Ideas for relationship links for genealogical data

Views

Watchers

1 Introduction
2 Current Uses
3 Proposal
4 Possible Uses
5 Other Thoughts
6 Resources

Introduction

Many blog websites today have in their <head> tag a <link> tag indicating the existence of alternative formats of the website, usually in an RSS or Atom format. For example, on my blog I have this:

<link rel="alternate" type="application/rss+xml" title="RSS 1.0"
 href="http://brainshed.com/blog/rss/1.0/brainshed.xml"/>

[edit]

Current Uses

Both Opera and Firefox key off of this kind of link tag and display a small icon somewhere in the browser, that when clicked, offers the capability to subscribe to the given blog. Other aggregators (like Bloglines) can use this to make subscribing to a given blog easier, as you only have to paste in the website url (http://brainshed.com), as opposed to a harder-to-remember feed url (http://brainshed.com/blog/rss/1.0/brainshed.xml).

[edit]

Proposal

Why couldn't we extend this idea to genealogical data? Let's take for example, RootsWeb's WorldConnect. Let's imagine that for each individual display page, there were a relationship link like the above, that pointed to a url which represented a GEDCOM format of that individual's information. Or on each page displaying a 4 or 5-generation pedigree, a similar link tag was embedded in the page indicating the url at which a GEDCOM representation of the data being displayed could be downloaded. I think something similar could be done on FamilySearch's Ancestral file, Pedigree Resource File, and IGI search pages.

And why limit this to linkage data? Let's continue with source data. Let's add <link> tags for each source document we display on a web page. Are there structured format standards for source/citation data?

[edit]

Possible Uses

This would make it easier for genealogically-attuned search engines to find and parse all this information across the web and make the results searchable. It would be nice to be able to search (metacralwer-like) across a whole slew of pedigree stores, instead of having to search over and over on a hundred different sites.
Browser plugins or toolbars could be built which would allow for different actions to be performed on the data made available in this way. For example, I could have a browser plugin that provided a "Genealogy sidebar" which would list all the individuals or pedigrees represented on the page I was viewing. Going a step farther with that idea, personal record manager vendors (like PAF, RootsMagic, Legacy, Family Tree Maker, etc, etc) could provide toolbars or sidebars that would let me see all individuals, marriages, pedigrees, sources, etc represented on a site, with options to import each into my own files.
Use relationship links to point to RDF versions of pages which display genealogical source images. Perhaps a project something akin to the LDS Indexing project could be built which would allow volunteers to extract data in machine-readable RDF format. E.g., So-and-so lived in AF, Utah, in 1847. Joe Johnson's father was Robert Johnson, etc.
Others?

[edit]

Other Thoughts

I previously thought microformats were the answer to making this kind of data machine-parseable, but as I have thought about it, I think relationship links are a much cleaner way of doing this. Any number of these links can be added to a page, in the case that you have multiple individuals, or multiple pedigrees represented by a given page. The benefit here is that we don't have to waste time inventing new microformat standards to represent genealogical data. We can use existing formats (such as GEDCOM), and existing methods (relationship links) to accomplish the same ends.

[edit]