Family Search org pilot records beta indexing Ancestry com genealogy errors online mistakes
Posted by
the philosophizer
May 8
I recently visited FamilySearch.org’s beta site and what a mess; this entire project needs to be reassessed; it is riddled with errors; natural man made disasters!
FamilySearch.org has begun the process of transcribing the data from their microfilm collection into an online searchable database. Here are the errors that I have spotted thus far:
- Surnames – In cases where a birth record does not contain a child’s surname, the transcriber assumed the child took the mother’s surname name; this is a huge mistake and you will only catch it if you know the child.
- Given Names – Males are assumed as females and vice versa based on the name; the transcriber has assumed the original registrar made an error and/or has assumed the name is gender specific.
- Misspellings – Surname misspellings are being corrected and should not be as the error is often carried over onto other official documents and/or it can become the name they continue using. The corrections should be written in brackets if being done or mentioned as a side note.
- Dates – Dates of birth are incorrect in some case off by years
- Informant / Witness Roles not defined – Witnesses and / or informants to births are being listed as the father; another huge mistake. Sometimes the informant is a friend, the person present at birth, a relative, the midwife or the person in charge of the child; the information could be anyone but we don’t know what their role is becasue this information is not being provided to us via text or a link to view the actual document.
- Certification – We are only given the certificate number but not the letters that go with it (i.e. “55” instead of the entire certificate number “LE55”). The letters are a must as it is the actual certificate number. It is also useful for distinguishing between two or more women having the same name and giving birth in the same year. It also helps to clarify the data found in the indexes.
- Locations - Districts, towns and parishes are not correctly matched. For example, you may see “Spanish Town, Westmoreland, Jamaica” and if you don’t know any better, you will take this as fact.
- The district of registration is being mistaken for the place of birth and they are not always one and the same.
- Missing Documents – There is no image (link to view the actual document) so we are denied the opportunity to see valuable information; date, name, occupation, residence, parents, actual age at christening, parents married or not, mother’s maiden name, witnesses, informant etc; confirm the data transcribed.
I use Ancestry and Family Search (LDS mormon genealogy site) religiously; I could not do without them and truly appreciate their efforts. I am extremely grateful to FamilySearch.org for making their online data available for free; they are not another money grabbing genealogy corporation. However, I am now forced to use them as I use my own site (CaribbeanAncestry.com), with great caution; as long as the information being presented is not supported with its documentation (so I can verify for myself), I must take it as hearsay or an assumption; I cannot take the transcriber’s interpretation as fact.
I have sent a letter to Family Search regarding the errors; I strongly suggest that you do not use their beta site (http://fsbeta.familysearch.org) until they correct these errors and take steps to ensure they will not be repeated.
Please share the experiences (positive or negative) that you have had with these two sites. Also share your suggestions and tips that we can use in the meanwhile to get around these issues.
In the days of old, family histories were passed down over the years by word of mouth without error; I’ll continue to use my mother’s oral history for my family tree research; modern day resources for genealogical research has got nothing on my mamma!
Written by Antoinette Forsythe Copyright © All Rights Reserved, Image Credit: Secret… by Vince Alongi Creative Commons Attribution 2.0 Generic License
Tags: addicted to genealogy, ancestry free, ancestry search, ancestry.com, ancestry.com error, ancestry.com review, birth, book, caribbean ancestry, caribbean ancestry com, caribbeanancestry.com, cd, census, certificates, civil registers, civil registration, databases, death, diaries, diary, discs, donald lindo, donald lindo errors, dvd, error, errors genealogy records, errors in genealogy records, errors in research, family history, family search beta, family search census, family search indexing, family search indexing download, family search labs, family search labs record search, family search pilot, family search records, family search version, family search wiki, family tree, family tree maker, family websites, familysearch beta, familysearch.org, free family genealogy search, free genealogy, ftbeta familysearch org beta, genealogical, genealogical errors, genealogical mistakes, genealogical research, genealogy, genealogy blog, genealogy errors, genealogy errors in research, genealogy forum, genealogy oral history, genealogy spelling errors, genealogy transcribing summarizing abstracting documents, historic newspapers, https beta familysearch org, jamaica genealogy, jamaica genealogy com, jamaicagenealogy.com, journals, lds, lds family search beta, lds records, letters, marriage, messages, microfilm, microfilm records, misinformation, mistakes, mormon, mormon church, mormon genealogy, mormon records, obituary, online, people search beta, research, research errors, researcher errors, search, transcribe, transcribing, transcription, transcription errors, typo, typographical errors, typos, vital records, vital records genealogy, washer woman, washer woman genealogy
Hi Terry:
Your observations on the FamilySearch beta site are quite correct. As one of the indexers I can tell you that it’s the way the LDS set up the indexing forms that is the problem! I’ve worked on birth registrations for Trelawny and St. Andrew, working with the digitized image. It bothered me that the form did not allow for the registration district letters, which as you say, are vital to finding the correct record. If birth dates are wrong then that could be a transcription error, as some of the digitized microfilms were incredibly hard to read. We were instructed to record the father’s name if it was there, but not to assume a surname. A good many of the records were for illegitimate children in which case one would record the child’s given name and the name of the mother. It bothered me that there was no way to record the informant’s name, if that were different from that of the mother or father. Neither were occupations recorded.
Indexers should not have been assuming surnames! However, you should be aware that illegitimate children are indexed under the mother’s name in the microfilm indexes. Still, I wonder why the LDS decided to set up the indexing forms the way they did.
As for other errors in transcription … well, I guess it may well be that some of the indexers are not familiar with Jamaican names, geography, etc., though the work was arbitrated. One of my colleagues, who is also Jamaican, who lives not far from me is also an arbirator and she has told me of some of the errors she has had to correct but not all arbitrators would have been familiar with Jamaica.
Hey Dorothy,
Yes the illegitimate children were under the mother’s name in the indexes and so I went to site the get the full DOB and in some cases the child’s full name.
To reduce errors they need to use transcribers that are familiar with the location and its history, I know I am asking for a lot as volunteers are scarce, however they need to decide if they want to do this quickly or correctly.
I see the transcribers at my research center struggling at times trying to make head or tail of registrar’s handwriting, and what I understand is they are given a list of possible names for these situations, so can you tell me why we are not being given the documents image so that we can verify and /or correct the transcriptions?
Even on my site I am repeatedly telling my posters to upload the documents they are using as sources so the membership can look it over for themselves; make their own interpretation of the data.
I haven’t used these particular records but children of unmarried mothers were registered, recorded and indexed under their mother’s surname – there being no other name to use. You mention that it is a huge mistake – why and what should the transcriber have done? Left the name blank?
However, there is a question over how to record slave entries where the child / mother / parents / couple do not have surnames and I agree that it can be a mistake to impose a ‘surname’ I have a James Grannum born to Polly Kitty – this is not a ‘surname’ but a forename that may later have been used as a surname on emancipation; Polly doesn’t have a surname so I wonder how the transcribers would have recorded this event. I also have plenty of marriage entries where one of the couple does not have a surname. I have to admit that my family history programme struggles with people who do not have ‘surnames’!
I realise that there may be transcription issues and I would like to see all sites have space for ‘alternative’ names and other notes. Personally, I would like to see full transcriptions rather than selective ones since informant & witnesses etc can give important clues.
I believe that the LDS cannot upload the images because they are not theirs to upload and would need permission from the Jamaican Registrar’s Office and maybe the individual churches / chapels to do so.
Hi Guy:
How are you? Nice to hear from you. First of all, I don’t think there will be problems uploading the images as the LDS already has images on line on their pilot site at http://pilot.familysearch.org/recordsearch/start.html#start. The images for Trelawny are up but not indexed. All you can do is browse. I understand from Rex Peterson at Salt Lake that image will soon be uploaded.
I agree with what you say about the selective way in which the records are being indexed. I would certainly not have done it that way! Speaking as indexer, I have found the actual process clumsy, frustrating, and not intuitive. I am also concerned that there are people doing the indexing who have no idea of Jamaican geography, nor do they understand how the records are laid out. People’s occupations have been entered as surnames!
I guess we’ll just have to wait and see how it pans out!
Heya Guy, its been a long time
In regards to transcribing the surnames yes they MUST leave it blank if no surname was mentioned; the moment you write something that doesn’t even exist on an original record, without knowing it as a FACT, you are making an assumption. EXAMPLE of some of the errors…
the ORIGINAL RECORD
Microfilm 1523302, Jamaica Births Register 1899
Father:
Mother: Mary Smith
Occupations: labourer
Child: Gillian Augustus
Sex: Male
Date and place of birth: Grange Hill, Westmoreland, Nov 1 1889
Informant: Present at birth, William Brown
Date of Registration: Jan 15 1890
Entry No. 9
Certificate No. LB123
TRANSCRIBER RECORD
Microfilm 1523302, Jamaica Births Register 1899
Father:
Mother: Mary Smith / Brown
Occupation: labourer
Child: Gillian Augustus Brown
Sex: Female
Date and place of birth: Grange Hill, Westmoreland, Jamaica, New York, USA, Nov 1 1889
or
Date and place of birth: Grange Hill, Westmoreland, England, UK Nov 1 1889
or
Date and place of birth: Grange Hill, Westmoreland, St. James, Jamaica, Nov 1 1889
or
Date and place of birth: Grange Hill, Westmoreland, Jamaica, Jan 15 1890
Certificate No. 123
Now, becasue I happen to know this family I am aware this is all wrong, but to you and others that don’t know this family, they will take the info as is and enter it and pass it on as FACT; in some cases, to their PAYING clients.
The FACTS of a made-up scenario are…
(1) Mary Smith is married to William John Smith, so the child is actually William Augustus Smith.
(2) Smith is the mother’s married and not her maiden name
(3) William Brown is just a friend of the family.
(4) Gillian is a male child not female and in fact the original registrar made a typographical error, his name is William Augustus Smith; if all other info had been correct we would have figured this out.
(5) The child was NOT born in New York or any of the other locations but in Jamaica, Caribbean however, the transcriber is not familiar with Jamaica’s history and only knows of Jamaica, New York etc and so writes that as the location.
(6) Next they have mistaken the date of registration as the date of birth
Now because the transcriber did not write the full birth certificate number here are the problems I will face:
– If I choose to get a copy of the birth certificate I will order the wrong film; 1890 instead of 1889.
- When I do get the correct film, I will have to a lengthy search becasue I have no clue WHERE in Westmoreland the child was born becasue the LETTERS to the certificate are missing
- If I go directly to the Jamaican registrar to order a copy of the certificate, I will be charged not only for the certificate but also for a search; all my info is incorrect.
What is the solution?
(1) LISTEN to those doing the transcribing becasue I am certain they have voiced their concerns as well as those visiting the site.
(2) Let those who are familiar with the location’s history transcribe that area’s records.
(3) They must provide the transcribers with forms that allow for ALL the info on the document to be properly recorded; especially if they don’t plan on linking it to an image.
This is what we should be seeing if this is what is written on the original, then the transcription must be an exact copy
Microfilm 1523302, Jamaica Births Register 1889
Father: ____
Mother: Mary Smith
Occupation: labourer
Child: John William Augustus ____
Date and place of birth: Grange Hill, Westmoreland, Nov 1 1889
Informant: Present at birth, William Brown
Date of Registration: Jan 15 1890
Entry No. 9
Certificate No. LB123
or
This is what we should be seeing if surnames or names are missing
Microfilm 1523302, Jamaica Births Register 1889
Father: ____
Mother: Mary Smith
Occupation: labourer
Child: ____
Sex: Male
Date and place of birth: Grange Hill, Westmoreland, Nov 1 1889
Informant: Present at birth, William Brown
Date of Registration: Jan 15 1890
Entry No. 9
Certificate No. LB123
Notes:
____ = name not known / surname unknown / not recorded / child not yet named
—————
Why are there so many errors on these sites, because WE are all willing to keep paying to use them. Granted the LDS is free, however if we don’t make enough noise, why would spend thousands of errors to revamp their site.
Queenie you made me roar with laughter; yes, occupations as surnames I have seen that as well; what ever happened to common sense?
When I wrote this blog many thought I was exaggerating or making things up to make a point, some thought the errors were isolated to the Jamaican records; it is not and many are slowly coming forward with all types of complaints; transcribers and users alike.
I feel bad about complaining because:
- They have to deal with poorly written documents
- Small fonts
- Blurred images
and so on.
I’m still gonna complain
but I so get how some errors can occur; minus the surname / occupation thing LMHO
Terry, you make excellent points, but I don’t know what the answer is. My first complaint was that we were not recording the complete registration number with the district letters. The numbers go from 1 up to maybe 1,000 (later maybe 10,000) AND THEN START ALL OVER AGAIN! That’s the reason for the district letters. If the indexer enters the wrong date then you can really screw up the record.
I have found the indexing to be really, really frustrating. If you make a mistake it’s almost impossible to correct. Really important stuff is just left out. I’ve seen early records where — because of political correctness — no room was made for recording colour or whether the individual was slave or free. These are important genealogical facts! How can you leave them out?
I’ve tried to point this out to Rex Peterson, but I’m getting nowhere. The thing that kills me is that they’ve given the Jamaican records the EASY description, so of course indexers with no knowledge of Jamaica do them and screw up, as you pointed out with the geographical mistakes.
I’m going to see if I can find a way around the problem; tips and tricks till they fix it. I will pass on whatever I come up. Try sending Mr. Paterson this article and see how he responds; if he does nothing then that is too bad; eventually we will all find other resources.
I revise my comment – you are quite right – the transcriber should record as seen – in the absence of a father on the entry it is correct to assume that the child would have the mother’s surname – but not the informant’s! Adding information should not occur and is prolific on all genealogical websites -I’m sure that many use drop down fields for locations. One example I saw recently was a 17th century will for Washington USA – instead of Washington, Durham which is recorded on the first line of the document – they got everything else right – name, occupation, date of probate etc so why make up a location?
Hey Guy,
A transcriber‘s job is to transcribe; make an exact copy of the original; basically copy/paste. A researcher you, I and all genealogy hobbyists is to interpret the data.
As long as the child’s surname is not given and no father is named the transcriber must leave it blank because there is no way to know if the last name of the mother is her married or maiden name and becasue that is what the original document states.
I did see a few new additions that have the full certificate number so FamilySearch might be hearing me scream or perhaps that transcriber simply decided to input the entire number.
The problem is not the transcribers but FamilySearch.org I just went to their site to see what guidance is being given and the first thing I see is this, “Indexing is simple. You can help… No special skills are required.”Are they kidding me?!?!?
Then it allows the transcriber to make corrections in certain cases. Why? They are transcribers! We can all see what mess has been created from these “fixes.” If these transcriptions are going to be coupled with the actual document why on earth are they fixing anything?!?!
An index is supposed to be a reference point that facilitates one in locating the data, so even though the documents will eventually be attached on their site, their corrections will prevent us from finding the data; the search results will say “not found” because the index has been manipulated.
Copy/Pasted from their FAQ page
Problem
“While indexing, I found information that the enumerator or clerk appear to have recorded incorrectly on the original document. Can I fix it while indexing? Clearly there is a mistake, some wrong information was recorded on the original document.
Examples:
1. On a death certificate, the clerk recorded the death date as February 30. The attending physician wrote that he last attended the patient on February 3. February does not have 30 days.
2. The state name or other items are clearly misspelled.
3. The sex does not seem to match the given name and relationship.
Resolution
Generally the answer is no; you should simply type what you see. The indexer is to make a record of the information on an image. However, there are exceptions to this rule, and such exceptions will usually be noted in the Field Helps. These helps will guide you along as you index each field in a document. Also, realize that the rules will vary from project to project, so an exception for one project may not apply to the next project.
Remember that the index will be coupled with a digital image of the original document used to create it. When researchers use FamilySearch to find an ancestor, they will be able to see that original document and be able to make corrections as they assemble their own family history.
One exception that will not be noted in the field helps is the following:
On projects that give both a given name and relationship, the Sex field can be interpreted using the combination of that information.
If the Given Name field and the Relationship field both contradict what is in the Sex field, then indicate the correct sex using the information given. For example, if the given name is Rebecca and the relationship is Wife, but the sex is M, type F in the Sex field. Make this change only if both the given name and the relationship are an obvious contradiction to the sex.”
This certainly explains all the incorrect dates I have been seeing; February 30 could be 3, 13, 23 and so on; how on earth can they know what is was supposed to be. What if the error is February?! Transcribers are not supposed to be doing any thinking; let us figure out what the date is supposed to be.
Secondly there are so many girls that received their father’s name in cases where no male heir was yet born, or simply becasue it was the name passed on or just becasue daddy wanted it to be so. As well many men have female names; Tracy, Kim, Shannon, Sam (short for Samantha), Kasey.
I can not believe they are telling people to make corrections; unacceptable, at that level of professionalism absolutely unacceptable! Leave those types of errors to people like us; the regular folk.
This is copy/pasted from my Site’s Terms
“Please note that some records have typographical errors, misspelled names or inaccurate dates, CaribbeanAncestry.com did not correct any of the information. It has been transcribed as we interpreted to preserve the records’ integrity.”
This has been my policy since I created my Site and that won’t change.