Friday, March 19, 2010

The Far Off Future of NFS

FamilySearch product managers, Ron Tanner and Tim Cross, talked about the long-term plans for New FamilySearch (NFS) during their presentations at the South Davis Family History Fair and the St. George Family History Expo, respectively. NFS is the FamilySearch tree product.

I’ve mentioned before that Tanner warns his audiences that he is a product manager and product managers can’t say when an improvement will be made. They can’t even promise that the improvement will ever be made. They can only talk about what might be. (Sounds a lot like a fairy godmother, doesn’t it?)

I’ve also mentioned before how funny Tanner is. True to form he had his audience rolling in the aisles. One funny part of his presentation was a series of hilarious traffic signs. This was one of my favorites:

In large letters: Caution! This Sign Has Sharp Edges. Do Not Touch the Edges.  Little letters: Also, the bridge is out ahead.

Tanner related the funny signs to the frustrations users have with the current NFS system. Speaking to the current limitations, Tanner quipped, “The general public doesn’t want to deal with this junk!”

Tanner explained the “Source Centric Open Edit” (SCOE) approach that he is driving towards. My article, “Top Secret FamilySearch Project,” covers a previous presentation by Tanner in which he gave a lot of detail about SCOE.

Source Centric Open Edit

While not mentioning SCOE by name, Cross called it “empowering the patron.” He stated the objectives of these changes to be:

  • increasing collaboration/communication
  • providing opportunities for users to serve
  • allowing patrons to be stewards of the information

There is a tricky balance between allowing users to make good changes and preventing users from making bad changes. The current system makes it much harder to correct bad combines (merges) than doing bad combines. SCOE seeks not only to make it just as easy to undo as to do, but to implement a way for users to receive notification when someone changes a person the user is interested in.

Person Page Prototype

Tanner showed an early prototype showing what the SCOE person page might look like. Parts of the interface used the same low-contrast, hard-to-read design as the beta FamilySearch.org website. Once Tanner tried to point something out on screen, only to discover that the light gray text on an off-white background was illegible. I hoped he didn’t rack up the problem to projector limitations. Certainly if a 50 year old product manager can’t read the text when projected, a 70-something patron will have problems, even on a computer screen. But I digress…

Unlike the current NFS, which uses tabs to divide up personal information, parents and siblings, spouses and children, possible duplicates, etc., the prototype had all the information on one, long page. I suppose this was done because it is felt that scrolling is more natural for users than clicking the tabs. In my experience, I find that many users never think of scrolling, which is why it is so important to keep vital information “above the fold.” But again I digress…

The prototype page included:

  • Possible matching records from Record Search.
  • Possible duplicates in the NFS tree.
  • Theories, with discussion and links to sources.
  • List of recent changes. A complete log of changes allows detection of pedigree wars.
  • Banners similar to those on Wikipedia, indicating ancestors that needed attention. Here are some example Wikipedia banners:
    Example Wikipedia banners 
    Tanner’s example banners included:
    • This ancestor has no sources.
    • This ancestor needs attention from an expert.
  • Community requests. Send a request out to the community if you need an artifact, gravestone photo, record translation, and so forth.
  • Artifacts

Sidney Tanner
FamilySearch is working towards
linking artifacts to the NFS tree

Artifacts

Artifacts are the sources, pictures, stories, web links, and other things about an ancestor besides the basic facts. Tanner showed an example artifact, a picture of one of his ancestors (right). Then he mused, “Why is it that the only surviving pictures of our ancestors look like their driver license photographs?”

Genealogical best practice is to keep source artifacts separate from conclusion trees. For whatever reason, perhaps to kick-start the NFS tree, FamilySearch poured in a bunch of source artifacts: Ancestral File, the International Genealogical Index (IGI), most of the Pedigree Resource Files (PRF), temple records, and membership records. To maintain the integrity of temple records and membership records, the information from them could not be corrected.

Now it is time to pull these artifacts back out into separately searchable collections. The information from the artifacts will remain in the tree, with links to them. This allows users to fix any information in the tree, regardless of its source. And the sources can be searched independently, allowing specialized searches, such as by batch number.

When asked about patron submitted artifacts, Tanner said it is one of the top three changes he is working towards, the other two being discussions/disputes and undo/logging user changes.

You’ll want to listen closely to your fairy godmother. Be home by midnight. And with any luck, the prince will show up with these changes, albeit in lots of little changes.

2 comments:

  1. I really appreciate your insight and contributions to understanding where we are going with NFS. Those of us out here in the far reaches of Mesa, Arizona obviously do not have the opportunity to gain your insight. Thanks for the good work of keeping us informed.

    ReplyDelete
  2. I missed this when you published it because of my South Pacific Adventure.

    Thanks for more info here - putting it in the grist mind for chewing over.

    ReplyDelete