Re: SVGT 1.2: Appendix C3: SVG Error Processing

On Jan 9, 2006, at 11:37 PM, David Woolley wrote:

>
>> I would hope that W3C specifications are meant to be engineering
>> documents, not marketing documents.
>
> Time to market has been stated, on this list, as one of the primary
> driving factors for SVG Mobile.

Even if this were an appropriate goal for a technical specification  
(and I don't think it is), that still does not make it a marketing  
document. After all, we are doing a technical review here, not a  
focus group.

>> plans to do so. (Or, for a lower-level example, consider TCP/IP,
>> where being "tolerant" of invalid packets is considered actively
>> harmful.)
>
> TCP/IP is a fuzzy term, but one of the core principles of the RFCs is
> be correct in what you send but be *liberal* in what you accept, so
> the key design principle is actually to tolerate non-conformance.

Depends on what layer. I mean specifically the TCP and IP protocols,  
where acceptance of malformed packets is not encouraged. Higher-level  
protocols like http are treated differently. Since SVG is a higher- 
level spec, I think well-defined tolerance of ill-formed content  
would be a better policy.

I'm actually not sure what the point of disagreement is. Are you in  
favor of specifying intolerant error handling, specifying tolerant  
error handling, or leaving it unspecified? How does your view of  
market conditions apply to this?

Regards,
Maciej

Received on Tuesday, 10 January 2006 09:36:37 UTC