> The published APIs didn't seem to lend themselves to a
> simple screen-scrape.

Well, they're mostly XML/SOAP, which is a pain, but they're not too rough as
such things go. Certainly they change a lot less than the web pages do, so
even if you're just using exactly the same techniques on what's returned it
should be an improvement from a reliability perspective. UPS's seem a bit
better than FedEx's (I don't recall looking at USPSs). FedEx has a "web
integration" thing which might be easier to use here, but I don't know details.

If you don't want to do that, I think the only other option is another round of
updating the HTML scraping that's there now.

Anthony