New Beta Family Search

[rmrichesjr]

Is this Beta for Family Search going to include a Web-based Version of PAF? I'm a Mac user, though we are small in numbers, we are a growing community.

I hope that the church is looking into a web-based Paf system.

If you might be interested in a chance to maybe run the current PAF on an X86-based Mac, you might be able to use the WINE program to do it, if not now, some time soon. The best way to get it to work is to try it, file bug reports, encourage someone to fix the bugs, try it again, file more bug reports, etc. I posted a link to the WineHQ AppDB page for PAF 5 on Linux to the forum thread about PAF on Linux. HTH.

[rmrichesjr]

Thanks.

This makes the system very hard to work with as you can not produce an accurate FGS. These are living children and 2 of my children are duplicated and the other 2 are not. I guess I just don't understand why it is this way. Maybe someone can explain it to me.

Roger

Regarding the children who are duplicated, have you tried combining them? There's an overview on that topic.

A grandfather of mine was duplicated, as were most of his siblings. It was not difficult to combine them. It's even possible to separate again if combinings were done in error. To test the functionality and to learn more about it, I combined and separated my grandfather a few times.

A lot of my ancestors, both parents and children are duplicated. The "tree" right now has a bit of resemblance to a briar patch, at least on that one branch. Unless the real, post-beta system does more thorough automatic combining of duplicates, there will be a lot of combining to do when the system goes live for real.

Hope that helps.

[thedqs]

That could be one of the reasons for the beta testers, maybe, to give the tree the once over and remove a lot of the duplicates that already exist. I am sure that on my line we have tons of duplicates too. So that is something to look forward to.

[rmrichesjr]

The purpose of the FamilySearch API is sorry developers can create read and update routines with the New FamilySearch. Many PAF Add-in developers are looking at this possibility so that a PAF using can upload and stay in sync with the new FamilySearch without using GEDCOM

From a technology standpoint, it wouldn't be too hard right now to export from the current new FamilySearch beta system to whatever format one wanted. Family Group Record format can be traversed by clicking on the "show family" links. The HTML for the pages could be saved, parsed, and converted to GEDCOM or other format. The Pedigree form could be used, but it would take more page loads per step.

The only thing I could see in terms of use that might possibly oppose this would be that one is not allowed to slow down the system, and that could be solved by putting in a delay of a few seconds per page request to keep the load on the servers reasonable.

The adventurous side of me would love to start design and coding first thing Monday. The conservative side of me says I should probably not do it unless somebody authorized to give a green light were to give one.

[thedqs]

The thing about the API is that it reduces the amount of information that has to be transfered across the network and thus reducing the bandwidth needed for a single user and so more users can connect to the FamilySearch website. For example to send a person through the API would be:

Code: Select all

<person>
    <name>
        <givenname>John</givenname>
        <lastname>Doe</lastname>
    </name>
    <birthdate>05051755</birthdate>
    <deathdate>09211825</deathdate>
    <marriagedate>07231780</marriagedate>
    <notes>
        <note>
            <text>Link to marriage certificate</text>
            <link>http://churchimageserver.com/images/555555555.jpg</link>
        </note>
    </notes>
</person>

Now for displaying this info to a webbrowser, even in a pretty simple form you'd need twice or more of code giving the table, style, position. Linking the image in. For more advance websites you'd need to include the javascript, stylesheets, etc. That all takes up time and room for the little information that you really are looking for. And for millions of people, the less amount of info to send means more time people can spend doing the work instead of waiting for the servers.

[rmrichesjr]

Agreed, an API of the type you described would be much more efficient than parsing human-readable web pages.

Is there somewhere we can get a bit of information about this FamilySearch API? There are about two postings in this thread that mention it. A search on this site for +familysearch +api returned a bunch of threads, most of which have no mention of "API" in the thread. Google didn't turn up much of anything.

[rmrichesjr]

That could be one of the reasons for the beta testers, maybe, to give the tree the once over and remove a lot of the duplicates that already exist. I am sure that on my line we have tons of duplicates too. So that is something to look forward to.

Actually, the current beta seems to be only to test the software, not to clean up the data. The instructions for the beta test say all the data will be thrown away, so we can add ficticious stuff, dispute anything, "be creative", and "have fun". I haven't added anything ficticious yet, but I did dispute the birth and existence of Fred Flintstone, born in 1776 in Ohio, US. Hey, they said to have fun.

[thedqs]

A search on this site for +familysearch +api returned a bunch of threads, most of which have no mention of "API" in the thread. Google didn't turn up much of anything.

The Search Engine for this site doesn't look for anything under 4 letters so API even though it shows up in the searching criteria, was not used to search. This was talked about in another thread.

Anyway I haven't heard anymore then you on the API, but I am sure if you start a thread asking for info, people in the know will respond to your questions.

[rmrichesjr]

Thanks for the explanation of why searching wasn't doing what I thought it should. That (inconvenient) feature of the search system (only looks for words with four or more letters) sure makes it difficult to search of instances of certain TLAs. (Three Letter Acronyms)

Also, thanks for the suggestion to start a thread asking about the new FamilySearch API. I think I'll do that.

[russellhltn]

(Looking at my email in-box) <sniff>

Whoo-Hoo!!! <cartwheels away>