most providers which use the irr see it as how one publishes the parts of one's routing policy one needs to make public, i.e. what one's peers need to filter prefixe announcements, that and no more. so basically, the list of prifixes for an origin is needed, and to which ass one provides transit. (some see even this as too much to reveal publicly) all the rest is cruft for configuring one's own routers, which one likely does not wish to publish and which one manages internally. i believe the rpsl project feels that providers will use private rpsl databases and the tools the rpsl folk make available to configure their routers. i am not aware of tier one provicers which do so. small and medium providers may, hence this would be useful. so the jury is still out on most of the rpsl syntax. if you don't need it, ignore it. randy