Tech Forum

RussellHltn wrote:I think that may have been a limitation of the Beta. Of all the databases pulled from, only PRF allowed anyone to say what the real source was. Had you pulled from another NFS user you probably would see their source notes.

The real sources (certificate, gov. or church books, etc.) were available by clicking on the link to the right of the datum (place, date, name, etc.). That should be available in the 4-way split panel, in the 'individual details' tab of the lower sub-panels, when comparing to combine duplicate records.. In any case, I'm looking forward to the real, live system!

Here are some more screenshots showing sources and notes in NFS. Screenshots

GarysTurn wrote:When you compare to merge, all the vital information is there. NFS (newFamilySearch) does not combine information, it links it together, so any notes and sources are added as additional information.

The API will allow programs like Legacy to add functions to their software to work with the NFS data, so you may be able to do all your work on NFS directly in a future version of Legacy, Family Tree Maker or some other desktop program, without ever needing to go to FamilySearch.

Question: Can you "unlink" people?

The New Family Search has been rolled out to at least two Temple Districts so far. Was the beta testing that was done just a stamp of approval? Or was it a simple case of the developers really believing that it was ready?? I know that there would have been pressure from further up the line to get this out ASAP. Would the programmers have had time to fix all the issues that it had in such a short turn around time?

thedqs wrote:Question: Can you "unlink" people?

Yes, you can unlink people. Given the likelihood that some (hopefully small) fraction of the merge operations will be wrong, an unlink facility is necessary.

However, in the Feb-April 2007 beta, the unlink facility was not very easy to use. A lot of people are made up of over 100 records (I saw about 360 on one). The unlink facility shows you five at a time, and during the beta it took a good 10-20 seconds to move one step. Then, you can only separate out one of the records. So, if you need to remove the last three records of somebody made of 150 records, it's going to take a couple of hours.

During the beta, I submitted a suggestion for improvement, so let's hope something is done, either in the basic web-based system or through the API.

sbarnsley wrote:The New Family Search has been rolled out to at least two Temple Districts so far. Was the beta testing that was done just a stamp of approval? Or was it a simple case of the developers really believing that it was ready?? I know that there would have been pressure from further up the line to get this out ASAP. Would the programmers have had time to fix all the issues that it had in such a short turn around time?

That's cool that the system is going live now!!! There is a "special" beta going on April 30 through May 2 to check specific things in the "final data load." Somewhere, I heard it will take about 18 months to from first temple district to last district to go live with the new system. I also heard that there were a number of issues with the system (and saw and reported a few) but a decision was made to go with what they have. The response to some of the issues I reported was that the issue would be fixed before release of the final system. I hope most of the fixes made it (or will be added before too long).

While I have expressed concerns about the beta, I think some of the bigger ones are being handled at a different level then the technology visible in the beta. I think everyone agrees there is room for improvement, but I think the second beta probably showed that it's "close enough" to get started.

Keep in mind that in the current situation, not all of the FHCs may be checking against the on-line IGI resulting in a lot of duplicates. The "cost" of a premature roll-out has to be balanced against the "cost" of delay.

Over all, in it's current state, I consider it a step forward for reasions that may not be obvious.

Someone who I know who tested NewFS had two issues that are to me rather glaring, who can I contact to get the word to about the problem as that person got no response during the beta. The bigger one has to do with combining names.

JamesAnderson wrote:Someone who I know who tested NewFS had two issues that are to me rather glaring, who can I contact to get the word to about the problem as that person got no response during the beta. The bigger one has to do with combining names.

I don't know anyone to contact. My guess would be that the report(s) your acquaintance apparently made through the beta test feedback mechanism will do the most good that can be done.

Yes, there are a few issues related to combining. I reported a number of issues and have described in this forum a few I saw.

If you would be willing to describe exactly what the two issues are, perhaps those of us who were/are beta testers could tell you about any response we may have received if we reported the same or similar issue. Or, perhaps we could explain a workaround we may have found during the beta test.

Here then are the two issues.

1. Combining names when the order of the names was entered into a previous submission by mistake. This person had a hispanic husband who is long since deceased. He entered some things into PAF and sent them into one of the legacy databases where the records now sit, awaiting proper combining. Basically, it's like John Smith was entered Smith John, and she wanted to find some way of combining them so that others know this was the same person, John Smith rather than Smith John.

2. There was an excommunicated member in her line that had the work done for them after the individual involved died. Apparently it would not let her combine the record for some reason. If this isn't enough I'll be talking with this person tonight and I'll be able to get more on this particular issue.

JamesAnderson wrote:Here then are the two issues.

1. Combining names when the order of the names was entered into a previous submission by mistake. This person had a hispanic husband who is long since deceased. He entered some things into PAF and sent them into one of the legacy databases where the records now sit, awaiting proper combining. Basically, it's like John Smith was entered Smith John, and she wanted to find some way of combining them so that others know this was the same person, John Smith rather than Smith John.

2. There was an excommunicated member in her line that had the work done for them after the individual involved died. Apparently it would not let her combine the record for some reason. If this isn't enough I'll be talking with this person tonight and I'll be able to get more on this particular issue.

I don't see enough information about issue 2 to have an understanding of what might have been wrong or what could be done as a workaround. I do know there are were some issues in the Feb-April beta (beta2, I think) that prevented combining of some records. I reported one such case, was instructed to combine the children and try again. (I would also make sure the parents and spouse(s) were combined, too.) In the regular beta, it never did let me combine the "Unknown name" wife with the named wife. I plan to revisit that case this afternoon in the special test of the "final data load".

Let me see if I understand issue 1: Is it the case that both records are in NFS, but the names are different enough that NFS is not finding them as possible duplicates? Is it also safe to assume that one of them is in your pedigree, with children listed? I would try a few things, and I think they would eventually allow the records to be combined. First, I would add another opinion to both records giving the exact same name.

Second, and probably more likely to succeed, I would use the "find a ..." links to connect _from_ the one not on your pedigree _to_ someone who is, preferably a child. I think that should turn on the "multiple parents" asterisk. I would use that to combine the records as a parent of the child you linked to. I'm fairly confident that should allow combining as individuals, and from there you're home free.

Third, I would use the "find a ..." links to connect _from_ someone who _is_ on your pedigree _to_ the record that needs to be brought into the fold. Between these three methods, I'm pretty confident you can bring the record into your pedigree and combine it up.