Enhanced Filtering Functionality for Suggestions Report [closed]

+15 votes
371 views

I, for one, see no value is using my limited research time repeatedly to tell the Suggestions report that MY date and location data is sourced, while Findagrave's data apparently is not.

Can current functionality be added enhanced for to the Suggestions report to allow us to hide or ignore all Findagrave filter a larger group of related suggestions in from our report?

This filtering functionality would NOT permanently remove any suggestions from the report.

edit: updated tags (these may be more relevant!)

edit-2: edited question title; executed strikethrough for irrelevant text; highlighted new text

edit--3: What happened to my previous edits? restored previous edits; added new sentence to clarify intent of functionality option

closed with the note: Discussion went off topic.
in WikiTree Tech by Lindy Jones G2G6 Pilot (211k points)
closed by Lindy Jones

The primary point of my question is to determine if adding such functionality is possible.

The secondary point is to determine if such functionality would be useful to my fellow WikiTreers.


I am not asking how to fix Findagrave errors - I already know how to do that!

And thanks for everyone's input so far!!

When using the http://wikitree.sdms.si/default.htm page there is a function where it skips the errors reported. However it does need a certain amount of errors which is more than your current amount of errors per suggestion. (It may be 30 or so). 

Also by looking at some of the errors in the suggestions under your management there are several that you do not have the accurate date whereas FindAGrave does. An example is Finlayson-143 where he was born in December 1894 as can be verified on https://bdmhistoricalrecords.dia.govt.nz which you also have as a source. Searching the year 1895 for his birth will show he is not recorded yet searching 1894 will show him up. Doing an exact date search shows the birth was recorded on the date specified on FindAGrave. Just cause the registration date is 1895 doesn't mean he was born in that year as Late registration can and will happen. 

 

 

Thanks, Darren! That functionality partially accomplishes the targeted task.

My objective is to determine f that functionality could be tweaked to hide all Findagrave errors simultaneously instead of by individual error number.

 

12 Answers

+11 votes
I have had luck with setting the status of the suggestion to "False suggestion (hide forever)" and using wording that explains why FAG is wrong.

For example: "FAG Dates and Location are unsourced. See WT profile for sources".
by Steve Harris G2G6 Pilot (432k points)
That is what I don't want to be wasting my time doing, Steven.

I want to filter out the meaningless Findagrave "errors" so I can focus on legitimate errors.

I wouldn't call all of the errors meaningless. This error in particular is good for those of who may have 'fat-fingered' a date, or may even have conflicting sources and are not sure in which direction to go for a specific date.

While I am quite aware of the errors and lack of unsourced profiles in FAG, the headstones, cemetery locations, and other features on FAG can be of use to users here on WikiTree.


With that said, I may have misinterpreted your statement of "repeatedly" telling the system that your sources are correct and not the information on the FAG record. In my answer, my process is only done once per profile and it clears the error - I will never see it again unless something is updated, a new or changed FAG link is added to the profile, etc.

Also, if you are not handling those suggestions, and ignoring them like they do not exist, could they then be caught up in a sourcing or data doctors sprint since they technically still exist as an error? If they were, your profiles are now going to be reviewed and possibly edited by other members or having to be cleared by other users.

I myself am tired of seeing the DBE_503 error, "Probably wrong gender (male)", and the DBE_511 error, "Unique Names (spelling)" - but I handle those errors as they come up in order to lessen the burden on myself and others:

  1. So I don't have to see the error again in my suggestions report; and
  2. So other users are not wasting their time trying to edit the profiles I am working on or trying to clear my errors for me.
Perhaps my choice of word was misleading!

It is not a situation in which corrected errors don't disappear from the Suggestions report. I no longer work these errors, which are the bulk of my Suggestions report.

My point is I want to filter Findagrave errors from my view without having to mark them as false (they will still be available to be fixed by Data Doctors). I want them hidden from my view, but not removed from the Suggestions report, so I can see what REAL errors exist for the profiles I manage.

And I am ignoring them because, in my view, they do not exist since MY data is sourced (or marked as uncertain). Errors at Findagrave are not my concern when I am working at WikiTree.
Steven, Yes you can "fix" the suggestion easily but it is a waste of our time to tell the system that we have it right and another website has it wrong, that time would be better spent adding actual good sources and just allow the sites like FAG to be used just as breadcrumbs that allow us to look for the real facts. We all know that the majority of FAG memorials are shaky at best so making a comparison between what we have and they have is never going to improve quality here, too many just change our data to match theirs because they think that we are saying that FAG is right.

I don't think that any of the errors are a waste of time. Yes some are annoying (The 511 unique name error springs to mind) yet to ignore an error on my family is not something I would like to even think I could do. 

If it means I have to suggest error corrections on other sites then I will do it. Otherwise what is going to stop someone with good intentions coming to wikitree and adjusting my family data to reflect the data on the other sites. It does happen now as I have seen examples of people who are using the new Gedcom system adding info which is not accurate. 

It doesn't matter where the information may originally reside but to leave incorrect information about a family member out there just isn't showing respect for those that have come before us. That is how I see what this problem is. Fixing my family members information no matter where it is. It doesn't matter to me if it is on another site, If I have referenced it I want to make sure it is accurate. 

It also comes down to the fact that FindAGrave is seen by many as a source of information and we would be remiss in making sure the information coming in is correct. The database can not make the call which information is correct, it needs human input to tell it which information is correct. The Database can only deal in what is imputed. 

A suggestion could be implemented for different data from Ancestry Family trees but that would probably run into the millions and crash the suggestions report. 

We can do False error which hides the suggestion and doesn't take that long. It can be done with three clicks on a web page. I can find a source with three clicks sometimes and add it with a couple more. What takes the time is assessing the information between the clicks. To do a false error suggestion shouldn't take as long as you will already have found the correct information and already assessed it as accurate. 

 

​I was hesitating from making any additional comments, but reading through the thread again, I almost feel compelled to make at least one additional comment.

My point is I want to filter Findagrave errors from my view without having to mark them as false (they will still be available to be fixed by Data Doctors).

This statement seems very selfish in nature, not to mention counterintuitive to collaborative genealogy. If I am reading this correctly, and I am hoping I am wrong, you are saying that you would rather hide errors for yourself so as not to be bothered with them, and instead of taking the upfront time to clear the errors, you want to leave them for someone else to work on.

When another user now takes the time to cleanup database errors in an effort to better the goals of WikiTree, they are now forced to waste their time researching your profiles and other sources in order to make an informed decision on how to best handle the error - An error which you could have cleared out in a much shorter time since you know and understand the profiles, as well as the sources.

To close this out, I want to make sure you understand that I completely understand your frustration and the time that would be involved in trying to clear these errors, and the point where we differ is how best to handle those errors. There have been quite a few options presented as answers (and comments) in this thread, and while they are not perfect for your situation, they do present much less of a risk than changing the suggestions report we have now.

My fellow Wikitreers are not forced to waste their time working on errors for profiles I manage. The majority of Findagrave date and location "errors"  associated with the profiles I manage are errors at Findagrave, which is where any corrections would have to be made.

I simply would like to be able to view my Suggestions report with the all Findagrave "errors" filtered. I want to know what other suggestions are in my report; they are generally overshadowed by the Findagrave "errors."

I am asking if programming such functionality into the Suggestions report is possible. I feel such functionality would benefit the WikiTree community by saving us time and allowing each of us to focus on our chosen areas of work.

My question was not asking how to clear these "errors." My question was simply asking the feasibility of a possible improvement to the Suggestions report functionality.
The problem with that is you are throwing all your FindAGrave errors into this idea that they are wrong & you are right. As proved further up this thread I have shown one green locked profile where you have the wrong info while FindAGrave has the correct info backed up by independent information. This profile is a New Zealand profile. You also have messages on all your profiles saying keep away unless you prove your sources first.
By ignoring the suggestions that will waste other people's time which they could use fixing other errors or sourcing unsourced profiles. As a Data Doctor you should heal thy self first. If they are false, mark them as false. But you must be too busy to actually check properly your profiles to bother. The errors I just looked at are from January and could have been easily dealt with. If you count all the replys to your message it is probably as many as the suggestions you have.
Lindy, It could be done but the majority of the Database Doctors want the error to remain as it is and fight every change. The fact that WikiTree even states that FAG is not a good source, and that is the reason a template can be used for it but is not recommended for sources, falls on deaf ears. I do wish there was a way to put the FAG link into the profile without generating a lot of useless suggestions that make extra work for us but since there is no way for that to happen I will not use FAG on profiles I am manager of. Not the best solution but the only one that works for now.

My question was not intended to start a debate on the value of Findagrave "errors," Darren. Had I wanted to debate that issue, I would have asked a question related to that issue.

If or when I choose to work on my Suggestions report is entirely my decision. If a fellow WikiTreer chooses to work on them, that is his/her choice. And those choices are completely irrelevant to my question regarding a potential enhancement to the filtering capabilities of the Suggestions report program.

All the profiles on that I manage are works in progress and will likely remain so for several years. I don't doubt that the same is true for most of the profiles at WikiTree. As with my fellow WikiTreers, I am free to choose to work at my own pace on specific areas of my choice when I edit profiles.

Dale, I am not asking for a filter that removes the errors from the overall Suggestions report. I simply would like to have the filtering capabilities that apparently already exist be enhanced so I can filter/hide ALL Findagrave "errors" (or other expanded group types) at the same time. Currently I can filter each individual error code; I would like to be able to filter all Findagrave error codes simultaneously.
+7 votes
I will actually not add and even remove FAG as a source on profiles I manage because the quality on that site is getting worse and not better. Most of the time I can find a better source for the death information and like you I am fed up with the suggestions that just because my sourced data does not match their unsourced data I should make the change. I agree that it can provide clues but even the clues can be wrong without extensive research and I have seen far too many who will treat the information from that site as if it were Gospel.
by Dale Byers G2G Astronaut (1.4m points)
I don't use Findagrave as source documentation for my data, Dale. I link to the memorials mainly as a courtesy to anyone who views the profiles I manage and/or edit.

Just as with other genealogy websites (including WikiTree), the data at Findagrave is only as good as the source documentation behind it. Some Findagrave memorials have source documents attached; most, unfortunately, don't. But they may have useful information on family members.

We link WikiTree profiles to various other genealogy websites. Rather than ignoring some of those sites that we view as inferior, we should attempt to educate our viewers of those sites' flaws rather than simply ignoring those sites that are popular among casual viewers.
So its permissible( I remove them anyway) to remove a good hearted data doctors and others addition of FAG info to profiles we manage? I have found more errors on FAG than anywhere else.
Lindy, I use the site but only as research and I check every fact from other sources so to add unsourced or even wrong information here is in my opinion worse than just ignoring it.

Jenn, I don't know if it is permissible but I feel that adding FAG is doing more harm than good so I will not do it.
If you are the sole manager of a profile, you have the final say on what information goes on that profile, Jenn.

In my opinion, the errors on Findagrave and other websites are only relevant to WikiTree if we use that erroneous information and present it as factual and sourced. I present Findagrave as a potentially useful resource, but not as a source of the data on a profile. Any data I mark as certain will have a citation to a source document or its transcription.

Errors found on non-source websites should not be relevant to our work as WikiTreers. If we want to correct their errors, then we would work at those sites.
+5 votes
I have also had some success in notifying the FindAGrave manager of the problem, along with telling them about the sources.
by Kay Knight G2G6 Pilot (370k points)
That, too, is not what I want to use my time at WikiTree doing, Kay.

If and when I want to work at Findagrave, then I make those corrections.
I report errors, but leave it up to the FAG manager to change it.

I view that much like the contributions on WikiTree. Anything that can be done to get correct source data out there is good. We know that the data from FindAGrave gets replicated all over (without checking sources), so if we can get it corrected there, all the better.

Its much easier to get things changed on FindAGrave than on Ancestry or FamilySearch (and yes, there are record errors).

I might have a different perspective, since I spend most of my WikiTree time sourcing, with the thought that it will help all the replicated data. Unfortunately that means I don't spend a lot of time on my ancestors, but rather on 'strangers'.
Replication of erroneous and unsourced data is probably a core problem for nearly all genealogy websites, unfortunately.

As long as sites allow ANY unsourced data to be added to a profile or tree, this problem continue to grow beyond our capacity to make any dent in it. If their members don't care to do quality work, these sites will become less relevant to our work at WikiTree.

Thus, I focus my efforts on building well-sourced profiles at WikiTree. Any site that doesn't have a repository of source documents will, at best, be a minor resource and noted as such on profiles I manage and/or edit.
+5 votes
Many Find A Grave Bio"s are filled in by person at a later date.Often there

is no documentation.My advice always prove the records.I have found some correct and some in error.
by Wayne Morgan G2G6 Pilot (915k points)

Not much different than some WikiTree profiles, Wayne! At least we care enough to improve our poorly sourced profiles. That's the main advantage of working on a single tree.

I mainly work here by sourcing (meaning adding proper source citations - mostly from FamilySearch.org) to the profiles I create, as well as profiles that I adopt, before starting on the biographies.

Our source documentation is the foundation of our profiles and their biographies. Similarly, Findagrave's foundation is the photos of the headstones/grave markers. Without those photos or images of death certificates, their memorials haven't even met their own minimum requirements!

+5 votes
I would consider Find-A-Grave a level 3 source. Level 2 if it pictures the gravestone.. Other level 2 sources might be a news article. Level 1 source would be death certificate or coroner's report.

Many deaths on some of the older profiles are dated the last child's birthdate and then selected as "after". So this does have the possibility of narrowing it down enough to find "better" sources.

If I know my sources are better I always do the false suggestion. That way it keeps my list cleaned up.
by Steven Tibbetts G2G6 Pilot (279k points)

As I stated, Steven, I don't even consider Findagrave memorials as any level source. At best, it is a loosely constructed and poorly sourced collection of family trees.

We can presume that most members have offline source documentation for the data and information they add; but until Findagrave requires citing those sources, the site will never be a legitimate source for genealogical data.


I don't see any value in using my time even to mark the Findagrave errors as false. If we absolutely must resolve conflicts between our properly sourced data and Findagrave's data, then let the Data Doctors Project handle that. But allow me to hide Findagrave errors from my Suggestions report so I can find and work on the REAL errors that I can resolve with my source documentation.

I want to know what errors exist on WikiTree profiles, not elsewhere. And I don't want to wade through a list of outsider errors (that I am not going to bother to "fix" or mark as false) to find the real errors.

+6 votes
Whenever I have sourced information that is different then what FAG has I always suggest the edit on FAG and usually the manager has no problem with correcting the information.
by Stephanie Stults G2G6 Mach 3 (37.0k points)
That means I have to use my time at Findagrave, Stephanie.

Sometimes I will work on the errors there. But when I am working at WikiTree, I want to use my time fixing WikiTree errors, not Findagrave errors - which are merely discrepancies between their data and my sourced data.

For some of my fellow WikiTreers, splitting your focus between/among 2 or more sites may be an easy task; but for me, I need to focus on WikiTree and FamilySearch only to do my best work!
+4 votes
Lindy, many of us share your sentiments, but you don't need any special functionality to ignore it.  Just simply ignore it.
by Dennis Barton G2G6 Pilot (385k points)

Yes, Dennis, I am aware that other WikiTreers share my sentiments (one of the reasons I am asking my question!). However, ignoring those errors while they still populate one's Suggestions report doesn't solve the problem.

My goal is to filter those errors (would be a temporary measure, an option hide Findagrave error while viewing the report) so a member can see what real errors exist on his/her report.

It's no real problem for me since my Suggestions report is small; but for a WikiTreer with hundreds of errors that are mostly Findagrave related, such functionality might be the difference between completely ignoring his/her report or correcting valid errors on the profiles he/she manages.

Well I think it solves the problem unless you really care about the list.  My current "suggestion" report for profiles I manage has 42 entries, 41 of which are inconsistencies with FAG.  I do look at it now and then, and I do acknowledge that there have been a few legitimate errors - e.g., multiple WikiTree profiles referencing the same FAG memorial, thus identifying duplicates to be merged.  In all cases the other PMs and I have taken care of those, and I have fixed the other real errors   For the FAG remainder, who cares?  In general, I think having software that can find and report obvious errors or inconsistencies is a good thing, but comparing our data to that on other web sites was just a bad idea.
Dennis, The problem with your "solution" will be very evident soon. There is a Clean a thon happening soon and during that many will be making changes fast to rack up points without doing much if any research and I have seen too many members just change our data to match FAG just because they think of FAG as a source. So ignoring the suggestion will only cause more work for some of us and could put false data on profiles that might not be changed back if the manage is not watching the changes. The best solution is to not have any checking of FAG with WikiTree, No one check familysearch.org or Ancestry against WikiTree for errors and they have actual sources there so why check FAG?
I agree with your point, it is a potential problem.  You're preaching to the choir here.  But I do watch the changes, and I really haven't experienced much of a problem with over-zealous data doctors making false "corrections" to my work.  I have even seen comments in G2G by several of them saying they'd ignore these FAG inconsistencies.  So I have just found that I waste less time by ignoring the whole thing.  Granted, it isn't the ideal solution, that could change, and your mileage may vary, but that's my conclusion for now.
+4 votes

If I'm not mistaken... there's a template you can use when you add your find-a-grave "source" to your profile(s), and one of the settings in that template basically tells the Suggestions report to ignore any potential differences.

Unfortunately, I have trouble finding the list of templates and how to use them. (edit: see Deb Durham's post)

by Dennis Wheeler G2G6 Pilot (535k points)
edited by Dennis Wheeler
+6 votes
From my reading of your answers to other peoples answers and comments You seem to be someone who is passionate about having great well sourced profiles.

Why then are you adding a link to an unsourced profile on some other site? If you don't trust FindAGrave then why link to it. It would be the same as if you linked to an Ancestry Family Tree which only references other Ancestry Family Trees or saying the info came from "Uncle Bob".
by Darren Kellett G2G6 Pilot (147k points)
There is a valid reason to use and link Find-A-Grave: there is a tombstone photo that shows a birth/marriage/death date or year. In the absence of (and until) a better one, I will use a tombstone photo as a source.

 

The challenge is that the errors aren't being checked against the tombstone photo, they are being checked against the almost always unsourced Find-A-Grave profile information. I said my piece last year about the wisdom of checking discrepancies against unsourced material, but personally I have started removing "courtesy" Find-A-Grave links from profiles I am actively managing except in the case above and marking "no error" when the flag is caused by the unsourced Find-A-Grave material.

Hi Ellen it is the same for Wikitree errors. You could have a document uploaded proving a date of birth but may have an issues where the child is born before the father according to the database. A database search is only as good as what data is entered into it. The FindAGrave database error suggestions just check what is in their database compared to what is in the database entries on Wikitree. 

I have seen profiles on Wikitree where the information in the Biography does not agree with what is in the profile database which is what is searched on. 

I have also seen some FindAGrave profiles that have excellent sources with Birth, Death and Marriage certificates loaded as photos so they can be independently verified. 

The original poster of the question is having data mismatches between her sourced information on Wikitree and the information on FindAGrave. She said the FindAGrave information on the profiles in her suggestion list are not sourced.  My answer to her was asking why she was linking information that to her knowledge is not sourced accurately seen as she seems passionate about having well sourced profiles. 

My question is about enhancing the filtering capabilities for the Suggestions report.

My question is unrelated to any specific Findagrave "error" on the report, the validity of any source or resource, or any other issue.

+6 votes

Please scroll down to "Advanced Usage for WikiTree+" at the following link. Adding the parameter "sameas=no" to the Find A Grave template will prevent the program from comparing the data contained in the profile to that on the linked memorial and therefor will not generate a suggestion report for the discrepancies..

Find A Grave Template Usage

by Deb Durham G2G Astronaut (1.0m points)
|sameas=no is more for marking the FAG entries on the person's wife/children/parents/next door neighbors listing that are in a person's profile as part of their bio, not to mark the FAG entry for that person's own memorial there as you can view it but do not use it.

it also takes care of marking a person's own memorial if there are duplicates waiting to be merged, or if that memorial is a cenotaph and not the primary memorial, as in killed during a war, buried overseas, but family put up a headstone in the family graveyard...
I do understand that was the purpose of creating the parameter, but the result would be the same, I believe. It gives her what she wants.
I actually use the sameas=yes template, Deb.

However, template usage is off-focus from the question I am asking, which is about adding useful additional functionality to the Suggestions report.

(Don't feel bad; most of the answers and comments have been off-focus, too!!)
I don't feel bad in the least.  If you want to avoid suggestions that come up as a result of conflicting data on Find A Grave then I believe using sameas=no will give you the result you are seeking without additional programming time spent.
+8 votes

I would lobby for removing the FAG errors completely from the data-errors/suggestions checking. 

As reflected elsewhere in this thread, FAG entries are not much better than Ancestry.com family trees, and we don't have a suggestion algorithm for those (and I hope we never do).

 

by Jillaine Smith G2G6 Pilot (764k points)

I think that's overkill, and unnecessary.

The whole point of the Suggestions Report, is to automate finding areas of WikiTree that are incorrect in some way, so that they can be corrected.

To that end, many, many people "cite" (or at the very least, reference) Find-a-Grave. That will never end, I'm afraid. So when there are differences discovered, those differences should be accounted for and reconciled. Not all differences found are false, I would venture to guess that most are legitimate errors that should be corrected. So we are in fact improving WikiTree one profile at a time.

True that often, the errors are only on the Find-a-Grave side. As WikiTreers, then technically, that's not our problem, and as such could be considered a "waste of our time". But I think that's a poor attitude for genealogists in general. Because I suspect that most of us also have Find-a-Grave accounts at the same time, and would like to see corrections (and sources) made on both sites ultimately.

I, for one, am glad to be given the chance to systematically re-verify my profiles and keep them as correct as possible.

Hey other Dennis, I don't think anybody is really campaigning against the detection and correction of legitimate errors in WikiTree.  But for many members, when the mechanism for doing so involves a very high false alarm rate, we get to the point where it seems that researching the inconsistencies is no longer a productive use of one's available time.  The point at which different members conclude that is subjective, and may vary considerably from member to member.  But it's really not fair to label that as an attitude problem for those on the lower end of the tolerance scale.  It's just a cost-benefit trade-off.
I agree that Find a Grave errors should not be flagged, except perhaps in three cases--where one Find a grave entry is attached to more than one profile, which helps one find duplicate profiles; and where there are links to merged or nonexisting profile IDs.

It doesn't matter, though.  We've been discussing this long enough for me to realize that this is extremely unlikely to ever happen because people with power consider them valid.

The added functionality I suggest could be used when comparing our data to other websites that we cite as sources or resources, not just Findagrave.

When the inevitable addition of those sites to our Suggestions report happens, we would be able to filter the sites we don't want to work at the time so we could focus on that day's chosen work.

In my opinion, we accomplish more when we can focus on the work areas that we chose ourselves. For myself, I want to focus my time first on the errors in my Suggestions report that are not related to Findagrave.

+4 votes
Simply avoid adding Findagrave as a source, if you think that little of it, then you won't get any "Suggestions".
by Gillian Causier G2G6 Pilot (232k points)
Find a grave can be an accurate source about the location of a burial.  Frequently it is the only readily available source for the location of a burial.  Sometimes other details of a Find a Grave profile are correct; but without attached sources, there's no way to tell.  So it can be used as a probable location of a burial.  Having worked extensively with cemetery records, I can tell you that in my local cemetery of 4800 tombstones, at least 30 tombstones represent people who are not actually buried at the cemetery.
And by omitting FindAGrave then you lose the 'hint' to search in that specific area for the person or their family or to search for records that connect the family.

Yes there are tombstones without burial, tombstones with incorrect dates engraved, death certificates with the wrong birth date (common) and wrong death date (less common). The information from Find A Grave still contributes to exhaustive research.

Findagrave memorials can have useful information, Gillian, such as headstone photos or images of death certificates. The memorials may also be linked to those of family members.

I don't always follow the new leads that Findagrave provides, but others who view the WikiTree profiles may find the leads useful in their research.


But again, we are going off-focus from the point of my question: the feasibility of additional functionality for the Suggestions report!

I agree with you Lindy, but getting Ales to make the change--how can we do that?  He has the power to decide everything about that report, and he doesn't agree with us.

Related questions

+7 votes
2 answers
+2 votes
0 answers
+2 votes
1 answer
129 views asked Jul 16, 2019 in WikiTree Tech by Susan Keil G2G6 Mach 5 (53.8k points)
+27 votes
3 answers
201 views asked Feb 6, 2019 in The Tree House by Dennis Wheeler G2G6 Pilot (535k points)
+7 votes
2 answers
187 views asked Dec 7, 2018 in WikiTree Help by Susan Keil G2G6 Mach 5 (53.8k points)
+3 votes
1 answer
+7 votes
0 answers

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

disclaimer - terms - copyright

...