Repetitive Sources

LegacyUser

Lori Conner said: Hi - I love the hints that show up to guide us to missing information. But they show up whenever the information has been added. The only two choices are to add it to the source list or choose it's not a match. Could we please have a third option? Maybe - it's already there? I spend a LOT of time reworking my line because people have added those sources and there are times there have been 8 or 9 copies of the same thing. Identically the same thing. I deleted 28 of 35 sources on one grandparent just yesterday. All 28 were identical duplicates of the 7 different sources - not a similar record from another source. The only workaround I have found when a source "wants" to be added is to go ahead and add it, then detach it. I love finding the hints, because sometimes it's something I haven't found. But the duplication is taking a lot of time - sometimes more than an hour a day. Heeeelp!

Thanks for all you do!
Lori

LegacyUser

David Newton said: Please provide the PID of your grandparent. It is very often the case that these repetitive sources are different transcriptions of the same data and so from the point of view of FSFT are actually different sources. This is a very real problem, and there are databases which need to be weeded to get rid of these repetitive transcriptions (worst I've come across was 5 different entries for the same christening). However the issue is the sheer scale of the task. We are literally talking about going through tens of millions of entries to remove millions of these entries. There's also a considerable programming challenge to set thos up.

LegacyUser

Cindy Hecker said: It is better to have multiple sources on the same person than for the source to be attached to the wrong person or possibly a duplicate gets created from that source not attached. Until they figure out how to combine sources, keep all of them attached to your person.

LegacyUser

Alan E. Brown said: It can be a bit confusing when the source titles are all the same, but it is often the case that the images are different, or sometimes the indexed text is different. In those cases, if you attempt to "clean up duplicates" you may be removing important information that is only in the sources you detach.

A good example of this is in marriage records. They may come from the same record collection and thus have the same title. But they might not be duplicates -- one record might be the application for a marriage license, another might be the license, another the marriage certificate, and yet another the marriage return. These would all typically have the same title, something like

Opal Elaine Berrett, "Utah, County Marriages, 1887-1940"

Yet the images are different, and the indexed data might well be different.

A concrete example for the above title would be these three sources. Each of them has a different image, and thus they are distinct sources. One of them has distinctly different information from the other two. So it would be a loss if someone detached all but one, thinking they tidying up the source list.

https://www.familysearch.org/ark:/619...
https://www.familysearch.org/ark:/619...
https://www.familysearch.org/ark:/619...

So I would certainly encourage anyone who is tempted to clean up source lists to carefully look at the details of each source and make sure that they are duplicates in every regard. I have found some wonderfully helpful nuggets of information among the unindexed information on the images of marriage records.

Note also that the simple act of detaching a source may well cause it to reappear as a hint, and someone is likely to reattach the source. The only way to avoid that is to see the hint yourself, and mark it as "Not a Match" -- which is misleading. It's generally best to attach all the sources. If anything needs to be corrected, it is on the record collection side of things, which is nothing you can do yourself.

LegacyUser

Lori Conner said: Thank you for your quick reply. I understand what you are saying about the images or possibly the indexed information being different. Yes, I will look closely. However, I am in the process of printing the sources (Using family with sources) and what is printing out is identical. In the case this week, it was printed 9 times for each family member. Since the family had something like 12 or 13 children, you can see how many times that printed. I'm not sure what the solution is, but surely there must be a better way! Even if there is a slight difference in the document, surely it would be better to somehow choose just the different bits and only have one printout from the original document.

It would be nice if the programmers could give us a quick way to print the original document. Right now, we have to make a screen dump, transfer the image to a word document (or something similar), crop the image and then print it out. If the original was too long or printed on more than one page (think a will), we have to marry the documents. If we could paste that into the source files, it would be wonderful. But it is nice to have it in my paper files to refer to.

This is wonderful software. It has improved tremendously over the years and I very much appreciate the opportunity to use it. Thank you for all you do!

Lori

LegacyUser

Lori Conner said: The grandfather I cleaned up last night was Dr. Daniel Greene Hunt Jr. KPHS-T4G. I have not looked at his father yet, but it shows he has 37 sources, so I am pretty sure there are duplicates. The father, Dr. Daniel Hunt is L64Q-XQ7. I will go back and look at the sources I deleted last night and double check them. I don't want to detach something that needs to be retained.

Do you think the programming task would be easier if you just added the option to print the sources you wanted to print? Something like ___ Print All Sources or _____ Select the Sources You Wish to Print. Then the user could select one of the duplicates and all the sources could remain attached.

There used to be an option to compare two files to see if there were differences. I haven't used it in years (maybe decades) but it should still be around in some form. That might help to remove the ones that are true duplicates.

Sorry, I am brainstorming here and I should let you do your job. Text was never my thing.

Thanks for all you do!

Lori

LegacyUser

David Newton said: Ah you're producing it for paper files. So what do those paper files actually do? What practical use are they? Do they provide the quickest access to the information? No. Do they provide the most up to date version of the information? No. Do they take up an enormous amount of space? Yes. What will your relatives do with them if they ever inherit them? Probably throw them straight in the bin.

Store the document copies electronically with adequate backups. Much more accurate, much quicker to find, much less physical space taken up by the records. Use one of the three pieces of software that interfaces properly with FSFT.

As for repetitive records showing different information how about this:

https://www.familysearch.org/ark:/619...
https://www.familysearch.org/ark:/619...
https://www.familysearch.org/ark:/619...
https://www.familysearch.org/ark:/619...
https://www.familysearch.org/ark:/619...

Same film number. Same event. Same person. Five separate index entries. Why do these five separate index entries exist? They should have been weeded long ago and reduced to one entry which is as correct as possible. This is one of the individuals who are part of the worst case I have come across that I indicated above.

LegacyUser

David Newton said: Oh boy.

I've just had a look at what's been done to that profile and the results are, erm, mixed in their usefulness.

Some of the source removals are very, very much correct. For example the removal of the Ancestry Family Trees "source" is justified on two counts. As rightly noted in the reason statement it was essentially blank apart from the title of the source, so it cannot be traced to its origin making it useless as a source. Beyond even that in this particular case the "source" itself was invalid. An Ancestry family tree is not a valid source as it is simply a way of storing family tree data like the FSFT system or a GEDCOM file (also a somewhat common and completely invalid "source" some people insist on using).

Other blank sources of that ilk are also very much valid in being removed as they have been attached willy-nilly without a proper citation allowing tracing of the original document which defeats the entire purpose of a source citation!

However there are some very much more dubious removals. As of now the profile has census entries for 1790, 1800 and 1810 attached to it. Therefore removing Ancestry versions of those citations is valid since they are duplicates. However 1820, 1830 and 1840 census source citations were also removed with the reasoning that they were from Ancestry. Where are the corresponding Familysearch versions of the census entries? They weren't completely blank sources and gave enough information that someone could at least have a go at finding the correct part of the each census. Once corresponding entries were found, and assuming they were actually the correct person then substituting the Familysearch version for the Ancestry one would be quite correct. However the corresponding Familysearch version for each hasn't been found.

Turning to the most dubious removals we have the "Identical Duplicate" removals. Now the URL for some of those removals produce a permission denied page for me. For example:

https://www.familysearch.org/ark:/619...

However three of them are precisely the issue that is being referred to by Alan Brown:

https://www.familysearch.org/ark:/619...
https://www.familysearch.org/ark:/619...
https://www.familysearch.org/ark:/619...

Ironically enough they are also precisely the issue that I referenced as well.

These appear to be the correct person. They are different versions of the same source with different Familysearch URLs. They thus conform to what Alan was talking about.

However 1:1:XFSV-YVC and 1:1:XFSC-32F appear to be duplicate indexings of the same document. They do appear to be different filmings of the same document since the lighting is different in the two images. However like the examples from Chellaston that I quote further down this thread they should have been weeded out years ago.

So like I said mixed results for your work!

LegacyUser

Alan E. Brown said: The fact that multiple sources have very similar citations doesn't mean that the image or the indexed information is identical. The citation simply lists the person, the record collection description, and the ARK (the special URL that uniquely identifies the source). So, yes, they will look nearly identical when you print Family with Sources.

I can certainly see how it would be frustrating to have multiple nearly identical entries appear in the sources section of Family with Sources. Of course, they aren't entirely identical, since they will have different ARKs. That is really the most important element of the citation.

LegacyUser

Lori Conner said: I am 63 and use the computer daily. However, none of my relatives older than I am even own a computer. I am sure that my children and generations to come will toss the paper copies, but there are still people who prefer paper.

I agree that digital is best. It's the way I store my personal copies. But...

Thanks for your help!

Lori

LegacyUser

Lori Conner said: Thank you for catching those census records. I had meant to attach them, but then couldn't find them and got frustrated and went to bed. I would probably not have discovered it for a few days - and then I would be trying to remember just which records it was!

Thank you for your help!

Lori

LegacyUser

Lori Conner said: Thank you for your help!

Lori

LegacyUser

Chas Howell said: When the same Title refers to a different image I go ahead and modify the title (after I look at each) so it is more clear. For example same Title may refer to an Index or the image of either a Marriage Application, License, Bond or Certificate. So I would put Certificate or whatever at the end of the Title.

LegacyUser

Lori Conner said: Thank you,
Lori

LegacyUser

Jeff Wiseman said: But REMEMBER, the source that you are attaching is a citation to an INDEX file and *NOT* the digital image. Having the image "appear" to be the same does NOT mean that the different index files are the same.

Any citation that you attach to the source list using the source linker is NEVER a citation to the image! It is always a citation to the INDEX FILE itself (whether the image can be seen or not)

The image file is NOT your source. The Index file that was derived from that image is the source being attached. They are separate things/sources