Would you find it useful to have an index of all WikiTree profiles by date of birth?

Question

Hi WikiTreers,

A comment by John Atkinson in a discussion on the AutoWikiBrowser bot made me think of this.

We could have an index of all WikiTree profiles born before a certain year organized by date of birth.

This might be a way for ambitious Arborists and other WikiTreers to find duplicates. You can sort surname pages by date of birth but with early profiles the surnames aren't reliable.

It might also be used by those who just want to improve important profiles. Early profiles should be important ones, but they're often junk.

What do you think, would you use this? How?

Chris

Comments

Hi Chris,

Agree.  Understand the technical constraints.  I had perceived your list as being a straight "date order" from 0 to 1000 (Birth / Death) or wherever the "funny" names cut off --

Name (Hyperlinked to profile) then Parents; Wife; then whatever seems useful and will "fit".  (children maybe)

Most items  like the profile manager, edit dates, profile lock, etc., I would venture to guess are rarely, if ever used at a list level and can be accessed direct from the profile page anyway.

Although further thinking about this - if you were to do a range call (i.e., Date from/to; and/or surname, etc., that would be really helpful in identifying duplicates.  (sometimes names and or dates are not a match, but a range would catch that).

Just a thought. (the "compare" works -- but you have to identify parents/wife/etc., (same with Surname call) and the only way to do that is by "opening up each one" you think applies, then going from there.)  At least with the surname call, you can kinda pinpoint who might be "family"

Regards,

---

Would really be a great help for many projects I think, finding duplicates isn't always easy with the many variations in Last names and patronymics, so comparing them by date (and maybe place/ country if that's possible) could be very helpful .

---

I do hope this is in the works.

---

Me, too...

Answer 1

It would be helpful to have the spouse and/or parents names displayed.  Seems to take too much time having to open each profile to see who/what is there.

Regards,

Comments

I would love this...

---

Sandy and Dawn, where do you mean? On this theoretical new list, and/or on search results, and/or on surname index pages, etc? In many places, we cannot display everything that some members would like to see for technical efficiency reasons.

---

I think anywhere and everywhere would be nice...

Ok, seriously, I meant in the surname index, but search index would be nice too.

---

I second this - would love to see it in the search index that we now have.

Answer 2

Thanks Chris, as you say with some of the very early profiles on Wikitree (from year 0 - 800 for instance) they can be known by many different names - both first names and LNAB or CLN, and being able to have an index or even some sort of search function based solely on dates would be an invaluable addition to other search functions on Wikitree.

Answer 3

How do you know what list will be useful in the future? You can't make every possible list. What you can do it properly categorize individual entries. I can't help but point out that categories make lists automatically.

Comments

Once you have a category, the list is automatic, but adding and removing categories is not automatic. However, it could be, if they're based on database fields.

This is one of the ways in which WikiTree is different from Wikipedia. We layer database fields on top of the simple database used by Mediawiki.

I see no reason to ask members to add, edit, and remove two things instead of one when they add, edit, or remove a date.

Answer 4

Don't see that making such a list separately would be very useful except for really early stuff where the number of people gets smaller, otherwise just too large to be any good.

One improvement that could possibly be put in right away is when you do a search for particular names, have the profile number show up as well as the dates and manager.  Right now it does not show the profile number so finding a specific profile can be a pain with the more common names.

Comments

"Like".  I like displaying the profile number, because that's actually what makes a profile unique.  (And it would be nice to have a "Like" button on G2G, especially for those of us who use it a lot on Facebook!)

---

Jack, Don't we already have the equivalent of a "like" button in the up-vote capability?

---

Gaile, in a sense, yes, but on FB one can see WHO it is who has liked a particular comment.  That's a feature I like!

Answer 5

In addition to the uses you've already named, I could see using such a list (especially if I get to select the date range) to look for specific people that I know were born in that range but I'm not sure about.

Answer 6

Yes, I think it would be very helpful. NNS has difficulty finding duplicates with all the spelling variations.

Also, an index list for place of birth would help projects. It's hard to search for profiles that fall under a specific project.

Answer 7

Chris,

Such a list might have some value when looking for duplicates with wildly different last names.  To be useful some method of handling the host of one year differences in dates will need to be developed.  The one year off problem applies to not just those Jan.- April dates in the 1600s that we are so familiar, and see as 1652/3 etc., but to many draft records for WWI and early census records and I am sure many other records where someone subtracted wrong.   This suggests that some fuzzy logic will be needed for date ranges - when that is done - will the number of profiles be small enough to make the list worthwhile?  No idea.

Answer 8

I like this idea in theory, but the usefulness of such an index might be diluted by all of the profiles that use false estimates. And this goes back to why I think it is a mistake to force us to estimate birth or death years for new profiles in the first place - it's easy to build a bad structure on faulty foundations.

Comments

As I have said before - there have been plenty of instances where I have known the names/marriage dates of mothers and fathers, but have declined to create profiles for them until I have more source information. This is because I don't want to post my blind guesses for their vital dates, nor have someone trust my bad guesses and build whole flawed genealogical edifices off of them.

---

Having a date range to work with when browsing search results is indispensible.  You do not have to and should not connect profiles based on date estimates.  Look at other evidence.  As long as estimates are clearly labelled as such, they are important to keep.  If you think people make bad links with these dates, just watch what happens with no dates!

---

I don't trust anything that is not sourced, especially not someone's guess.

---

I've seen a lot of profiles where the creator clearly made very arbitrary (and often unrealistic) guesses on dates -- for example, giving several people birthdates of 1600 and death dates of 1700. I imagine this is done to get around the requirement for dates.

---

This is exactly the sort of thing about which we should be concerned.

Answer 9

Sounds like a great idea, but you wouldn't need all 2015 years at once.  Currently I've been frustrated using an index for a specific name and surname because it only displays the first 100 names; if I'm searching by birthdates, that's the 100 oldest.    Great if my target is in 1400, but not so great if the one I'm looking for is in 1850.  

It's already possible to search by full name, or by surname only.  I'd suggest a change where you could also search by date only -- and get the first 100 (or 200) names beginning with a certain date.  So I could enter just 1450, and I'd get the first 100 names in the WikiTree database born after 1450.  Or I could put in "John Harris + 1525" and get the first 100 John Harrises born after 1525.  Mostly we know within a century or half century what we're looking for, so anything outside the range is simply clutter for a particular search.  

Since place is also in the database, it would be helpful to put in place and date.  So if I put in "Shrewsbury, England, + 1625", I'd get the first 100 people born in Shrewsbury after 1625.  If the person I'm researching lived in Shrewsbury, that would give me her contemporaries -- many of which might also turn out to be family relations.

Comments

That's what I would like - place and date - so I can narrow it down to relevant matches (potentially).

Answer 10

From an Arborist point of view it would help greatly with the standard problems we have with alternate spellings of the last name....just the Andrews/Andrew/Andrewes, the simple ones....and this happens more with 1600-1700 when they were immigrating to the Colonies and many did not know how to spell.   My personal family Wallis/Wallace would be helped also.  

I do believe being able to "Bracket" the dates by 25 or 50 years would help keep the list manageable.

Great thought!

Answer 11

I agree with Robin. Having a 25 - 50 year search bracket would work well. I don't know what cutoff everyone has in mind, but pre-1700 seems like a good place to start. Also, would the resulting list be able to be seen with the names in alphabetical order?

Answer 12

The most important tool for pre-1600 profiles would be to have Wikitree generate and allow download of a Gedcom of all the ancestors of a certain person. This would allow finding duplicates be matching positions in the trees of two purported matches.

Comments

Hi James. What's the obstacle you're running into on GEDCOM download? Chris