Not a lot to add at this point, but this demanding project that had me find a way to determine an accurate reference point to measure the impact of SBAS, does tend to show what regular users of consumer GPS systems have found over time, i.e. that SBAS will not have a measurable impact on the user experience. Certainly not at the level that "specs" would have us believe and make everyone want to go out and get a SBAS enabled GPS system over a non-SBAS system, even if we had maps that were accurate down to one meter, wich we know is not the case.
In terms of accuracy, it seems that more can be done by the chipset designer at the firmware/software level to improve the accuracy as SiRF have apparently done between the SiRFstarII and SiRFstarIII generations although based on my side by side static accuracy testing Nemerix seem to have quite some advantage. This will be the subject of another article, but with the wider availability of very accurate aerial maps (thanks to Google Earth), I've started taking a closer look at dynamic accuracy and there again, Nemerix appears to excel as seen in the picture below in a very demanding area of Paris, where the Nemerix v4 iBlue (green) tracks perfectly while the SiRFstarIII BT-338 (blue) and Fastrax WBT-100 (red) wander quite a bit.
If you have questions or comments, you can use this thread of the forums and don't forget to parcipate in the SBAS Poll (May/June 2006) !