Stop the sourcing chaos!
Over and over again I run into records of people in the Tree that have up to 30 sources.
It seems as if patrons believe the more sources of ANY kind, the better. But sources should appear as PRIMARY and SECONDARY sources.
- PRIMARY sources do indicate CLEAR evidence of a person's vital data: Birth/Christening dates and places, marriage date and place, death date and place, and where the source is located. Included MUST be first and last names of the parents.
- A SECONDARY source provides HINTS (only!!) for family relationships between people and indicates that hard evidence is not fully there or still missing.
In reference to German church records: Many of the sources offered by familysearch originate from the International Genealogical Index which is based on extraction records and privately submitted records over a long period of time. Back in time it was well known that the IGI is NOT a safe place to go when it comes to finding accurate information about an ancestor, due to numerous duplications, spelling problems, misreading of records and the fact that technical possibilities we enjoy now were not available in IGI-times.
Other sources so profoundly promoted by familysearch are based on records that have been indexed in recent years (and ongoing) Especially the older indexed records several flaws in regards of accuracy and detail. That is changing to the better but unfortunately the older indexed records are still around and in use: as sources!
It seems to me that patrons, looking at a source provided by familysearch believe that this is automatically a good, solid (PRIMARY) source. I cannot wrap my head around the sheer number of sources for one single person found in the Tree.
We need to READ the content of a source before we do the left click with our mouse. How much hard evidence does my source provide in regards of birth, death, marriage of the person I attach it to?
Your comments/concerns on best demonstrated practices for sources reflect your passion for accuracy and aspirational excellence.
That similar type of passion for attempting to "do the right thing" is part of the spirit of this relatively new concept, a single shared collaborative Family Search Family Tree ( FSFT).
It isn't lost on me as I run into all types of users and experience levels, and yes, the resulting "pause moments" that at times I am reminded to step outside the tree growing process and details and look more broadly at the forest, and long term nature of building this thing called FSFT, itself.
What helps me, and maybe others, is to widen the lens we view the platform itself. First, and maybe helpful to understanding IGI, "legacy", transcriptions, etc. ....all are moving parts of the process which are comparitively different from say a "wiki tree" for collaborative sharing...Wikitree unlike FSFT starts at "nada" and users build the tree from there; the FSFT is different and it is basically the "child" of multiple legacy databases (likely including those 30 transcribed IGI or German church baptisms that got entered twice...smile) ...so a bit of baggage comes with our experience of using, reconciling, merging, cleaning, etc. Yes, to your point, our collaborative tree came with a tad bid more of "weeding to do's" than a Wikitree starting from ground 0.
While I understand and strongly respect your frustrations, I focus on FSFT as an infant platform and see most users coming to the table with a spirit of "doing the right thing" based on their understanding and skillsets. ..Sometimes when frustrated, I'll send links to those users who I can only guess are just as new as I was at one time, and in some cases to help them understand what a primary and secondary source is. That is life in my FSFT neighborhoods.
....In 100 years, maybe the aspirational tree you seek will be realized as more collections are digitized, indexed, searchable and available to the public....and profiles are more mature, cleaned up, ....users more educated on "best demonstrated genealogy terms and practices", etc. ...for now I am just thankful the platform is available for free to me to share years of research, documents, photos and sources....and am thankful to volunteers, the newbies, the experience and passionate researchers such as yourself, who bring small "adds" to the growing tree.
Hang in there!3
Well said Harvest8
I believe it's more a 'tech experience' issue than anything else. While FS is 'relatively' easy (no pun intended, ha!) to use---it can be super confusing with each 'new' window that pops up. Add that confusion to the excitement of finding a 'scrap' of info about ones ancestor--- and into the source pile it goes. I cannot tell you how many times I've run across a family member attached to someone they shouldn't be---simply because of a common name. To my knowledge, FS doesn't offer a 'shoebox' to store "possibilities"---so info may get stuck w someone---simply because they don't want to lose/have to search for again/just in case....or sometimes...it's just all kinds of wrong <shrugs> When I see it---I fix it. Thats the beauty of it. We're all here to find those connections---and it's FREE. To have this much information---and the freedom to 'claim' it-- is a blessing in of itself. We're all a part of this GIANT puzzle---sometimes ---we need to help someone else find their "pieces" so that we can find our own.
Peace, Love and Puzzles!
I am anticipating there will be some FT users who will have strong disagreements with you on this issue!
My own position is that I hate the clutter that results from having a large number of sources - especially when there can be, say, four or five for the exact same event. However, if they do relate to the individual concerned, it is better to attach them, rather than another user attaching these sources to the "wrong" individual, or creating a duplicate ID for the same person and attaching the sources there.
In many cases, the "duplicate" sources have been produced because the same records have been indexed multiple times. The "source" is often, but not always, the same. For example, in English records, the "same" record might have been indexed from (1) the parish register and (2) the Bishop's Transcript. As they might not contain quite the same detail, it is worth having the different versions attached for the possible extra information provided. What does annoy me is when users attach several completely unrelated sources (say for a census) in a hope one of them might apply to their relative. Or, if they attach a record for a sibling, which makes no mention whatsoever of the person to whom the ID in question relates.
I must disagree with you on one main point, however. I believe the IGI (especially those records relating to the extraction program) is very unfairly criticised. Without it, I would not have found many of my ancestors and, in my experience, the accuracy of contents of IGI records is no better or worse than that found in many recent indexing projects.
In summary, I am supportive of any thoughts regarding omitting irrelevant sources from attachment, but can see no way of not adding items that do apply to a particular ID - even if this does produce an unwieldy amount of sources. To help reduce the problem, FamilySearch could carry out careful checks to make sure it does not carry out "unnecessary" multiple indexing of the exact same records - e.g. indexing each film when there has been multiple filming of the same material. However, you cannot change the behaviour of individual users - if they want to attach sources that are not relevant there is nothing to stop them. But, not attaching sources that do relate to the person / ID concerned could prove counterproductive - in that someone else might choose to attach them (incorrectly) to another ID, if they are just left alone (when found in a "Results" list), by you or others.4
I agree with Paul that the IGI is no worse than any other index out there, and that those citations should be attached to the appropriate profile, to prevent them from being attached incorrectly, or from extraneous profiles being created based on them. This is especially important when the IGI has an error, because left "in the wild", the error will keep getting propagated by people who take the index at its word. (Yes, I know that attaching the index entry to a Family Tree profile is of limited utility in preventing error propagation on other genealogy sites that have copies of the IGI, but every little bit helps.)
Your definitions of "primary" and "secondary" are more usually used for degree-of-confidence scales. The definition of a primary source versus a secondary source doesn't normally consider the clarity of evidence provided, only the circumstances of the record's creation. If it was written down at the time of the event, based on the report of a participant, it's considered a primary source. A marriage register entry that only names the bride and groom and the witnesses, without ages or birthdates or parents, is nevertheless a primary source -- if you can figure out who it's for. If a document was created based on a prior record, then it's a secondary source. An extract of a church register is a secondary source, as are most indexes. And if that secondary extract or transcription was later printed up and annotated in a book, that's a tertiary source. In terms of clarity of evidence, the tertiary source may actually be best -- for example, if it explains an obscure abbreviation from the original register entry.
I believe that there is value in sources of all degree, and all types can and should be included in Family Tree. For example, one of the ongoing indexing projects on FamilySearch is the Hungarian civil registrations that I've spent nearly the past decade finding the long way. This means that every now and then, a bunch of hints show up on my relatives for the indexed versions of the records that I've already attached as images. I go ahead and attach those index entries, without removing the older citations, because this way I get the best of both worlds in terms of the two broad types of source citations possible on FamilySearch, namely individual versus group.
Individual source citations are index-based and are meant to be attached to one and only one profile. Such attachment is shown in search results (with the little tree-stublet icon), and the association between index and profile is used by the hinting algorithms to refine further source suggestions. (I think the Possible Duplicates algorithm also takes the index-to-profile links into consideration.)
Group source citations are image- or link-based and are meant to be attached to everyone mentioned in the source. The great advantage to this type of source is that if I notice a typo, I only need to fix it once; it'll then be corrected in all instances of that citation.
Thus for example my grandfather's sister-in-law currently has four sources: her marriage, the index of her marriage, her first husband's death, and the index of her first husband's death. When I created the citation for her husband's death, I wrote a full transcription of the entry, and attached it to four profiles (husband, husband's parents, wife). The index of his death consists of four linked "personas", and I have attached them each to the appropriate profile. There was a slight typo or misreading in the index of his mother's surname, which I have noted in the index citation that's attached to her. If I wanted to note that correction in the rest of the index citations, I'd need to go to each profile and make the edits three more times. If, on the other hand, I want to add a translation of my transcription, I only need to do so once, on whichever profile out of the four I happen to be looking at.
Yes, there is a world of difference between doubling two sources and doubling thirty sources, but in practice, it's generally not linear: when someone's only mentioned as a parent, I usually only add the indexed version (if the image is freely available), reserving the image-and-index treatment to sources where the person is one of the primary participants. Likewise, if the proliferation of indexes is due to multiple filmings of multiple copies, there is seldom any reason to add image versions of any of them. (One does, unfortunately, need to add all of the functionally-duplicate index entries, for the algorithmic reasons mentioned above.)1
Don't forget that sometimes a repetition of the same source is NOT the same information, whoever was adding the source failed to customize the title. I have seen 10 listings from the same birth / baptism collection, all titled the same, only to find each was a birth record for a different child of that person as the parent.1
Yes, I have seen that many, many times. In order to make the Tree more workable and less time consuming I suggest to limit the number of sources being attached to one individual. Wouldn't it make sense, to mindfully determine upfront to which profile I attach my source? It would eliminate the sorting through a long list of sources that do not really say much about the profile of the person they are attached to.0
I would not advise that at all. When researching people who lived several hundred years ago, the landscape of sources available is changing rapidly. Plus, I hope people (like me) who are taking stock of what they have inherited (or discovered at their church, cemetery, local courthouse, library ...), are trying to digitize, upload and create new sources from this new evidence won't be stopped because that ancestor has reached the quota of allowable sources. If you absolutely cannot scroll past messes, please try to be a good citizen and clean up duplicate sources! And don't be so quick to say a source is without evidence. For someone trying to prove lineage, every little bit can make a difference. I have a couple of ancestors whom I've sourced their every move. Of course I'm trying to answer a mystery on those people, and so many sources is nor normal for me, but you might think it a mess if you looked at those particular people.4
I would like to thank everybody for his/her response to my initial "question". It shows me how many different aspects and opinions there are in regards of "sourcing". I became genealogically acquainted with various regions in Germany throughout the last 30 years, ongoing, and I am glad that FSFT offers the opportunity of sourcing. While, in my opinion, sourcing had been neglected over a long period of time, it has been moved much more into the focus of the FSFT community. It would be nice if "Sourcing - the Do's and Don'ts" would become a category where all its various issues would be discussed and, in the long run, improvement on a larger scale would be initiated.2
So my father currently has 200 sources attached from time when sources were new and all the newspaper articles about him were added. Sources are now limited but I don't know what is the limit to go back and remove them down far enough I can attach the 1950's census...0
There is another angle to this great discussion. One could look at Family Tree as a card catalog of sorts of all of Family Search's holdings indexed by person. From this angle, the tree becomes better as more and more of the holdings become cataloged (or attached, as they say). Most libraries, I think, would consider it ideal for all the books they hold to be entered in the card catalog. Similarly, I would consider it ideal if all of the data holdings of Family Search were entered in the Family Tree, person by person.
An example of what I am getting at is the card catalog of the United States Library of Congress. On can certainly find many, many entries for a particular famous person. This is not a bad thing. It means that, thankfully, most all the holdings--some more valuable than others--have been accurately and completely cataloged. Pray for the day when Family Search's holdings will be so accurately and completely cataloged! Pray for the day! Meanwhile, let's keep on attaching those sources, and let the database engineers figure out the technicalities.1
I've come across one instance of a person with a common name, who was merged with everyone else of the same name. He ended up with 76 sources. There was a message on his page flagging him as probably not correct, and to use care with his data. So it appears, there is some threshold where excessive sources are flagged. And a warning of possible errors is shown.0
Dennis J Yancey ✭✭✭✭✭
probably simply a message from a user - just like you or me - who was indeed concerned and shared his concerns on the page1
I believe the number of sources someone has is conditional based on many factors.
For instance, a father of a large family of 12 or more children or someone who married multiple times, especially if they were born after 1840 and lived a long life will naturally have more sources as more vital and census records were becoming available during that time.
It is quite common for fathers of large familes to have 45 or more legitimate sources such as birth, death and marriage records for all of their children.
I have seen people born in the sweet spot after 1840 show up on every available federal census record (1850, 1860, 1870, 1880, 1900, 1910, 1920, 1930 and 1940), including all their state census records. This alone can give them some 9-20+ more legitimate census records (depending on the state).
I agree that duplicate sources can be an issue, and things like tax records and city directories are more busy work than helpful.0