1- every full text from every release gets stored to a data table
2- so instead of searching against what's "live" in the RSS feed, it's searching against ALL of the feeds going back in time (from whenever we start storing them)
3- the searching would still be much the same (matches are displayed, and the link to the original URL would be shown)
4- Also - some of the feeds that I'm accessing on the RSS admin side do not seem to be capturing some of the items. I'm not sure whether this is the fault of how the RSS is configured by the publisher, or whether the tool is actually missing something, but part of this next wave of work -- if we do it -- would be examining those problematic RSS feeds to make sure they are being properly indexed.
5- One other housekeeping item - I will need to make the search app password protected.