Konrad, See THIS FAQ to set the match criteria for the online tree. It doesn't give you very much control at all.
In Family Tree Builder, click the Help button and be sure you have downloaded the latest version of the Users Guide. Then go to p. 439, Configuring Smart Matching. It doesn't have much of an explanation (in fact, none at all!), but play with the settings. Set the First Name field to NO and set the Last Name field to Very strict, or Exact names only.
It's not entirely what we as users want, but this will get rid of some of the worst of your matches. It will possibly filter out some good matches, too, unfortunately. Back the settings off until you get a level of quality that is tolerable.
Eva, yes, we all look forward to the release of the New Improved Smart Matches (NISM) and hope that they live up to their name.
I'll take this opportunity to mention that Geoff does not simply merge the SMs but does due diligence and careful fact-checking before he incorporates them into his tree.
I hope the NISM incorporates your excellent suggestions, Geoff
Tom I put this here because you also had a couple of apposite (great word, I only learnt it a few years ago...) comments on the system; also so it becomes the last in the forum and people are more likely to read it...
It is a mystery to me why, when a post is deleted, it is ascribed to "Deleted Member".
I saw it a few times and thought,"Oh, golly, someone has been, if you will, drummed out of the regiment. In front of a full ceremonial parade, his sword broken across the knee of the regimental sergeant major and his insignia of rank and regiment ripped from his uniform... "
But I think it might mean 'We have deleted the post and the name of the writer has also been deleted'. Dull boring stuff, no slow, dramatic tap, tap, tap of the single side drum...
So why don't they SAY 'Member's NAME deleted" ?
Whichever way you read it, 'Deleted Member' is incorrect. He has not been killed! And he apparently has not been removed from the membership roll.
I am one of the lucky ones. Nearly ALL of my SmartMatches come from a relatively small group (never thought about how many but maybe fifty close and what? three hundred? total), most within Australia, who actually share tree members, many or a few. Maybe one per cent of matches (good or bad) come from others.
I reject quite a few where the proposed match is with maybe a father with the same names but of course different wife, different dates....different person.
And I match item by item, meticulously (though I still have mistakes in my tree...).
And a lot of my time is taken in correspondence with others about details; was this person born where it says in my tree or theirs? Are the children in their tree actually the children of his SECOND wife, who is not even in their tree?
An old lady actually asked if I was a school teacher because I seemed to like finding mistakes: I look at it the other way, I like to find the truth, and to work with others so there are less errors in the MH trees.
Ah. I guess part of my problem is with all the Swedish patronymica you get a few generations back. There are a lot of Anders Svenssons and Anna Nilsdotters out there, most of whom are false matches. It would really help if the list didn't contain "matches" where there are birth years for both parties and these are decades or a century apart. But as things are, it does.
Matching "dead babies" with persons with an adult lifespan ought to be possible to avoid. In cases where I have entered "dead babies" in the tree, to make the family structure visible, I regularly "ignore this person" as soon as I get a false match. (Don't know if I've ever had a true match for the poor kids).
Also, I have started ignoring Danish and Norwegian trees on a regular basis. Won't be missing much that way - but there's never an end to them, either.
I recognize the item by item matching as what I do with relevant matches and trees; turning back to the parish books to sort out the mismatches in details or hunt up new leads. That's what matches should be for.
I wonder how other people handle their lists of matches.
As I am currently visiting the site on a daily basis, I also keep my list clean on a daily basis, confirming (a few, but a few of those have been real findings) and rejecting (masses, which often takes considerable time, even done in bulk). I'm just thinking that if I don't handle new suggestions as they drop in, the whole thing will get too messy.
Have you given up housekeeping, or have you found some good methods to reduce time spent rejecting false matches?
This seems like a time waster to me. Half the matches are from sites that have had NO activity in years. What is the point of matching with someone if they never update their site? I think the way to do it is to just answer if someone else asks for a match. The way Wiki tree does it is a better system.
I wonder...Does it MATTER if sites have no recent activity? If details are not updated?
Surely all that matters is that the information was accurate in the first place??
What if I die? The information in my tree, which I have spent over two years compiling, will still be valuable to others. Facts about the lives of the people in my tree will not change, though someone else may make a correction or an addition to the information...(and this is something I think MH should make possible; it could be an addition to my "deceased owner" tree with the name of the person who made the comment shown)
Mary Kathleen - I understand your idea: having a site which shows no obvious activity for some time seems to be odd - but I agree with the suggestion from Geoff: In case of Genealogy, it is much more important that the information in the tree is accurate – valuable data does not necessarily mean that is has to be data that has been updated recently.
However, you might – although you do not see obvious recent activity – still try to contact someone from another family site regarding a Smart Match. For that you can use the Contact Site Manager link, which is located at the top of a particular Smart Match.
Let me know if you have further questions on this.
I think the biggest issue with matching is that - if (say you) post a tree with incorrect data- might be easy like a birthdate - wrong.
Some time later you become offline- either pass on or just simply cease to subscribe.
Your incorrect data CANNOT be altered and many others see the incorrect data that has been "copied- ad naseum" and THAT data becomes the FACT and - is inherently wrong- the error just gets perpetrated.
WHERAS- if ytou pots dat that is in error and you are still live and valid- the "matcher" can contact you- a debate ensures and if you are proven in error- you would edit you data- and all is sweet.
But that can only happen ifd the poster of the data is still current.
I agree pretty much with what you say, there is a problem where sites like this cannot be ALTERED. But that does not mean that the information could not be placed in the tree!
That is why I suggested MH allow COMMENTS to be added by MH members though they may not be able to change the original tree. With a suitable heading and icon....(Another genealogy program allows comments in their birth certificate etc. section with a big pencil icon and the comment in heavy green with the name of the member who put it in)
So in a recent important anomaly I found; if I could not contact the site owner (deceased or not now a subscriber or whatever) I might place this COMMENT in the tree:
"The photo of the gravestone in this tree does not seem to be of (not the name) Mary Ann born Smith; a check of the birth and marriage records shows that the person shown on the stone as her husband, John Jones, actually married Mary Ann born Tremellen. I can find no record of a husband of Mary Ann born Smith. Geoff Stewart".
And it would be great if new comments could be 'tagged' (? whatever) so that they show up in exactly the same way as smart matches
This is actually a further development from the simple idea of a big green question mark 'watermark' that could be placed on suspect profiles or matches. Which is still good but this would allow a place for explanation.
Went to a Genealogy workshop yesterday and in the morning session, the presenter, a professional genealogist of thirty years' experience, started off with a list of the Ten Most Common Mistakes of Beginners.
Here's Number Two:
MYTH: If It's On The Internet, It Must Be True!
"The Internet is a valuable genealogy research tool, but Internet data...should be approached with skepticism. Even if the information you find seems the perfect match to your own family tree, don't take anything for granted....The trick is to learn how to separate the good online data from the bad, by verifying and corroborating every detail for yourself. Contact the researcher, if possible, and retrace their research steps. "
That applies to Smart Matches. In every case where a SM has had data that's not in my tree, when I contacted the site manager to ask, "How do you know that he was born in that place on that date?", the answer is, "I don't know. I got it from a Smart Match." One piece of data I chased back through seven trees before I gave up: Everybody had just copied blindly. Nobody had bothered with a source citation. NOBODY knew where that data came from.
I have seen the most ridiculous errors listed in tree after tree because somebody copied the data without looking at it carefully and without checking original sources to verify that it was copied into FTB correctly. Mind you, citing sources will be easier once MH gets proper citation handling in place (it's pretty crude right now, but even so, please use it).
Go ahead, find those SMs. But when you confirm the SM, be sure to also confirm the accuracy of the data. Otherwise, you're probably just copying somebody else's mistakes.
The issue of people just copying data from somebody elses trees is really a concern for me. The validity of the tree is impacted gravely if you can not be sure that the data is correct ie. validated against formal information sources. The possibility of marking the information as positive or doubtful is of course in the system, but when you look at the tree it self, you cannot see this information. A colour code could be the solution or a different font on the data. In relation to smart matches, a marker on the information that will show, when you check the information to see if it is relevant for you, could be a great help. The mark could be a notice that this information is copied or is from original sources, defined as formal information about birth, death and similar facts.
But it is also a FACT that if you "take" a smart match-- discover an error (or query) - and then try to make contact with the site manager or site managers- and discover that they are NOW unfinancial subscibers- you can do NOTHING to have the error corrected and the error WILL CONTINUE to be perpetrated and myheritage in their infinite wisdom will not take steps to rectify the error or even make an annotation or some alert. But I think the error is set for all time and cannot be altered.
Thus if you alter the data in YOUR record -YOU are the one who appears out of line with EVERYONE else and your correct data WILL BE IGNORED based on majority - even in error.