Banjiman wrote:

1: "OK, I understand that there is a need to catolouge (as with stamps) but what do we hope this will tell us? Is it about the correct/ original context of the song, evolution or something I'm missing?"

Yes to your second sentence. Tradsinger and Tom B. have already answered this partly, but to me it's a simple matter of human curiosity. For instance, my early experience of the folk world was in North West folk clubs (particularly Harry Boardman's), where the 'local tradition' was seen in terms of the songs of the cotton mills and other industries. Two of the best-known songs of that genre - both of which I've been known to sing - are "The Four Loom Weaver" and "The Handweaver and the Factory Maid". I can't understand why anyone would choose to sing those songs without being at least mildly curious about where they came from, who composed them, who actually sang them a century or more ago. In both cases it turns out that their provenance is murky, with probable or definite editorial intervention by Ewan MacColl and Bert Lloyd respectively. That doesn't mean that they're not worth singing, but it does mean that I would feel the need to be careful about introducing either song to an audience with words like "Here's an old song that Lancashire cotton mill workers used to sing."

2: "Is this to inform how a song "should" be performed?"

Some people who have spent a lot of time listening to traditional/source singers find it more difficult to enjoy the efforts of modern performers (particularly with respect to things like accompaniment, pace, storytelling, etc.). All of us have our own ideas about how songs should be performed (usually defined as the way we do it ourselves), but most of us are aware that these are only opinions. Telling someone that the way they're singing is 'wrong' is plain bad manners. Suggesting to that person that they might enjoy listening to a particular performance - perhaps a 'source' version, perhaps not - would be a bit more subtle.