- From: Heitzso <xxh1@cdc.gov>
- Date: Thu, 3 Aug 2000 18:09:10 -0400
- To: "'www-rdf-interest@w3.org'" <www-rdf-interest@w3.org>
I've been searching for clues for (future) web standards based data portals. I am a consultant at the Centers for Disease Control and am aware of a dozen or so data projects that have one or more of the following components spec'ed: web/internet based metadata registry/repository (as in 11179) data portal for humans via browsers data portal for other servers/apps distributed data sources Most of the context for this data is common relational db type data along with survey data. For instance birth and death counts by diesase, etc. So the focus is *not* document oriented. Many of these projects will hopefully be used around the world and have long lives. I, personally, automatically add the following to the wish list: meta-data is data, so the same tool operates on both discovery of data, i.e. you don't know in advance what data will be found at a site, only that a site will provide data as per some standard technology Does anyone monitoring this list have clues for me re the current-best-guess technology to focus on? (more follows after the sign-off if you want more background info) Thanks, Heitzso Information Technology Branch Centers for Disease Control and Prevention xxh1@cdc.gov ============== The idea of the semantic web is a tease. Yes, a triple can be used to describe the contents of a relational db (tease, tease, tease). But the discussion always pulls back from the brink of full data exploration and retreats to the realm of metadata. (what I mean by 'describe contents of' is that it can represent the data stored in a relational db). What I'm searching for would be for structured data what html/http does for documents with the added benefit of being fully traversible by both humans and robots/apps/other-servers. In other words a semantic web that includes repositories of structured data. SOAP is a tease because it provides a nice language neutral way to ask a question with structured data and get a structured response. But it's just a small piece of the puzzle and there are other ways to skin that particular cat. Clues? Thanks, Heitzso
Received on Thursday, 3 August 2000 18:09:59 UTC