I am indexing Military records. The image shows, "last day of February 1919 for the date. Do I index the day as "Feb 28" (as it is an odd numbered year this would be correct), or "BLANK"?
In the instructions on one of the previous projects we were instructed to put the date for the last day of the month when the roster read "last day of" so I have continued to follow that instruction and I would put Feb 28.
I just did a quick check and I no longer see that instruction but I assume you are safe putting the date.
Then you'd have to figure out if it was a leap year or not, in which case it would be 29.
Every leap year is an even number, so 1919 would not be one. Here is the leap year story per a web search.
"Every year that is exactly divisible by four is a leap year, except for years that are exactly divisible by 100, but these centurial years are leap years if they are exactly divisible by 400."
Interesting information, John Empoliti! Something to remember.
It is, Anne. I knew the first part and that there was more to it than that but didn't remember the century details. Now I know - at least for a while.
lol for awhile is same for me! Of course when the need arises to remember it then there is at least a chance we will!
Wow, whatever happened to entering just what's on the document? I don't know if we're supposed to make it difficult by researching things like that, are we? Most people aren't going to do that.
I know. But, we are, in a sense typing what we see. Until the indexing example a few parts ago for this project that used that substitution I blanked the day field for a case like this. For this project the century issue doesn't arise, and the Researchers can do their own substitution if we blank it. I suppose the Moderators can check what the current Project Managers want done for this part of the series.
Something ive done when i come across those types of dated roster rolls is to just look up, "what was the last day of febuary 1919" and it brings up the calander and everything. It didnt feel right to me in my opinion to just put
so I looked it up and then put the date. This is just me personally, so I hope this is ok.
We are researching this question.
@annewandering, Hi, lots of the Muster Rolls I've indexed and reviewed have "last day of the month" language. Regardless, with the other 11 months, we know its 30 or 31 -if we aren't sure, we look it up like @Colin Yamamoto_1 suggested for February, so why are we making this so difficult? 😎
It is a question of following the contract with the owners of the document. The best answer I have been able to find is to mark the field blank as barbaragailsmith1 has reminded us. It might be frustrating but we need to remember that we just index. The researchers get to make the conclusions.
Thank you all for this informative conversation. It has been enlightening.
It really is not about making things more difficult, but, actually easier. It is also a matter of setting a precedent in indexing. If they allow one to calculate the "last day of the month", then it may give people the idea that they can calculate dates for things like, "died last Tuesday" or "burial tomorrow" or "lived for 3 days", etc. I do remember the project that Larry mentioned where they said to index the day and was aghast to see that diversion from the general instructions to only index what is on the document and not calculate dates or ages.
Really good points, Melissa.
As Colin said above, just look up the calendar for the year, a simple search like "Feb 1919 calendar" and you'll get an image to look at. I've done it dozens of times for other things.
So you honestly think the record custodian is going to take issue about recording a known fact? The fact is the the last day of February 1919 was the 28th. In the same way that if the document had included a date as "the first day of February 1919" no one would have hesitated about recording this as "1 February 1919", it seems obvious that a similar decision should be made in this case.
@Melissa S Himes
I'm afraid I don't see how your analogy applies with this example. Of course there can be interpretation / ambiguity over things like "next Wednesday" - if you are speaking on the preceding Monday, do you mean in two days time, or the Wednesday of the following week? However, there are no two-ways of understanding what the last day of February 1919 is - it was the 28th, and could not possibly have been anything else!
I think you know my feelings about "project instructions" from my previous posts. They need to be seriously examined by an experienced genealogist, or - at the very least - have some semblance of consistency between them. Currently, they are leading to all manner of problems once the indexed records go online and are even proving to be a hindrance rather than a help in assisting users in finding their ancestors' and relatives' records.
I know that you don't like the way we index, @Paul W - but, we must all sing from the same hymnal or there is chaos. That is exactly why we have consistent instructions. In this case as in most, the field help tells us: If the day was not recorded or was written as a variation of the word "unknown," press Ctrl+B to mark this field blank.
The day was not recorded, thus, it is blank.
It is kind of interesting that they chose February to discuss how we index incorrect information:
Here is an example of a miscellaneous record error:
On a death certificate, the clerk recorded the death date as "February 30," and the attending physician wrote that he last attended the patient on "February 3." February does not have 30 days. Can the indexer conclude that the correct death date is February 3?
In general, type only what was shown on the document. The purpose of indexing is to create an accurate record of the information on the original image, mistakes and all. However, rules vary from project to project. Check the field help, which will give any exceptions to this general rule.
Thank you for your response, but as has often been illustrated, the project instructions are not always consistent - take the example of some instructions saying to index an event as a "Baptism" and others as a "Christening", when (in England, at least) these nearly always refer to the same type of event.
In relation to dates, what if a date is recorded as "xxviii February" - would you index that as "28 February"? If you think I'm being silly in asking that, who knows - some indexers might choose to leave it blank, as it does not literally say the "28th"!
There is also the issue (in some older documents) of a date being written as "in the seventh month of". In the 1600s this would almost certainly mean September, not July, so I wonder how you get around that issue?
On a "lighter" note, I too, have come across a parish register entry dated "30th February" - it followed all the other February entries, so I would certainly be confused as to how to record / index that!
I'm not sure if you index, but, records are being indexed under the Entry Form titled Baptisms/Births/Christenings. There are fields for both the birthdates and the baptism dates with project instructions as to what to if both dates are not on the record. I have never seen a project where we indexed a christening differently than a baptism. Can you refer me to a project where this was the case?
I think that on projects where roman numerals are known to be used, they usually have an instruction to record the roman numeral as an arabic number. Unfortunately, I can't recall one that is current with the instruction.
In those projects that might have documents from the 1600's, there is usually a project instruction:
Some months may have been written as a number, followed by the letters "bre." These months should be indexed as follows:
Unfortunately, there is a current project nonconformist project where indexers and reviewers are fretting over how to index the months. The project instructions do not include this instruction and the field help says to index the month 12 as DEC.
There are projects that allow one to use the language to determine days, like this one from Italy:
The word "ieri,"sometimes written as "jeri,"means yesterday, so if the day was recorded as "13", the day would be indexed as 12. If the day was not recorded or was written as a variation of the word "unknown", press Ctrl+B to mark this field blank.
That being said, the muster roll project does not have any instruction to use the "last day" to index a number. We should follow instructions.
N.B. Here is another very cool instruction that I read this morning while following up on another indexing question.
So, as you can all see the instructions are becoming much more precise than they have been in the past. If they want you to use a different calendar, or make a diversion from the General Indexing Guidelines, they will most likely let you know in the project instructions.
Thank you for advising these interesting points. No, I'm afraid I don't index, but obviously am affected by the nature / quality of the work once these indexed records appear online.
From what you say, many of the problems might be happening in a process following the indexing itself. For example, the (seemingly arbitrary) recording in a source of the event being either a "Baptism" or a "Christening" would appear to be the action of another team - who is responsible for adding the records to an online collection. Perhaps this also applies when a whole set of records has been incorrectly standardized - i.e. not connected to the indexers' actions at all!
Apologies to the original poster for drifting so far "off topic", but the overall process seems rather flawed at present and it is frustrating that requests to review indexing, and the "post indexing" process cannot be passed to a person / department that could review the issues that are currently making our work more difficult than it should be.
My comments at https://community.familysearch.org/en/discussion/comment/418728#Comment_418728 illustrate some of the problems I am encountering: in this case 5 out of 5 recently updated, indexed collections with issues - probably few or none due to the actions of the indexers themselves.
We encourage our guests to index the records. If you feel they can be indexed more accurately then please feel free to start indexing to improve the quality of the indexed records.
The two reasons I would not consider this are:
(1) I would not have the confidence I could perform a good job at it. Instead, I prefer to volunteer my time in adding records to Family Tree, which I am sure has proved far more useful to others than any indexing work I might have chosen to undertake.
(2) I guess I am just too much of a "rebel" to be able to conform to any project instructions that I would feel to be counterproductive to the purpose of the exercise. For example, one project I am aware of gave instructions to carry on indexing of records under a particular collection title, even if they were found to be completely unrelated to that (geographical) area. I doubt very much if I would be able to contact the project manager to discuss how unhelpful such an action has proved in enabling users to subsequently trace their relatives' sources.
I have a great deal of admiration for the work of individual indexers, but not for the idea of following project instructions to the letter, no matter the negative consequences.
When indexing I have come across 'last day of' I have found it is not possible to type that as the day date field (in all the indexing I have done) will not take letters that field will only take numbers. Hope this helps.
One thing that I think we may forget is that Indexing is to make records searchable. Getting a name and related event that a researcher may be able to find is our purpose. The researcher will be able to look at the image and make the additional corrections/assumptions to add to their personal records of that individual. Trying to make Indexing an all encompassing "expert" source on every detail is not what we are here for. When we keep things in the proper perspective, it simplifies things a lot.
@judyannehenry1 , Hi and thank you so much -- your comments about making records searchable are perfect for what I think so many of the rest of us were/are trying to say about indexing and reviewing all these precious records. Bless you. Mary😎
Very true and thank you. Today I for the first time I came across 'this day' for date of birth instead of the date.
Whilst not disagreeing with the general direction of your comments, those expressed in one sentence is not always the case:
"The researcher will be able to look at the image and make the additional corrections/assumptions to add to their personal records of that individual."
In many cases the problem is that researchers are unable to view the original image (say when it relates to a FamilySearch "index-only" collection, where there is no direct access to the original from elsewhere). In such cases they are completely reliant on the record having being accurately transcribed.
In such cases it is very important to give as much detail as is available / allowable when making the transcription. So if, for example, the record reads that an event took place of the "first day of February 1819" it would defy logic to not record the date as "1 February 1819", just because the date is written in words instead of numerals. As it is a historical fact the the "last day of February 1819" was 28 February 1819, I really don't see why researchers can't be advised of that fact, as the position (use of text instead of numerals) is of absolutely no difference, whether written as "first day" or "last day".
In summary: in many, many cases the researcher just has no access to the original material, so it is especially important they should be provided with as much available detail as is allowable under the terms of the contract. It is being just pedantic to the point of silliness to hold back information that, as in this case, is not open to interpretation, but is merely being expressed in an alternative manner (words instead of numbers).