Surname suggestion

There is in my opinion an error in Geni which links profiles to the surname-db without respecting accents since (i.e variants) since you can have a surname-entry for all variants.

Since you can use wikitext in the description you can just add links to the variants like it is done on this example:

http://www.geni.com/surnames/Sunde

True, and I have done something similar on other surname pages, but that doesn't do anything but let a reader of that page know those are variations. What I was describing would add a variation section to the page, which could be queried and utilized by the tree itself.

Tried the links on my example?

I'm sure they work fine, but that doesn't truly associate them on a back end database. It's a hyperlink in a text field. It links the user to the other names. That's not what I'm talking about. It can't be queried outside of that page for use in the actual tree. What I was suggesting was a section that would build an actual database table for name variations. Whatever surname you go to, you can automatically see all the variations assigned to that surname. Those variations could be queried by the actually tree for matching purposes and reducing extended display names.

Using your example for instance, Sunde lists Šundić, but Šundić doesn't list Sunde. You had to manually add the variations to each variation page. In addition, this cant' be queried by the tree itself so that the profile of "Henry Sunde" automatically has a variation as Henry Šunde, since the Sunde surname database has all the variations for Sunde. This might help us when we're merging profiles and we have Sunde / Sundic / Šunde. We can choose a primary surname and all others will be listed as variations.

It would also seem to make sense that you could use a common "about" for the variations, so you don't have to explain the information on each page. If they're linked it could create a common description.

Mike Stangel

We will be introducting a "Related Surnames" module to the surname pages.

s/duct/duc/ :-)

Related surnames is OK, but the main problem is that Geni does not respect that in the link from a profile to the surname description and opposite.

As in my example above we have different descriptions on the Sunde and Šunde surnames and have even added cross-links between those two surnames (Related Surnames) pages.

Geni does however list and link Šunde profiles to the Sunde surname description and opposite.

Our search engine automatically strips diacritical marks, etc when matching, so even if we wanted to separate the Sunde profiles from the Šunde profiles on surname pages, we cannot.

Well, make a try, because I would classify that as an error ;-)

Since you allow exact spelling surname pages the link to the profiles should be exact too.

Another issue with names is that there are often alternate ways of spelling them, depending on different traditions and conventions.

A great feature would be to add "alternate spelling" to names on the Geni profiles. One would be set as the "most common spelling" or "main spelling" and would be the one shown in tree view etc.

I know we have the "also known as" field today, but that doesn't quite cover it, as it doesn't tie the different spellings of i.e. the same surname up to eachother.

This type of feature would also provide a possible solution when merging two profiles, where a name has been spelt differently, and both are correct. Instead of overruling one, by choosing the other, one could add both.

The different spelling conventions could then be further elaborated on the surname page.

Private User
The same story.
In Estonian the umlaut (diaresis) ones are usually totally different from others (without umlaut).
It has been already problem that Geni search (for PRO's also) don't make difference between them.

Examples:
Kään ≠ Kaan
Kuur ≠ Küür
Rõõm ≠ Rööm ≠ Room

Jadranka, - it is still the Šunde family.

Norwegians named Sunde will however react if they get linked to the Šunde page, but people which have changed their names from Šunde to Sunde will accept the text in the Sunde page telling people originally from Croatia can click on the link to the Šunde page to get the description.

A different issue I have with surnames is that the data entry fields say "first name" and "last name" rather than "given name" and "surname". My relatives in Hungary enter the surname in the "first name" field since this is how they write their names there.

The match function will not recognise these profiles as being the same, so merges all have to be done manually.

Is there some way to get the match function to check both ways? I would say if two of "first name", middle name", "surname" and "maiden name" are the same (although not field to field), it should still come up as a suggested match.

This addresses situations where people switch which name they generally use (so there is confusion as to which is the first name and which is the middle name) and others who fail to use the maiden name field correctly - we would still be able to see the match. It also addresses the situation where surname has been entered in the first name field.

Private User Your case is localization problem. For Hungarians there shall be setting to display or Lastname Firstame (in Hungarian way - Miklos Jessica) or Firstname Lastname (in international way - Jessica Miklos) . The Lastname (Surname) shall still be always written in Lastname (Surname) box to avoid mixing up.

I'm not sure how localized it is. I know that many eastern European countries do the same. I know how to use the existing fields, so it is not a problem for me personally, but when I look at profiles entered from Europe, the fields are being used as I described. The current method for finding matches cannot identify these as matches.