Proposal: One more option for place field status buttons: 'unknown' (or similar)

Question

This proposal comes in response to the new/upcoming Data Doctors suggestions on the birth or death place fields. Another proposal has been made to enter some kind of standard phrase in a place field when the editor cannot even be sure which continent the person was born (or died) in.  I think that using the actual field to explain that research has been done is not the best idea.  This is what the status radio buttons are for.  We have a few for the location fields: "certain", "uncertain", "blank for extra privacy" and (for the death place) "intentionally blank because still living".  The proposal is to add one more button, which could be "unknowable", "researched but unknown" or something similar (feel free to state your preferred wording below)*. I've been told that this would be pretty easy to implement but community support is required.

Please upvote the 'agree' or 'disagree' answers below.      

* Wording ideas so far: researched but unknown,  intentionally blank: evidence lacking, unknown, unknown continent, continent unknown, not found, undetermined, no current documentation, unascertainable.

Comments

I agree with "unknown" but disagree with "unknowable".

---

Ian,

After having read through the other threads, I've come to some conclusions. First: I saw some of the specific issues mentioned. Difficult? Yes. Complicated? Yes. BUT I really don't want the tail to wag the dog. Second: I agree with identifying profiles lacking either a birth or a death location and I tend to agree with Ros, that one or the other should be REQUIRED before a profile can be created. Third: I don't believe either "Unknown" or "Unknowable" are acceptable alternatives. If a person is unable, with the tools wikitree has available already, to place some location (Country, Region, Continent....) using the most likely logical place, marking it uncertain and then explaining the uncertainty in 'Research Notes' etc, I question the value of even creating the profile...perhaps place a 'Research Note' on a connected profile until you feel ready to commit instead. Finally: Echoing Emma's note about how this lack of even the most basic information complicates 'duplicate avoidance', understanding how much Wikitree is growing (37 M +!) and seeing things like 17899 profiles for John Smith's makes SOME location a necessity. A John Smith profile with no location is a disservice to all other Smith researchers, requiring them to spend their time ruling out possible matches when this most basic data is missing.

---

If we are supposed to enter a continent when that is the best we can do, then this button would represent 'Unknown continent' and although not a single word, using this as the label would remind people about what choosing this button indicates.

---

Good idea, Andrew. If "Unknown continent" is too reminiscent of Atlantis, then "Continent unknown" could be considered.

---

Nick Andreola. If you want to suggest that a birth or death location be required, you need to start a new thread to propose a policy change here in G2G. What I would say is that I see profiles where we can locate someone at some point in their lives - perhaps when they married, perhaps some other event - and we can determine some relationships for them, but we have no evidence of in what part of the world they were born or died - or else whether they were born or died in, for example, America or Europe.

Occasionally we may not have any location information. For instance, all we may know is an immediate family relationship, with enough to enable some guesstimate of a birth or death date, which may be just before or after year xxxx.

Further discussion of your suggestion would be off topic for this thread.

---

Nick there are times when continent is not known. I just worked on one of those this weekend. Did she die in North America before the British deported her family?  Or did she die at Sea or did she make it to Europe before she died?  Fortunately we know where she was born. But it does illustrate legitimate reasons to not know continent for some events.

---

Ian, could you add Michael's suggestion "intentionally blank: evidence lacking" to the list? It's worth considering, for the reasons he gives. It isn't any longer than the already existing "intentionally blank because still living" for death date, so it would fit.

---

"intentionally blank: evidence lacking" would fit my concerns about suggestions for no birth or death locations.

As shown in my answer on the original discussion on this topic

https://www.wikitree.com/g2g/1695136/new-suggestion-on-the-horizon?show=1695184#a1695184

There are occasions when no sources can be found for birth or death locations, there can also be sources for locations at other times in the life of the person that do not provide any evidence about birth or death locations.

Answer 1

I agree with this proposal.

Comments

How about "not found" for the label?

---

I would hope people would not just select the button without actually doing some research. 'Undetermined' possibly.  I don't think 'unknowable' is good since that indicates it can't be determined.

---

It's a bit long, but "unascertainable" would be good. Not "unascertained": that might just mean you didn't try to look. "Unascertainable" means you looked and confirmed that no location or source can be discovered.

---

I would favor a phrase such as "no current documentation."

I feel this phrase would indicate that one is aware of the situation and has done some research.

We should understand that birth (I include baptism as option) location and death location are specific to those events. As such, we should understand that current documentation for those events may not be available yet - or even exist or have existed.

Data fields should be devoted to information that has been found, not to speculation or to explanation. Speculation/explanation belong in Research Notes.

---

I would like a button. Currently we are using a category but a button would be faster and easier to use and see.  I would like it to indicate that research took place, otherwise people might take a blank to mean to check the button. Perhaps Researched, Unknown would do that.

Answer 2

I disagree with this proposal.

Comments

Jim Patterson:  Thanks for the best answer star, but I don't think that's how this kind of thing works.  

Would you like to expand on why you disagree with this proposal?

---

Flags like this make no sense to me. Clearly every person has a birth and death location, so there is a valid answer, it's just that no one has found a source that identifies it. That information can be put in the Research Notes section of the bio.

If the objective is to avoid Data Doctor flags, that should be dealt with in their reports. There's already a False Suggestion response which is supposed to suppress the diagnostic on future reports. Can't it be used here?

(Sorry, didn't know how voting worked - I've upvoted now).

---

I guess that for some people, a lot of research has been done, and as, due to lack of records, we simply cannot know everything about everyone who lived a few hundred years ago, we should be able to say that in some cases, despite everyone's best research, we do not know where this person's birth/death happened. For many, we can make assumptions, but there will be some event locations which are, for all intents and purposes, unknowable. Part of this proposal (or the next one anyway) will be the Help text which will stress that this flag/option is to be used only in these rare cases and that attempts to find the information should be explained in the Research Notes.   

I think others have stronger opinions than myself on the Data Doctors suggestion part of this.  I think the suggestions are suggestions.  If you know that the automated suggestion is not really valid on a profile, you are free to ignore it (or set it as false).  But I do think that there are cases when we can flag an event--like a birth location--as something like 'researched but unknown' in order to show that this isn't something that has been left blank by accident.

---

I initially thought I can't really agree to this as it would be far to easy for someone to use that option without doing the research. Though having said that, they would just leave it blank anyway. So I unvoted my vote :)

I'm not going to vote either way.

---

Chase, encouraging location information which is just a guess goes against current WikiTree policy, and would require a policy change, which would have to be proposed and debated in the usual way for policy changes. 

For current guidance, see

•  https://www.wikitree.com/wiki/Help:Uncertain_Locations 
• https://www.wikitree.com/wiki/Help:Uncertain.

---

Michael - Your post got me to reconsider mine and I deleted it. I'll post a new one.

---

I disagree with the proposal.

Per the current policy, almost any location (even a speculative location) is better than no location. You don't have to be "sure" where an event occurred; you just need to have some reasonable basis on which to speculate as to where it occurred. The situations where there is insufficient evidence to make a reasonable speculation as to even the continent that an event occurred on, are exceedingly rare, perhaps non-existent. Please provide examples.

---

Examples, Chase, exist. Here are two situations as examples.

1. There is a record of someone in North America or the Caribbean in the 17th century. There is no record to indicate where they died, and there are no records of family members which cast light on this, and there is nothing else to suggest they probably settled permanently in North America or the Caribbean. We know that a lot of people went to North America and the Caribbean in that century for a period and came back to Europe. In these circumstances it would, in my view, be unwise to enter a location in the death location field.

2. There is a record of someone in 19th century India, where they were, or a family member was, for a time a British official or in the army. We do not know whether they came - or returned - later to Europe, died in India, or went - as some of my collateral connections did - from India to eg Australia, New Zealand or South Africa.

Where there is a reasonable basis for entering an uncertain location, fine (as I have recently said several times in G2G exchanges, for pre-1700 profiles that basis needs to include some reliable source about the person). But in these situations it would be a pure guess.

---

… and an example for birth. We have records for someone in 17th century America, but no birth information, and do not know who their parents are. We have nothing to suggest whether they were born in Europe or America. This is one of the more common situations I encounter where there is no basis for giving a birth location.

---

I've run into this birth situation often when looking at 18th century Maryland Germans; there's some people where I know they were in Maryland by a certain date but have no idea whether they were born in North America or Europe and can't find that out with the sources I have access to. (That said, these folks generally wouldn't trigger Ales's proposed new suggestion, as they died in Maryland.)

Nick commented that if we can't figure out any location then we shouldn't be creating the profile in the first place, only putting notes in the relative's profile. One counterexample:

Suppose you have two profiles who the sources indicate are brothers who share at least a father, and you want DNA test information of each's descendants to show on the other's profile. To do that, you need to create a profile for their father. Unknown Surname; estimated date based on the birthdate of the older brother, research notes giving the sources for the sibling relationship. So far, no problem.

But these profiles are for two people who lived for many years in New South Wales, Australia. They're both known to have died in NSW, but it's not known whether they were born there, immigrated as children with their parents, or immigrated as adults. And absolutely nothing is known about their father's birth or death place -- it could be Australia, England, Ireland, India, North America.... You need the father's profile for propogation of DNA test information, and you shouldn't have to wait on creating it until you find out a place where he definitely lived.

Answer 3

Hi Ian. Thank you for this idea. As a technical solution it would be good. It would avoid complicating the data field with a marker phrase that isn't data, and could lessen the difficulties that have been discussed of even choosing what phrase to use, because the radio button label would by necessity be very short.

But so far it's only a technical solution. It doesn't address the questions raised, and partly answered, in the two threads you link to in your post, concerning what the conditions would be on when it's acceptable to use the new status, and how this would be conveyed to members. I hope I'll be able to vote in favour of your proposal, but first I'd like to see how aspects beyond the button itself would be handled.

It's even easier to select a radio button than to enter a marker phrase, so restrictions will be needed. This is something that is only intended for use when not even a continent can be entered without guessing. What guidance would be given on when to use the new button, the necessary steps beforehand, and the provision of research notes? The details aren't necessarily needed yet, but we need at least to know what process would develop them.

Edited to remove a sentence as indicated by striking out.

Comments

A shorter (and blunter, sorry) way of saying the above is that so far this is not a proposal, just a technical improvement request. In order to become a proposal, it needs to explain what style or help page would be modified, and give a draft for the new or changed text. 

---

People select a button because they want to or not.  It is safe to say that the majority of people are not referencing any Help Page to determine why or when they can select a button.  Also, remember, once a button is selected, it cannot be reset unless another button is set and do NOT say that it can be done with WBE, which many members are not using because they are not aware of extensions and / or they do not want to use extensions.

---

If people are not scrupulously following carefully written guidance from a Help page when they select the new radio button status Ian is requesting, then there is no hope of meeting Michael Cayley's criterion: "to have a way of clearly indicating that a blank location field has been looked at and that there is no, or insufficient, evidence to include any place - not even a continent - in the location field."

In order for Ian's request to become a formal proposal satisfying the procedure for Developing New Rules, which Help page section that will be and the draft text for the new guidance need to be stated here.

---

Jim, please suggest some text for the Help page. I also thought that this technical change would not need a proposal, but I've been told that it does. It may be too late to add the Help text to this current proposal, but revision of the proposal is part of the process. After discussion, I'll make a new proposal focusing on the Help text and the label of the actual button. There have been many interesting suggestions for the wording.

---

Sure, Ian, I'll try to come up with draft text for the Help page. As you say, it will be such a big change that it will need a new G2G question and recommencement of voting. It will take me up to 24 hours. In the meantime, it would be good to see discussion here continue, with the aim of coming to a decision on the wording for the label of the radio button.

---

Ian, in my next comment below will be my tentative draft for use as part of your revised proposal. I strongly recommend that you ask Michael Cayley to vet and improve it before proceeding. Be sure to copy the HTML, with links, block quote, and bold. "Continent unknown" needs to be replaced with the final wording chosen for the radio button label.

---

The last two paragraphs of the Almost Any Location is Better than No Location Help page section would be replaced with the following text. (The first paragraph is unchanged, but included here to give context. The words "(see below)" refer to the existing subsequent section of the Help page, relating to use of continent names as locations.)

Our community has agreed that almost any location name is better than no location at all. If you do not know the historically-accurate place name in the person's native language it is better to enter an uncertain, estimated, modern or translated location name than to leave all location fields for the person blank.

The only exception where both birth and death location fields should be left blank is if entering any location at all would be a guess. See Help:Uncertain for more about the line between uncertain information and guesses.

Leaving both birth and death fields blank will usually cause a Data Doctor suggestion (141, 142 or 143). In the particular exceptional case, you may use the radio button marked "Continent unknown" to forestall such a suggestion from being applied.

This radio button may only be used in the case where research has shown that there is genuinely no prospect of establishing and providing a source for a location, even a broad one like a continent (see below). By using the radio button, you are confirming that you have carried out this research. You must document the research you have done, and your conclusion that entering any location at all would be a guess, in the Research Notes section of the profile.

---

And if, as has been posited, the person could possibly have been born elsewhere than a continent - what then?  Using "continent unknown" would be as wrong as guessing at a place.

---

Hi Melanie.  I'm collecting suggestions for the wording of the label.  Jim's "Continent unknown" in the Help text there is a placeholder.  Please let us know if you have good suggestions for the label and/or the Help text. Thanks.

---

Yes, Melanie, on a different thread in the post please, so it won't derail the different discussion here.

Answer 4

Thanks, Ian, for suggesting this. I have given an upvote. This change would be helpful in addressing some of the concerns that have been expressed over the proposed new database suggestions for blank location fields. More generally, I believe it would help other WT members by indicating that an effort has been made to identify a location.

On wording, I wonder about something like "Intentionally blank: evidence lacking". That avoids any difficulties about referring to continents, is accurate, and allows for the difference between evidence standards for pre-1700 profiles (where we require some reliable source as a rationale for entering even uncertain information) and later profiles.

Answer 5

I agree with 'researched but unknown'. It's too easy to click unknown to avoid having to research it.

Answer 6

I would prefer that the process mimic the one for guesses at birth and death dates.  When we determine that one of those dates cannot NOW be determined, we have a template that says the date is "Estimated".  It is quickly added from the Template button above profile edit windows.

{{Estimated Date|Birth and Death}}

I would like to have the same available for location as well.  By providing this option, we maintain some consistency and don't really change either the database or requirements.

If such a process is instituted, Ales could possibly refrain from suggesting those profiles have no location, where the template exists.

If such a process is followed, and later the location becomes knowable and is entered but the template not removed, either Kay or Ales could flag the template as possibly in error. 

The advantage here is that (1) profile editors and managers are not feeling any pressure to do or not do something; (2) we maintain a process that is known and easy; and (3) we can still use the suggestion system to locate profiles needing attention without needlessly or repeatedly going over profiles with unknowable locations.

(My apologies to Ales.  I don't have the keyboard to spell your name correctly.)

- Aunt Peggy

Comments

Unfortunately, ”estimated" does not cater for the situation, which is the one we are considering, where there is no basis to form a view on what part of the world someone was likely to have been born or died in, and anything inserted in the location field would be a guess. We are rightly told not to insert guesses in the location fields. There needs to be some basis which makes it reasonable to include an uncertain location - and for pre-1700 profiles, that basis needs to be a reliable source.

---

In that case, I still prefer to follow precedent in developing a process.  

So, I suggest that "Unknown" be allowed in the location fields in the same manner that "Unknown" is allowed in Last Name at Birth field.  Once again, it is familiar.  It is easy to update when the location becomes known.  Searching can be facilitated to differentiate between an empty field and a field saying "Unknown".  The only coding change required would be to the suggestions -- very simple for Ales.

There would need to be some education to help people understand the difference between "this location is not knowable at this time" and "I don't know right now and don't intend drop everything to work at finding it."

But, I also stand by my suggestion for locations that can be guessed at.

---

Earlier, I proposed the idea of something like "Unknown" being allowed as a marker in birth and death location fields. I still think that should be kept in mind as an option, but at the moment Ian's newer proposal here is getting much more support.

Answer 7

Although, it may need more time to judge correctly, this proposal seems to be broadly supported here.  Before going on to a second, clearer proposal, I'd like supporters of this proposal to weigh in on what we have so far.  

.

1. The label of the proposed option (in order of my preference):

a) intentionally blank: evidence lacking

b) researched but unknown

c) no current documentation

d) continent/region unknown

e) not found

f) unknown 

g) undetermined

h) no current documentation

i) unascertainable

.

2. This draft Help text: 

.

Our community has agreed that almost any location name is better than no location at all. If you do not know the historically-accurate place name in the person's native language it is better to enter an uncertain, estimated, modern, or translated location name than to leave all locations for the person blank.

Birth or death locations should only be left blank when, after research, entering any location at all would be a guess. (See Help:Uncertain for more about the line between uncertain information and guesses.) In this case, choose the status option labeled "{label goes here}". This option should be used only when research has shown that there is no prospect of establishing and providing a source for a location, even a broad one like a continent (see below). To help other researchers, details of the research should be either mentioned in the main Biography text or added to the Research Notes section of the profile.

When both birth and death locations are blank and the "{label goes here}" option is not selected, a Data Doctor suggestion (141, 142, or 143) will be given.  

------

Thanks.

Edited to add placeholder text for the label in the Help text.

Comments

Thanks Ian. This looks good. For part 1, I like Michael's option a) the best.

The draft text in part 2 covers all the points well, and is nice and short. I assume that "researched but unknown" there is intended only as a placeholder, to be replaced eventually by the final selection from the options in part 1.

In due course I look forward to seeing the updated version of your proposal in a new G2G question.

---

Ah!  Thanks. I meant to put a placeholder into the text.  (I've done it now).

---

It will be good if people can please indicate their preference for the options in part 1 above and any snags they see with part 2. This will help Ian's proposal to go more smoothly at the next stage. (This is a boost for the Western Hemisphere.)

---

Ian, I also find Michael's option a) intentionally blank: evidence lacking the best choice.

It explains why there is no location given.

It tells me says that the decision to not enter a location was intentional, it tells me that research has been done and that there is no evidence currently available that can provide a location. 

It includes as an 'umbrella term or concept" options b to h. 

It does not imply that will never be evidence found that could support the use of a location. 

( It only took about 9 years for my own family history group of 5 people to find a birth location and a source for a 4 x GGF)

The phrasing is similar to intentionally blank still living which is familiar to our members.

And I completely agree that a Research Notes section should include information about the research that has been done and that provided no results. All of these record types should have been searched for possible location information, births, baptisms or equivalent naming events, marriages, death and burial records, obituaries, census records and many other types of records depending on dates and locations thought of as possible. 

---

Thanks, M.  As you see, I put Michael's wording at the top of my list of preferred options.

Feel free to suggest any edits to the proposed Help text.  We're not voting on this wording here. That will come after this discussion.  I think it's pretty good as it is--Jim wrote an initial draft; I've edited it, and I've had private input from others--but I'm more than open to further edits before it goes to the second stage of the proposal.

---

Hi Ian. Four people seem to be in favour of 1a, with no expressed opposition.

Unless someone offers improvements on the draft text (part 2 above) in the next few hours after the post returns to recent activity on G2G now, it seems to be time to go ahead with a final proposal. Although the new suggestions have not appeared yet, they might soon, and it would be good to have the mechanism you are proposing in place before then, or as quickly as possible after.

---

I reviewed the draft Help text: I don't have any experience with drafting anything similar, however it appears okay to me.

The current weeks suggestions do not include birth and death location except under Profile Completeness as have been included for quite some time.

.

Answer 8

My biggest problem with the place data fields is that the word "or" (and its equivalents in other languages) is considered a forbidden word in these data fields. Most of the time when I don't know where somebody was born or died, I can narrow it down, to a few places -- and usually a few very specific places, and I think it is misleading to pick just one place. 

Perhaps we are pretty certain that a person was born either Essex County, Massachusetts Bay Colony, or Long Island, New York. It would be absurd to say "North America" because the information is so much more specific than that, but entries like  "Essex County, Massachusetts Bay Colony, or Long Island, New York" don't fit nicely in the data-entry box, and tend to result in "suggestions" to select just one place. And saying something like "Unknown" fails to communicate how much is known.

And I have on occasion dealt with people who may have been born on one of three different "continents," or perhaps at sea while in transit. An example is a person with an uncertain year of birth whose parents lived in the Netherlands, emigrated to Pernambuco in what is now Brazil, and went to New Jersey or New York after a fairly brief time in Pernambuco. The best way to record this would be to list all three places in the birthplace field. I think that saying "Unknown" or arbitrarily choosing one of the possible places is at misleading and rather dishonest.

Comments

Hi Ellen. In the first case, if unwilling to put "North America", couldn't you enter the two possible locations separated by say "OR", and mark the resulting suggestion false? This would be at the expense of not having a little pin linked to a location on a map. But it would allow searching by location keyword.

The second more complicated case, where the birth place is vaguer still and a long series of alternatives might not fit well in a location field, seems suitable for the radio button status "Intentionally blank: evidence lacking" that Ian's work may lead to, with explanation of the different possibilities in research notes.