The top two items on my wishlist for improving searches

+15 votes
179 views

There are two improvements to the name-search functions that I believe are probably feasible and that would have a major positive impact on my productivity and personal satisfaction as a WikiTree contributor -- by reducing the creation of new duplicate profiles and making it less time-consuming to detect duplicates or find additional relatives. Here are these top two items on my personal wishlist:

1. Enable search on the married names of women whose maiden names are unknown.

Two examples of the problem to be solved:

  • Look at the profile for Agnes Curtis (this is a woman's married name -- see her profile by clicking on the link). Now enter the names "Agnes" and "Curtis" in the search box, and look for her in the search results. Spoiler alert: You won't find her. She's not in the results. This isn't a problem with "too many results, because the search finds only a few hits. The problem is that name search doesn't find people by their Current Last Name if their LNAB is Unknown.
  • If you don't like that name, try this with Annie Beebe. You'll get the same results.

Please make it possible to find profiles like these when the first name and Current Last Name are entered as search terms.

2. Teach the search function to overlook the underscore character _ in last names like Van_Dyke and Du_Bois and van_der_Walt when searching for possible duplicates.

  • How things work now: If I'm looking to see whether there's an existing profile for an Abraham Van Arnhem, born about, I can enter Abra* and Vanarnhem in the search boxes, and I'll see results for Abram or Abraham with last names of Vanarnhem, Vanaernum, Vanarnum, VanAernam, VanArnem, and more. However, if I want to find variant spellings for the two-word version of the name (and this was properly a two-word name), I have to search separately for Abra* with each possible variant spelling (Van_Arnhem, Van_Aernum, van_Aernam, van_Arnheim, etc.).

How hard would it be to teach the search function to treat Van_Arnhem as equivalent to VanArnhem, for the purpose of finding possible duplicates?

in WikiTree Tech by Ellen Smith G2G Astronaut (1.2m points)

I'm so in support of this. Many duplicates are created because we cannot find profiles on married names, or aka's, or (as has been illustrated before) when searching in combination with a first name, the result also includes profiles that has (in combination with the entered LNAB) the first name as second or third or middle name as well.

Would it be possible to create a separate field for every version of the spelling of a first name, as well as an aka?

We would soon have to create a sort of database within WikiTree (within our project then) to facilitate the searching of already existing profiles by coupling the validated LNAB-profiles with the standardized (such as in an Access-file I imagine) - people are using dated sources and searching for existing profiles on standardized or names with a slightly different spelling when all the research has already been done and the profiles with names spelled exactly as in the first primary record to be found for a person, not finding them and then creating new duplicates that have to be merged all over again. Just one of the many examples: http://www.wikitree.com/wiki/Gabrielsz-1 ... (I had to merge a duplicate created yesterday; also one of the spouses ...).

Philip, the search function that operates when we try to create a new profile does check all of the name fields, including Nicknames and Other Last Names. Thus, because you have populated those fields in the profile, if you try to create a new profile for Adriana Claassen, born in 1667, the system will identify  http://www.wikitree.com/wiki/Gabrielsz-1 as a duplicate.

Unfortunately, if the person has an LNAB of Unknown, the search function ignores the Current Last Name and (I think) the Other Last Names, and it doesn't match to Current Last Name or Other Last Names that contain more than one word. Thus, if I try to create a new profile for Agnes Curtis who died in 1585, the system shows me the two profiles that give this woman an LNAB, but it doesn't show me the PPPed profile for Agnes (Unknown) Curtis. And if I try to create a profile for Adriana van de Kaap, born in 1667, the system doesn't show me  http://www.wikitree.com/wiki/Gabrielsz-1.

As I understand it doesn't show up either way - people are struggling to find duplicates because of all the name variations, not knowing (me as well) if someone had a patronymic, semi-patronymic or not, an "aka" (that field is totally not searchable - and it includes other marital names, versions in the spelling thereof, and also other spellings of the patronyms if they are known).

Your (Unknown) is just another variation (even more murky because like a LNAB as "Baker" there are innumerable of them). So yes I get what you are saying, but we should also be able to sort / search with a functionality such as this one: http://www.e-family.co.za/ffy/surname_index.htm where different variations of one name are listed and can be coupled back to a single record.

There could perhaps be a functionality where one spelling of any (every used!) name in any field could be fixed as a descriptor and people can search / filter with Booleaan search strategy.

You are more optimistic (or should I say idealistic?) about automated searches than I am, Philip. I don't hold out hope for a perfect solution to the name-search problem. Every system I've seen has some drawbacks.  WikiTree's search function has greatly improved; I only hope to make it better.
I'm a information professional. Finding information is my vocation, have many years of experience in doing online searches. I'm neither pessimistic nor optimistic and certainly not idealistic. But have to confess that I'm being outdated by technological development - rapidly. Always confronted with the limits of technology in combination with my limits to adapt to innovations, but at the same time amazed and surprised by the innovations and the future possibilities. Data visualization is just one of the amazing vocational possibilities - wish I could (have) been part of that ... but feeling old and getting tired ... so next week I'm starting my year long vocational re-schooling from librarian to archive assistant. I'm just so thankful to be living in an age of internet where things are coming together at mind boggling speed.

I hear you, Philip. Search has come a long way since the days when I worked with information professionals who would submit Boolean search queries to a bibliographic database system (do you remember Dialog?) and wait until the next day to see whether the search had generated a massive computer printout or turned up nothing. (And speaking of Boolean queries, I agree that WikiTree definitely would benefit from some additional search filters.)

The inherent challenge I see with name search relates to the diversity of the variations needing to be recognized and accommodated. Name variants can exist for diverse reasons: different spelling practices, variant interpretations of spoken names, cultural differences in the treatment of names, legal changes requiring people to adopt different names, variant interpretations of handwriting, deliberate decisions to adopt a different name or alter the spelling of a name, shifting fashions, etc.  Given this complexity, it's difficult to imagine an automated system that could detect every possible variant of any world name -- and if it did, it would return so many unwanted results that users would reject it.

:-) I was taught with Dialog[ue I forgot how to spell it even - it was really horrible I remember - luckily I didn't work with it for too long] ... en it was just in the years that Google was "being" born ... and in complete agreement with you on the complexities of regarding the "context" of a complicated reality when automation still have to incorporate the human element to make it function. Perhaps you are right in that I'm idealistic = who knows the "algorithm of things" might bring something amazing as the internet, google and data-visualisation sooner than we think ... for me I feel like Moses seeing the promised land and knowing that he will never set foot on it ... so I'm going to end my days like a scribe with my nose in the old texts ...

Note on some related G2Gs:

Thanks Ellen - that's something else that I sometime have a hard time finding back through searches - related G2G threads. I've started keeping separate notes when I find them containing valuable information, but how is the best way to go about it? - though we tag the posts finding the posts / threads back on those tags do not work for me - or perhaps I'm doing something wrong ...

Please log in or register to answer this question.

Related questions

+8 votes
2 answers
+16 votes
12 answers
+5 votes
2 answers
478 views asked Sep 6, 2017 in Genealogy Help by anonymous G2G Crew (310 points)
+6 votes
1 answer

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright

...