21st century syncing of PAF / FamilySearch / IGI

[thedqs]

The current form of FamilySearch will be a web based program with and API for client programs so that you can still use PAF, Legacy, Gramps, or whatever else you use and it will just synchronize together. That way you can take your laptop, PDA, smart-phone/iphone, etc to the cemetery, update your info and then sync it with the world.

[tomj]

Just a couple of quick comments/observations I've had as I've read this thread:

1. we should never forget the typical end-user. As power users sometimes we have a stigma of build it just for people that are on broadband/Best Set-up . . . etc.. . . . If you build it they will come. yes this is true to a point, but there are still meeting houses on this earth that have limited to no electricity an offline client is a necessary reality.

2. Syncing - Syncing allows for the best of both worlds. My biggest concern with syncing is: accuracy. I don't know how many times things with my palm got messed up and ended up making a complete set of duplicates of every contact & calander appointment. My point is that I hope that this syncing is an area that he church is paying alot of attention to. I sure hope that they iron out the best possible way to do this so that there are not major problems with syncing. . . .

[jmasters-p40]

I recall a "tech talk" type of meeting in Pleasanton a year ago in which I suggested a wiki-model for attaching info to names put up on the web in a big family tree. I said that with trepidation and full knowledge that we're at the mercy of people making mistakes with the data, leading us on wild goose chases.

But I'm now remembering that there was an effort by an individual in Salt Lake City a few years ago to purposely create "random families and individuals" and upload them to databases under the guise of "making people do their research" and not just "take it from the net". From other stories about this guy, my impression was that he was trying to ruin the LDS church's desire to put up a good resource for family searchers everywhere. (Similar to dumping salt into a well to spoil it.)

How does this this connected database avoid small or large scale "data-spoiling"? It's a well known tactic to provide enough "false noise" to discourage the use of one's senses to seek out a reliable solution.

We see stories right now of someone rooting around in Ebay, putting up fake auctions based upon other people's accounts (and good seller ratings). Ebay's (and the sellers) credibility has decreased rather quickly.

Do we put in a "reliability flag" on submissions? How do we determine that someone is "untrustworthy"? If we put in links to documents scanned in, wouldn't we fall for photoshop-ed documents?

Comments?


Justin

[russellhltn]

But I'm now remembering that there was an effort by an individual in Salt Lake City a few years ago to purposely create "random families and individuals" and upload them to databases under the guise of "making people do their research" and not just "take it from the net".

I've not heard that. I did hear of a guy who manufactured false genealogies to attract people to his page to sell advertising. I don't think he went as far as creating false sources.

Frankly I think a more troubling problem is when two people agree that someone is the same ancestor but have different beliefs about the details. Once we resolve that in a way to keep both parties happy, dealing with mischief should be trivial.

[blackrg]

I recall a "tech talk" type of meeting in Pleasanton a year ago in which I suggested a wiki-model for attaching info to names put up on the web in a big family tree. I said that with trepidation and full knowledge that we're at the mercy of people making mistakes with the data, leading us on wild goose chases.

Well we already know that despite best efforts there will be mistakes, so I think we do best by just accepting there will be error and planning that into the model. I think the important part about people putting data up is the ability to preserve >their< data just as they put it up because there opinion, just like all the others out there, will need to be heard. I also think it's very important that conflicting data be preserved (regardless of the degree of conflict, sometimes "minor" differences in data are enough to help you discover there are two or more people with very similar characteristics that you're trying to treat as the same person). Even when you're dealing with identical data, if there are multiple sources, the sources should be preserved. Studying these duplications/conflicts/sources/etc can help those who care to find the real sources. For those who are "less detailed" in their research (either intentionally or unintentionally) I don't think false data is going to make much difference. For those who are detailed, the false data isn't going to make much difference either because they're going to see through it.

With the work the Church is currently involved in to put a tremendous amount of source documents within easy reach of all of us, I think it's going to be much easier to verify data (even for the novices) than it has been in the past.

As for creating false "source" documents, I don't think that's going to be a large concern either - just consider the tremendous amount of time needed to be able to do that on any scale... Also consider that the data will conflict with other source documents (which we all come across normally anyway which leaves any source document under a certain amount of suspicion), and if it's duplicating any easily obtainable source document it's going to be easy to spot it as a fake.

I don't think we have much to worry about from false data. I view the big challenge as being able to display all of the available information, conflicting or not, in a way that meets the needs of every researcher out there - particularly those coordinating their research with others (I coordinate research in my family - it's a headache keeping everyone's changes in sync, I'm hoping the new website will make a lot of this far easier).


BTW - just as an example of how confusing things can get sometimes, a few years ago I ran into someone who showed different parents for one of my (a few greats) grandmothers. I challenged their information in a kind and polite manner hoping they would point me to some sources or what not so I could figure out why they saw it differently. The info I had was based on my Dad's research and I had some source documents and felt that what I had was fairly solid. Unfortunately the person never responded back to me, but I'm a stickler for accuracy anyway so I pursued it. Come to find out I've got two women with the same name born in the same place (and grew up and died in the same place) with just a few years between their ages. Even once I had verified there were actually two it looked to be impossible to sort out which had actually married my (a few greats) grandfather. A few years later while doing some on site research at a genealogical society, a researcher over in another area suddenly poked his head up and said "Aren't you looking for Blacks?" when I replied I was he told me he'd stumbled on an obit he thought I might be interested in - turned out to be an obit for this grandfather of mine that showed that his wife was still living when he died. Of the two women I had identified, one had died prior to my grandfather, the other after, so from that I had my answer.

[garysturn]

Under new FamilySearch all individuals with the same name and dates will be linked together, and even though the term merged is used to combine all the information, each persons submission information will be kept seperately. You will be able to edit, delete, or update information you send in, but all you can do with other peoples submissions is attach comments or disagreements to their data. The version with the best sources will be the main person displayed in the public view of the person. All other persons submissions will be available to look at and review but not in the main pedigree, they will be listed as additional contribuitors. I am not sure how the Sources will be ranked as to which Sources are the best but I understand their will be some type of source ranking. You will have the option in your preferences to have all of your submissions displayed if you don't want the highest priority submissions listed when you go online. If someone were to submit bogus information, it would soon be downgraded in the system as people dispute it.

This is not a true Wiki but in some ways is better than a true Wiki. Here is a link to the Introduction to new FamilySearch which was given to the Beta2 testers. It explains a lot of how things work. This link came from rzamor1's blog.

http://freepages.genealogy.rootsweb.com ... sguide.pdf

[russellhltn]

Under new FamilySearch all individuals with the same name and dates will be linked together,

Which raises the question, how close is "close enough" to make them the same person? It wouldn't surprise me a bit to find two people with the same name living in the same town in about the right year. When it comes to people, never say "that can't happen".

[garysturn]

Which raises the question, how close is "close enough" to make them the same person? It wouldn't surprise me a bit to find two people with the same name living in the same town in about the right year. When it comes to people, never say "that can't happen".

That is always possible, new FamilySearch brings all the possible matches together, it does not do the merge of these individuals. Decendents working on the lines make the decision whether the matches are the same person. And if someone is wrongly marked as a match, that merge can be unmerged at a later point if it is determined to not be correct. This is a great improvement over Ancestral File which merged the names and they became one record, now merges just become contributors to the record.

The best thing we can do to get ready for new FamilySearch is to clean up our data and get our sources entered under sources. (many people have their sources in the notes)

[russellhltn]

new FamilySearch brings all the possible matches together, it does not do the merge of these individuals.

OK, that's good to hear.

[rmrichesjr]

OK, that's good to hear.

Letting the users do the combining of duplicates is undoubtedly the right thing to do, but it has a downside: there's going to be a whole lot of combining to do. In the current beta, many of my ancestors had several duplicated parents and/or several duplicated spouses. Many of the families have _many_ duplicated children, 93 children in one family I happened to look at. Getting that cleaned up is going to take quite a bit of work. I hope most people will be careful when doing that.

The issue of combining vs. separating different contributions is one area where, at least from what I have seen, it appears there is some potential for different researchers to disagree and cause each other difficulties by repeatedly changing the database back and forth. I hope people will be reasonable and civil about such situations. I wonder if there are plans for FamilySearch support to mediate or defuse such situations if they arise.

Lest the above give a false impression I am negative about the new FamilySearch, I have to say I am _V_E_R_Y_ excited to see it happening. I hope it is able to go live soon. It's going to be beyond Marvelous!