Is there a way to show weights of an error?
Charles Taylor KWNR-S97 and Quality Summary
He shows only one error for consistency, but many for Conflicts, yet both only lose one star.
Comments
-
@AnneLoForteWillson thank you for your feedback. Several updates have changed the look of this significantly. Could you give this individual another look and let us know what you think?
0 -
Can I just laugh?
So looking at Charles Taylor KWNR-S97 now:
First let's talk about Data is Conflict Free.
He has three errors listed. The first one, death happened before the residence...the residence is from his death and burial records, the date that I logged was the date those were registered, 29 Oct 1895 instead of the date he died, 24 Oct 1895. Out of curiosity, I changed the date to 24 Oct 1895 (since they are really referring to where he lived when he was alive) and suddenly 'Data is Conflict Free' jumped to 4 stars!?!? AND his profile quality went from Low to High! Really? Had no idea that a residence would have that much importance.
Next, dealing with the life events statement. I LOVE that you point this out. One question, do you account for large populations as well? I mean, if someone has events occurring on the south side of London and also on the north side of London...would it give a "Watch out! There are a lot of people living in this area, they might not have moved this far" even though it might only be 5 miles distant?
However, I happen to know that these locations are correct. Lonnnng story! So I clicked on the message.
YAY! I can DISMISS the error.
BOOHOO! I can't give a reason WHY these locations are correct.
BOOHOO! I dismissed the message, but it doesn't go away. It doesn't get crossed out, it doesn't get marked in any way.
Hmm...when I changed the residence, the message went away immediately. Maybe it needs to refresh? No change when I refreshed.
Hmm...maybe I didn't actually click it, you know how you jiggle the mouse just as you go to click? So I dismissed it again and got this error:
Windows 11, using Chrome Version 123.0.6312.107 (Official Build) (64-bit)
Last message, regarding 13 children. Yup, I also know that is correct. Wild story, but DNA confirms all these kids. So, dismissed. No reason statement. (YIKES! I really need that reason statement!) Nothing happens. So I Dismiss again and now I get that error message same as above.
So Dismissing isn't working quite right yet.
Second, Regarding Consistent with Sources:
The first several messages are actually all the same, that his surname shows up as Moore instead of Taylor. None of those appear to be given much weight, as the score for that is 4 stars, even though there are 7 errors listed. Since these first few are all the same error, could they be logged as a single error? "On this record, this record, this record and this record, his name is listed as Charles Moore, which doesn't match his stated surname of Taylor."
Hmmm....Does this account for additional names recorded in his profile? AKAs, married names etc.) Tested that! YES!!! THANK YOU for calculating against AKAs!
So additional thoughts,
THANK YOU for making the number of children being a single error message.
But what happened to old age at child's birth? I really would have expected that to show up on him...I mean his youngest child was only 7 when he died at age 80. (Yes that really is true!) Could you make this statement exist but, only count it once? So no matter how many children were born after he was 'old' the statement only shows up once? It really is something that I would want to be alerted about, I don't think that 45 for men is 'old' though. Maybe 55 or 60 (for this time and place).
I see the five stars for Data is Complete and Data has Tagged Sources...but what does that mean? Could you put an information link on those so I know what I don't need to work on and why those sections are five stars?
Also, there are still hints and Research suggestions for him:
'Possible Missing Child' research tips should mark down either Data is Conflict Free or Data is Complete.
Hints on the profile should mark down either Has Tagged Sources or Consistent with Sources.
Other Research Suggestions, such as looking for another spouse should also be accounted for within the Profile Quality Score.
Knowing the weight: Obviously we want everything to be as accurate as possible, but I am surprised at how much or how little these errors actually change the score. Could you color-code either the messages, or the little signpost icon to how much that error will affect the score? Maybe red for a lot, orange for medium and yellow for little effect on the score?
Hope some of this helps!
0 -
@AnneLoForteWillson Thank you! There was another update that processed on April 10. This helps tremendously to see where further refinement in needed.
0 -
@AnneLoForteWillson another update 17 Apr 24 fixed the "dismiss" issues. Yay! Thank you for your feedback and suggestions.
0
