Jim Lyon wrote:
I've been asked to defend the proposition that MARID records need more
extensibility than can be easily afforded by SPF syntax.
Well, it has been argued (successfully, IMHO) many times that the architectural
advantage of SPF's non-extensible syntax is one of its best attributes.
By freezing the feature-set, SPF is more stable, easier to understand and
implement, and more uniformly practiced. Less ambiguity provides the authors of
SPF records a larger degree of assurance that their records will be parsed
correctly. Frozen syntax and semantics make MTA implementations easier to write
and debug, and more likely to evaluate the records uniformly.
A hidden problem with extensibility is the fact that non-uniform practices can
emerge with multiple competing mechanisms fighting for dominance (imagine
browser on SMTP). With a frozen spec, things are a lot more predictable because
new mechanisms are thoroughly debated and vetted before being released
uniformly as a new version.
In general, extensibility is not a benefit to MARID.
Michael R. Brumm