Introducing the Geni Consistency Checker

Job Waterreus - re: "From what Mike stated you should not conclude "they are clearly only looking for existence of parent, not that the parent has a last name."
You could test it and come to that conclusion, but that is something else."
--- I ended with "If Mike needs a link to understand what I am saying, he can ask, and I will either post it or PM it to him."
because I tripped over one profile where it is the case, re-read what he said (the statement I quoted), and realized that apparently they were not looking to see if the Parent had a last name, just if there was a parent. Hence my suggestion to him.

The rest of your statement, starting with "You should not even conclude" is not referring to anything that I suggested I was concluding, so I am baffled by why you are making that comment --- not sure whether to ask what are you talking about there, or why are you talking about that? But if you could please enlighten me, would appreciate it.

I am working through some things, but have found issues with some of my Finnish family who have two, three or four last names depending on where they were born and later lived. (Different naming system from the more Western tradition.) That said, I am finding some things to correct and am thankful for this, such as sorting children of third great grandfather's two wives. Found correct evidence and cleaned it up pretty easily.

I am a bit hesitant in making changes in profiles that I do not manage but that are attached to my tree. I can and am contacting the profile managers in some cases, but in others, I am hopeful they will attend to the revisions. I am fluent in Google Translator, but it has its limitations. :-)

Private User Quote: "(...) In old days when people died during the winter , they were not burried before the snow was gone. (and it could take some months...) So three months are not enough in all cases".

My comment was not about burial after death inconsistencies, but events in future inconsistencies. These appear when there is an event in a week, in 2020 and so on.

However, you have a point that burial (jordfesting) was often months after the person died. According to Mike Stangel's reply in post #32, the limit is one year after death, which should be sufficient in most cases.

Susanne Floyd I think that Finland has the best school system in the Nordic region, so I would expect that most Finns would understand English very well, although this would not always be so. :-)

No Torbjørn Igelkjøn, that is not exactly what I meant. Sorry for the confusion. I try to read original documents and the translation is difficult for me. It is my issue in most all cases! I have been using the Google translate along with my English in my correspondence to managers/family. Hopefully, having the two versions side by side can make any inconsistencies clear with my translation. They have the best school system in the world according to many and I send out info all the time on it to my education colleagues. I hope to visit soon. I have lots of "new" relatives, thanks to Geni! :-)

Susanne Floyd, sorry for misunderstanding.

Nah, Torbjørn Igelkjøn. I am prone to rambling! It has been a long week.

>>> "You raise a good point about funerals, we'll put a grace period on those as well. What do you suggest? 3 months?"

Mike Stangel, I'd personally suggest a year. There are a number of cultures that hold a funerary event on the anniversary. Also, a lot of New England burials/interments and funerals aren't held until the thaw (when the ground loosens up enough to dig in the cemetery). For example, my grandmother died in early December, but we didn't have her funeral and interment until the end of May. I assume some other cold places have similar habits.

Private User, no special group of curators working on just the consistency checker. Some curators have taken more of an interest in the detail, some have more available time at the moment, some work in areas where consistency is more of an issue etc. Same with all of Geni, Volunteer Curators assist where we can, where our speciality lies, where our interest lies, where we have access to information, when we have available time etc.

It is just a lot easier for us to help if we have links to profiles rather than generic "what happens in this scenario questions". We can look at the detail and compare it to what we have seen elsewhere and maybe save Mike from some time in investigating.

Leanne M (Volunteer Curator - Australia) - on walkabout🇦🇺 - I had addressed someone else with a suggestion specifically to that person - I did not ask for your help there, did not want your help there, -- and you butted in.
Moreover - Your first answer to what you meant by "we" definitely did not convey to me what your second answer (above) is suggesting.
Please do not butt in to my business any more, since apparently we cannot understand each other.

(this consistency checker has me basically wanting to scream and attack - please, suggest Mike look at the comment of mine you first responded to, which I thought and still think is quite understandable as written, and please leave me alone)

Private User if you want to have a conversation with just one person and have no-one else involved - I would suggest using "Messaging" functionality rather than "Public Discussion" functionality - that way you can control exactly who you want to be involved in the discussion.

Just trying to assist.

Lois,

It seemed to me you where reading too much detail from what Mike was writing.
When doing the testing we also saw that you have to be very careful when drawing conclusions on what you see on a test with only a few profiles. There can be hidden problems that influence what you see.

Mike had explicitly stated "you won't get a Missing Birth Surname inconsistency if there are no parents. If there are parents, then the typical presumption would be that this person had their last name as their birth surname. "

From this I concluded you got the impression that he is saying the last name field from a parent is used for the birth surname.

Lois,

There is no reason to react like you did to Leanne.
Like she said we are trying to help.

Leanne M (Volunteer Curator - Australia) - on walkabout🇦🇺 - I purposely posted it in the Public Discussions because I thought he would be more likely to notice it there. But also so all would see the issue. I had absolutely no objection to others joining the conversation. For example, I explicitly said "not sure how it should handle it if Mother has a Birth Surname, no Last Name" and suggestions or comments on that were certainly welcome. As would a variety of other comments.

But what you said was: "if you post a link to the profile we can look at it for you" - that is NOT participating in the conversation - it is basically the total opposite.
And - You did that without defining who the "we" was, or why I might want whoever that "we" was to look at it or what that could possibly accomplish.
I made a straightforward suggestion to him, which required no looking at a profile on anyone's part unless he said "Hey, what you suggest is how it is supposed to be working, can you supply an example where that is not happening"

Moreover, in response to my attempt to figure out what you did mean, you said "there are a group of curators working with Mike on consistency checker"
then later you straight-out contradicted that and said " no special group of curators working on just the consistency checker. Some curators have taken more of an interest in the detail, some have more available time at the moment, some work in areas where consistency is more of an issue etc. Same with all of Geni, Volunteer Curators ..."

Either you are lacking in basic skills at communication, or I am lacking in comprehension skills at the moment - or possibly both.

Job - I was still composing when you posted your two comments.

re: your comment that you concluded I "got the impression that he is saying the last name field from a parent is used for the birth surname." - no, I had not concluded that at all - didn't even consider or think of it.
Thanks much for explaining.

When I discovered I got a "Missing Birth Surname inconsistency" when there was a profile with no Birth Surname with a father, no mother, and the father had just a first name, I re-read what he had written, realized what I saw was consistent with what he had written - so apparently no bug was involved -- and thought I would make a suggestion to him as to how I thought it might work a bit better.

Lois,

Thanks. We (the curators) have been testing this for a few weeks and we know how frustrating it can be.
I can understand that Mike thinks it is good enough now, but there are still some problems and there is still a lot of possibility for improvement.
But even in the form it is now it can be a big help in improving the tree.

When the list is automatically refreshed and there was a hove rcard visible for the inconsistency profile that hover card is still visible after the refresh

As I wrote before: some functions in the "checker" are good, but it is hopeless because of old Nordic name traditions.
So my wish: remove asap: "children with different last names", "no birth surnames",and "unrecognized suffix".

These functions makes a lot of unnecessary work , esp with Nordic profils. (I have maaany profils to click "ignore"...)

I have not time nor capacity to fix all my "warnings" about this.

I agree if the consistency checker is based on a naming system that is only used in one culture it will fall if it is used on profiles with a other naming system. But you can hit the ignore button if you know it correct.

Some messages are still from the test fase. And it is still not perfect, but it is better then nothing.

Private User feel free to ignore the "children with different last names", "no birth surnames",and "unrecognized suffix".

I must say the "unrecognized suffix" should be turned of until it works better or all the messages should be cleaned up and rechecked. As that check does give a lot a lot of incorrect messages.

Private User I'm unable to reproduce any of the first three issues you mentioned. If you have existing examples, please post them or send to me via private Inbox message so we can diagnose. As for #4, I would argue that an obituary is not an event, it's a document. You can tag people in documents (even documents with dates) without triggering an inconsistency.

I have found a lot of data entry errors with the unrecognised suffix check.
From memory, I have found that some incorrectly contained birth surnames, mothers maiden name, alternative spellings of the name and nicknames (aka).

Erica Howton you're suggesting we ignore English suffixes with "of" in them. I'm torn.. on the one hand I do agree that would fix your problem without accidentally skipping too many profiles that ought to be flagged with unrecognized suffixes. On the other hand I'm not convinced "of <place>" should be considered a valid suffix? What purpose does that serve that is not already served by displaying birth and death locations on the profile, side-by-side merge comparison, etc?

Suzan Martin according to our research the average age at marriage has been late teens or later since the end of medieval times (see for example http://www.todayifoundout.com/index.php/2014/02/teen-girls-stop-com... ). Not to say that your data is invalid or incorrect, but that's the information we were relying upon when setting the cutoff age for Married Too Young, and the cutoff year for when it should no longer apply. Exactly how many of these are you seeing, that you believe are flagged incorrectly?

Job Waterreus your van / de / het suggestion seems reasonable... we'll look into it. I believe the Event After Death inconsistency on Marius Adrianus Brandts Buijs, Jr. was probably left over from earlier implementations of the check. I see it's gone now, so evidently that's no longer an issue.

Arne Åsbjørn Drangeid please provide examples where Children With Different Last Names is a disaster.

Torbjørn Igelkjøn we will fix that case you posted ("died 1893") to treat it as the *end* of 1893 rather than the beginning. But that's not going to fix the second example (child born 1953, parent died 1952)...

Private User good catch on the blank parent surnames; we'll fix it.

Torbjørn Igelkjøn and Private User it turns out we're already using the 1-year grace period for events with "funeral" in the title, so we'll just leave that as-is.

I want to add that the curators have been immensely valuable in fine-tuning these checks and I welcome their input and assistance. They know from hard-earned experience that in almost every case my first response is going to be, "post an example so we can diagnose the problem."

Mike Stangel It is absolutely impossible to maintain separate profiles for even the 40,000 plus I curate without more distinguishing characteristics. The advantage of a location suffix is that it is citation supported by wills. In England and the British North American Colonies, Wills (a legal document) start with “I, Erica Howton, of Pima County ...”.

In addition, “well known targets” in genealogy, such as the passengers of the Mayflower (see that project - Geni’s first, I believe) are distinguished by that suffix. You can do a count on the number of merged profiles to see how many we’re talking about, and “new” Mayflower pedigrees are uploaded to Geni weekly, if not daily.

Let me put the question back to you: what is the advantage of “not” having a disambiguation tool in the name?