts from the 
>>local disk if the file system does not support such meta data. Among 
>>these formats are HTML 4.01 (<link rel=script charset=...>), VoiceXML 
>>2.0 (<script charset=...>), XHTML 2.0 (<script charset=...>) and various 
>>derived formats. Please change SVG 1.2 to provide similar means to 
>>ensure interoperable encoding detection for external scripts.
> Personally I consider those attributes to be a huge pain in the neck. They 
> have caused any number of problems on the Web (e.g. the same resource 
> being treated in two totally different ways purely due to the linking 
> mechanism). Please rely only on the resource's own metadata to make this
> kind of determination.

I agree, charset attributes must die. If you've written your script in 
some baroque legacy encoding and stored it on the file system you've 
screwed yourself and deservedly so.

Robin Berjon
   Research Scientist

Received on Monday, 17 January 2005 16:45:57 UTC