What do you think of this "weakest link" Correct-a-Thon idea? [closed]

Question

Hi WikiTreers,

Last week we discussed a "Correct-a-Thon" marathon event idea. Through that discussion and subsequent testing, it became clear that my idea was a bad one. I proposed working on error types that Data Doctors have been working on for years. Too many of the remaining errors are unsuited to a broad-based community challenge.

Here's a new idea: Working on the weakest links in our connection paths.

It might work like this. We would have a list of 500 or 1,000 celebrities, historical figures, and other notables for the challenge. They'd be people like those in our connection features, like Sean Connery and the others in this week's James Bond feature.

As participants, we would look at the chain of connections between ourselves and each notable, one at a time. We would pick one link in the chain that is particularly weak — a profile whose connections cannot be considered reliable. Then strengthen it.

Maybe we'd recommend some standard of profile improvement, e.g. that it be improved to the point of being "genealogically defined" with a primary source that supports each parent, spouse, and child.

We'd record the profile we improved by the name of the notable. That would be one point. Then we'd move on to the next notable and find a weak link in our connection to that person. We'd see how many we and our team members could get through.

If you yourself aren't connected to the tree, or if you prefer, you could participate on behalf of someone else. For example, someone on Team Italy might choose to work on the connection paths between Christopher Columbus and other notables.

What do you think? If you like to participate in our marathon events, is this one you'd participate in? Would it be fun and productive for our mission?

Thanks,

Chris

Comments

Any update?

---

There isn't going to be a January marathon in 2021 but we are planning a year long challenge. Details coming soon.  

---

Here's the current plan for our year-long WikiTree Challenge:
https://www.wikitree.com/wiki/Help:WikiTree_Challenge

Answer 1

Anything to do with celebrities I avoid like Covid-19. That's just a personal thing but is why I wouldn't get involved. I'd much rather spend my weekend sourcing deficient profiles

David

Comments

It's true that the connection path would be between you and a notable profile of some sort but the profile you'd likely end up working on would not be. Just something to consider. :)

---

Sometimes it can be fun to look for, or at, connections -- as it was when I adopted Desmond Llewelyn and upgraded his profile, and (with help) got him connected.  (Actually found another few notables in the process.)

I don't always check each of the featured profiles to see how we are connected -- because when the path almost always goes through my dad, it gets boring -- but I have checked for the strength of some connections when I don't recognise the names.

When the path goes via my late husband (this is new, but has happened a few times), I get a warm feeling because I know the connections I have made are good ones.  (I can't answer for those after they diverge from what I know.)

---

I can't find a profile for Covid-19, is there a URL available?

8^]

---

Are you also averse to historical figures?

---

For placing a person's life and perhaps death within an historical context and from where the US news gets the Covid-19 numbers: The reference is at the bottom of the page.  Consider taking a picture of the screen, as the webpage may only be temporary.  Hope this answers your question, Rob.

Answer 2

I don't think I would have much value to add in this challenge.  And I'm not aware of too many celebrities or notables (a few) in the era I work.  However, could we consider having some tough ones that need better answers - I do have some of those - for some small teams to work?  Or perhaps we could work on "Brick Walls" - many of mine are far back and I know others would have those ancestors too (for small teams).  In fact, checking my brick walls, I could keep Isabelle and her team busy for quite a while!  Or I'd love to help people with Brick walls that end in places/dates where I could find, source and connect them.  The Brick walls might expand participation because we all have them and some people would love to get connected further back.

Comments

Great idea, Cindy, to 'trade' brick walls.  Mine might be easy for someone and others might be within my 'been there; done that.'  Hope you are visualizing a way to set that up.

Answer 3

To be completely honest Chris, not interested in this idea, I would have to pass this particular "Thon". as a team leader. I'd rather see us just do sources, edits and whatever on all profiles and leave the connecting to the "badged" and experienced connectors!!  They already have a connection_combat and connection_challenge event. 

My reason: There are so many profiles without sources, without biographies, and various other basic errors (ref: weekly suggestions reports) that this would be one, time consuming, and two, hard to be a team leader and help each person who may get stuck in the process, especially our newer members who tend to sign up for these "Thon" events. 

Anything that has a time limit and a "Team" competition element to it would cause more errors in the long run. 

Answer 4

Based on many previous connections to the weekly profiles I have seen, many of these potentially weak connections are probably in the 18th or 17th century colonies (Unites States), running through two specific lines. I do not feel comfortable working on these. (In some cases the weakest may be my own because of my inexpertise). I thought I would try with this week's people and was surprised (and pleased) to see connections going through a variety of lines in my/my husband's family. I have even found one connection where sourcing is needed on 19th century Dutch profiles (this is what I typically do). In this case, I am not looking for the "weakest" link, but moving methodically through the connecting people and seeing if sources are needed, so there may be several profiles being improved in this chain to Bernard Lee. I am still extremely dubious, but it seems  slightly better than my initial reaction.

Since I don't really care abut these connections it seems to be a very arduous path to finding profiles that need sources (especially when there are many much easier ways to find profiles that need sources). However, I never participated in the scanathon, so skipping this would be no big deal to me.

Comments

Thanks for at least taking some time to try it out. :)

Answer 5

I think this could work.  I see this as a focussed Source-and-Clean-a-thon.  Rather than improving profiles for random ordinary people (maybe based on time or place) that need work, we'd be improving mostly ordinary people profiles that need work (and possibly still time or place based) but they'd be related to notables in some way.  They could be on a connection list (the weakest chain concept) or they could just be in the circle around a notable - kids, parents, cousins....   It's all still improving/fixing profiles, so fine with me for a Thon.  And I expect that it would not be hard to find profiles that would benefit from anywhere from simple to complex fixes.

Comments

Yes, Paul, exactly.

Answer 6

WikiTree Jail, Here I Come!

As a Team Captain, I cannot commit to this thon. Seems as though everyone who thinks they are Someone Special will always work on connecting to a Notable. Not for me. I had rather make meaningful corrections to multiple profiles. Our Tree suffers from far too many unconnected and unsourced profiles. Each and every Thon we make a small dent in what is becoming an overwhelming task. 

The Rich and Famous always come out on top because they have the money to pay "to be found" and published.

@Whitten-1
OUR TREE is suffering from an explosive Growth Spurt and if we don't cordon off the old gedcoms and associated profiles and WRITE New Positive Precise Goals and Rules, OUR TREE is destined to wither and die from the neverending flow of New Unsourced Profiles. 

I would venture to say we all understand this, but everyone is so afraid of offending others, so nothing gets done about the root of the problem. The silly thing about all of this, no one has to be offended. We all know how to separate business from personal life, except in WikiTree. We are a business of sorts, we are all working on building Our Brand i.e. Tree. Sometimes, one has to make abrupt changes in businesses, say Re-tool, to become more accurate and competitive. One can continue on the slow path of massive sourcing BURNOUT or make changes and begin the new year with a Precise New Mission.

Otherwise, I will find my own little niche and work like a trojan cleaning, sourcing and connecting "Lack Lustre" profiles of ordinary people instead. "The Forgotten Ones"

Like I said in the beginning, WikiTree Jail, Here I come!

Comments

This really isn't about famous people. The reason we suggested the list of notables type profiles is it gives everyone a consistent list to work from. It isn't about the notable on the end as much as it is about finding a weak link and  strengthening a profile somewhere in the middle of the connection path which is extremely likely to not be a notable and far more likely to be a "forgotten one" thus making a more certain connection which improves multiple connections, not just the one that happens to end in a notable profile.

---

Eowyn, I understand the concept and it could be fun, I guess. But, I believe it is time WikiTree address more serious issues before these impending issues do irreparable damage. 

Fluff is pretty but I believe we really need to clean up the "Old Hot Messes" to advance our Goal of Accuracy for 2021.

I am terribly sorry if my opinion is not received in a positive light. We just can't keep sweeping these "Hot Messes" under the rug, eventually we will have a mountain under the rug and lord help us all, if it turns into a "Landslide".

Sort of like the Elephant in the room, everyone sees it but no one will admit to it. 

Final thought, how can we move forward if we are towing "Old Baggage" that holds us back? 

An executive decision needs to be made one way or the other. Let's face it, if those who created these "Dump and Runs" obviously do not care about their files or they would have stayed around to clean them up! I venture to say, if the entire WikiTree community worked tirelessly on these "Hot Messes" for the next 12 months, we would see the same thing happen in 2021 if the rules aren't changed and adhered to. Sorry for getting off topic.

---

Loretta, you might be making valid points, but you need to start new discussions for them. Please see https://www.wikitree.com/wiki/Help:Discussion_Rules rules #1 and #5. Answering your many concerns here would bring the discussion further off topic.

---

Thanks Chris!

Answer 7

Source a Thon concentrates on profiles identified as Unsourced.

Why not have a 'mini' Source a Thon, which is done only with profiles have a Gedcom associated with them and / or have the Gedcom junk in their profile.  Sources must be added to each profile, so if a profile already has sources, they don't count.

Along with this could also be profiles that have a family tree associated with them, but have no sources on the profile.

Both of these are the same type of profiles.

Comments

I was thinking the same thing, Linda. Have a Gedcom-Thon. Every day I come across profiles with Gedcom junk in them. Some are semi sourced and others are completely unsourced.

---

I love working by gedcom because working by an entire family at a time is easier--you've got a straight line from one source to the next.

---

Linda, this is an excellent idea. Old Gedcoms are a huge problem. 

By focusing on a list of Gedcoms (thus giving us a common theme), we would in actuality be accomplishing three things at once. 

Buy One, Get Two Free. 

(The buy one part) We source and clean profiles for one branch of a family (possibly completing that family), (first free part) connect that branch to the Main Tree and (second free part) get rid of the old gedcoms at the same time. 

That is a Win Win Thon.

---

Loretta, I love "Win Win Thon" 

---

Thank you!

---

Just to add another perspective: We've considered this, but if we focus on specific GEDCOMs, we're also focusing on the mistakes of a particular individual, and that just doesn't feel right.

That doesn't mean you can't work on them during the Source-a-thon, Clean-a-Thon, etc. We just don't want to make a community-wide effort of being critical of the work of specific members.

---

Julie,

There are sufficient GEDCOM created by Closed Account WikiTree (or last active say in 2014) to go around.

There are also a couple of (large!) GEDCOM with a pretty common name that show the profiles were created by different PM using the same GEDCOM name.

---

I assume there are plenty of 'old' Gedcom profiles that were loaded years ago, never been attached or only by Data Doctor Suggestion work, that have no sourcing.  Those are the profiles that need work to determine if they are duplicates to other profiles or just need sourcing done.  

Most people have no idea how to find profiles created by those old Gedcom files, other than just happening to run into them at different times. I don't consider working on those as being a mistake that someone did, more that wiktiree allowed lots of profiles at the time to be created with no sourcing or checking for duplicates.  Currently, it is a slower process of allowing profiles to be added with no sources, but we are still seeing that done.

My suggestion was not to use specific Gedcom, just to try to fix 'any' gedcom profile that is unsourced or profiles that only has a family tree associated with it.

---

Julie,

Most of the Gedcom generated profiles I come across are between 2011 and 2014. Many have been dumped by the original uploader.

All,

We need to get them tagged so they can be easily found. Perhaps a new tag e.g. {{Gedcom Generated}}.

I have seen the occasional weekly challenge focusing on cleaning up old Gedcom junk. I think the scale of the problem requires a big event. A thon or a 52-week challenge.

---

That is a good idea, if the Gedcom profiles are unsourced. We also have a ton with only a family tree mentioned. If it is a source a thon for those type of profiles, the bio would or should get cleaned up at the same time hopefully.

Answer 8

First and foremost the name should not begin with C or S.

Now if I understand you correctly, we will pick a path between us and the celebrity and strengthen the weakest link. So this is NOT working on the celebrities, it is working on the common people in between.

So, who determines the weakest link?  Can multiple profiles be strengthened or only 1 between you and each celebrity?

Will most of these devolve into looking at the sources and clicking the "certain" radio button?

Although I hate doing biographies, this seems to be a major weak point. I once more put forth my "Bio-a-thon" idea. This can help more people learn how to do biographies. After all, these "thons" should help the tree AND provide training.

We have sourcing with the Source-a-thon and partly again with the Scan-a-thon. The Scan-a-thon also touches lightly on biographies.Then we have the Clean-a-thon for fixing suggestions. Finally we have the Connect-a-thon whose goal is to add new profiles even if nothing connects. (profile-a-thon?)

What we need to target are weak areas in our tree. This is Biographies, Merges, template useage, and other potential areas I might not think of.

Ideas?

Comments

I think in a simple version, one picks a notable from the list, uses relationship finder to figure out set of profiles from you to notable, then start looking at profiles along the list.  Then largely up to you if any of them seem poor in some way that you want to or can fix - lack of sources, lack of a decent Biography....  There is no formal "weakest link" definition.  And I think it would be fine to decide to improve the biography as your choice of how to improve the profile.  From my random profile checks for sourcing, I note that almost half profiles do have a source, but a much smaller fraction, around 1%, have anything like a decent biography.

Answer 9

Well, I guess being heavily involved in the Notables project, I see quite a bit of value in this approach. My only concern would be how much time and effort it takes to identify this "weakest link" and who would be doing all this research. If it is a sort of challenge and we pick our own, it would seem sort of random and possibly arbitrary as to how we determine what we see as our "weakest link" or if we have multiples. If someone chooses the celebrity and we have similar paths, I could see us stepping on each other as there are many of us who have similar connections at certain stages and we might all try to edit the same profile. If this proves to be a challenge, we might have to provide multiple celebrities as suggestions and that way if too many pick on the 5th paternal relative out from someone, then we might have to give some options.

And we don't have to use "movie stars" - lots of alternatives. Sports Heroes, Nobel Prize winners, Mount Everest explorers, Famous Ship Captains, etc. It's almost certain the profile being updated would be someone relatively normal, and hopefully someone we've neglected for awhile.

Comments

"It might work like this. We would have a list of 500 or 1,000 celebrities, historical figures, and other notables for the challenge"

It sounds like the leaders would make a list and not just of movie stars. Then we would pick one, find our connection, and improve a weak link in between. Then next notable.

Answer 10

Wasn't there a push a few months ago regarding 5-star profiles? Couldn't each Project submit a list of profiles they'd like to see all the family members worked on? For instance, someone recently asked the community to help with the descendants of Thomas Gardner, one of the original planters of Salem, Massachusetts. He is PGM and has notable descendants, but the profiles in between need work and/or to be found or created. Some of the projects that may like to participate are:

• Beyond PGM
• Southern Colonies
• Titanic
• New Netherlands
• US Black History
• Italy

Perhaps some of the projects could create a list of 100 persons they'd like to see worked on. I'm not talking about working on the actual profiles, but either their ancestors or descendants.

Comments

I'd love to see this as a monthly thing--especially for our US Black Heritage Project.

Answer 11

This is something that I do every week when the new connections come out. I choose my closet connection and review to find the furthest back profile from myself which has not been marked as confident. I then try to find a source which will allow me to mark the connections to parent and child as confident.

If I can't find a source, I add a research note and do something else to improve the profile.

I'm willing to lead Team Cornbread Catchers on this one too.

[edited to complete my thought LOL apparently I got distracted]

Comments

I agree with Debi - linking it to the weekly profiles would be good and more managable

---

We're also thinking about having 52 "Weekest" Links thing in 2021 to go along with 52 Ancestors/52 Photos.

---

When I have time, I do this, too, Debi. 

It's not about the person at the other end. It's a way to find a path that I might not otherwise examine. 

Answer 12

My first thought it that I don't quite understand. I have a hard time getting everyone genealogically defined. when I do sourcerer challenges. so if that is involved, there would have to be some education on everyone's part. I do like the idea of a GEDCOM-a-thon. I do come across old GEDCOM profiles a lot. It could be old GEDCOMs or just finish working on our own GEDCOMs that are not finished. I thought when I joined Wikitree, my main job would be to put my GEDCom on Wikitree. But, I got distracted and never finished it. I like that we do it one at a time now, but i still come across these old ones.

Comments

I've been going back and trying to clean up old GEDCOM profiles from 2010, 2011, etc. A lot of them are locked, and the profile managers are no longer active, so we'd have to request for them to be opened so they can be fixed up. Quite a number are for people who may still be alive.

Answer 13

I say this often, but I'm not really interested in notables the way many others are, especially when it's the everyday profiles that are crying out for help. I just cleaned several entire family lines missing dates, locations, and correct family connections.--no notables were connected to these people.

I still advocate working through entire family lines, especially gedcoms that are a hot mess (often called dump and runs).

We could also work by surname lists. There are TONS in the surname lists with no sources, no dates, no locations, lots of suggestions. Until we have these basics taken care of, I can't justify spending my time prettying up profiles on the way to notables.

Comments

Emma, I couldn't agree with you more.

So much that needs to be done and so little time to do it in.

I would love to see all of those early gedcoms cleaned and gone.

---

I'm working on a 2013 one right now--over 3000 in the upload with no dates, locations, or sources. Plus, I'm finding all sorts of false connections (parent/child and spouse). It's a serious hot mess. Like you said, I'd love to see all this old stuff cleaned up to make the tree healthier so we can move on to other stuff to make the tree more robust.

---

I agree Emma. I often pick up an unloved profile from a gedcom and find myself cleaning up an entire family. It can be  both interesting and satisfying. Some of these 'ordinary' people are just as interesting as the notables.

I think one of the things that needs doing most on WikiTree is adding dates or locations to those profiles without them.  Surely a thon could be organised around them.

---

The proposal is to look at all of the profiles between you and the notable, not the notable.

---

We understand the proposal. We feel there is a more useful way to spend our time. :)

---

Keep in mind that the proposal is not to make profiles pretty -- it's to improve them with sources that either prove the accuracy of the connection or correct it if there are mistakes.

The Connection Finder profiles merely act as 'anchors' to give us paths to follow. Personally, I would start at my end and start working with a connection I have not marked as "confident."

---

My overall point about notables is these lines tend to get a lot of attention already because people get very excited about being related to them. All my personal family lines to notables are solid because they've already been worked on a ton over the decades. I personally wouldn't have anything to work on.

Answer 14

I didn't answer straightaway decided to sleep on this and this is what I suggest.

Since many of us have a weak connection to many of the notables that are highlighted should we create a list on a FSP,  from say January 1st until next Thon date, of those who we feel are weak links.  We need to ensure we include as much location details as possible for those profiles.

It maybe that if the cut off was a bit earlier some information could be extracted prior to the Thon by  Editbot.

That way geographical teams can improve profiles in their area and we are truly collaborative.

I know one of my weak links that frequently shows up is someone from England marrying someone who appears to have been born in the USA. I would like to get this one sorted.

Answer 15

I think "GEDCOM-A-Thon" would be a good idea. I didn't upload a GEDCOM but I come across many while sourcing profiles. So that would

1. Clean up GEDCOM-junk

2. Source many profiles

3. Get us rid of some of the most problematic profiles we have on the Tree.

I think that would be the best option for the Tree.

Answer 16

The original post has been closed, so would like to add  some ideas here (although they are not an answer to the question really, please forgive me) and thus forming an opinion about the idea.

- Gedcom clean up (already given by others) will also increase the sourcing and look-and-feel of many profiles.

- Connections: Connect-A-Thon (oh, we already have that one)

- Notables: there is a team/project for it. I am not a member. I come from a humble family without any interaction nor connection to older famous people... sorry.

- Mark profiles "genealogically defined" and thus state they are cleared from the basic work... Ah. New as far as I can see. The original question was about the rather broad discussion on accuracy of profiles. If there would be a way to easily mark a profile as genealogically defined (or not), the thon could be to improve a number of profiles and get them into a  defined state.

Own review: difficult to do for all, need new technical addition and discussion how to review. But in the end that is what it is about right?

- Mark as many connections certain/uncertain and then improve number of certain connections. No real tech stuff needed, just a counter and a way to search for the uncertain or non marked profiles. Interesting? No idea. New, yes. This would really help all the people that work with the data and make it visible for viewers which profiles are more accurate than others.

- Rate the quality of sources used. It is not so much about the number or lack of sources, but the accuracy and relevance of the source. I sometimes see profiles with 20+ links to family trees. Looks impressive, but value of those sources is low, very low. We are not allowed to put the unsourced template on those profiles, but for me those are the real issue: they just repeat what others have copied without anyone reviewing the real value. This is how science and peer-review and all related scores work by the way.

Sorry for the long post. Hope it is of some value. Please do not ask me to start new posts for all these ideas.

Answer 17

I performed this exercise with some of last week's profiles of the week, or perhaps it was the week before, and added sources to some of the profiles between me and the EPOW. It can be quite time consuming. I also didn't have access to some records of people from some nations whose primary language is not English, so I couldn't follow up on all profiles.

"We'd record the profile we improved by the name of the notable. That would be one point. Then we'd move on to the next notable and find a weak link in our connection to that person."

What do you mean by record the profile we improved by the name of the notable? Why not just record the Wiki ID of the profile we improved?

What happens if all profiles in my line are sourced to WT's level of acceptability, even if they aren't done well? Then I don't get any points for checking numerous profiles. Or if I only get 1 point per notable, I could have several profiles in the connection that need improving and I could be 30 degrees away from the notable, whereas someone else may only be 15 degrees away. The proposed scoring system doesn't place everyone on a level playing field.

 

Comments

Perhaps the points could be weighted. 10 points for improving a profile and one for taking a look and determining it's in good shape.

Or we could do this as an event without points being awarded. I would still participate. The attraction to the event is the camaraderie of us all working on the same thing at the same time to improve our shared family tree for a short weekend. It could also a teaching moment by familiarizing everyone with genealogically defined profiles or something else, if that is decided for the focus of the event.

Answer 18

I don't think I would support this idea. I don't care about being connected through several unrelated marriages to anyone.

Now a thon where only primary records for Birth, Marriage and Death are added, would in my opinion add the most value to WikiTree.

Comments

I answered about the requirement of primary sources already on a different thread, but here I go again:

My mother is from Serbia. There are no primary sources online from Serbia, because of law restrictions. No sources, no matter how old. I created profiles for the siblings of my grandparents, more or less documented on that what my mother told me about them. She said, the older of her uncles on the maternal side was mayor of their birth town. I found a book about the town of my grandfather, ca. 5km away from my grandmother's town. The towns of my grandparents always had relations and in the book I found the mention of "the mayor of [my grandmother's town, my granduncle]".

So this sentence in the book corrobates my mum's information, I cited the book and the page on my granduncle's profile. And you want to tell me basically that this profile of my granduncle is unsourced? Because I "only" source it with a secondary source?

In my opinion, the big strength of WIkiTree is that it allows a huge variety of sources. It only requires you to "tell where you have the information from". As long as you do that, the requirement is fit.

---

Dear Jelena, please don't get carried away. I never said anything about the topic "unsourced" nor did I say your family is unsourced. Neither did I say that the primary sources needed to be online. It is really sad to hear that there are laws in some countries that prohibit online sources, but that does not prohibit you from referencing a primary source in an archive or library.

---

Oh yes, it does. The Serbian archival law only allows Serbian citizens to look for sources in archives. I don't have a Serbian passport. So, I am *not* allowed to look for sources in archives, no matter if it concerns - for example - the military records of my uncles.

---

We all have other types of legal constraints, thou mostly on the age of the records. You are complaining about not being able to get legal access to your family's archival records. This actually has nothing to do about the records not being online.

---

But it does have to do about your wanting to add "only primary sources to profiles" which I simply cannot deliver for my family, obviously because of legal constraints, because I can neither find them online nor actually get access to them. 

Your opinion of "...primary records for Birth, Marriage and Death [...] add the most value to WikiTree" actually devalues millions of profiles that don't have primary sources.

I don't want to say that all sources have the same value, but you cannot dismiss profiles only because there is no primary source listed on the profile. 

---

Have you ever heard a saying that "Genealogy without sources is mythology" The "unknown" author was making referrence to how important the quality of sources are in genealogy. You get mainly two types, primary and secondary. The use of primary sources does not devalue profiles without them. It actually just takes those properly sourced profiles further away from mythology, a standard we should all strive to accomplish. Some of us will succeed and some of us won't, whatever the reason.

---

I know that saying. But it says "Genealogy without sources", not "Genealogy without primary sources". Secondary sources are not created in the moment when a father came to the priest to announce his child was born, but they are sources. 

I don't get where you get the reference to the "quality of sources" out of that quote. That sentence basically says: "As long as there are no sources, genealogy is mythology." That does mean on the contrary: "When there are sources, there is no mythology anymore." 

---

Trust me, the quality of your sources matter the most.

---

In some (many) parts of the world primary records have been lost due to bombing or fires, or other catastrophic events. It isn't practical to insist that those records must be present. We do need documented sources that can be verified in case of suspected error, and where those sources are less than ideal, then the collaboration occurs to find other corroborating sources, or more reliable sources. What we need to get away from is using unsourced family trees or personal knowledge as the sole source. It is encouraging bad habits in those new to genealogy and being abused by people who can't be bothered with good research.

---

In (some) many parts of the world primary records have not been lost and are available for use. I never insisted that lost records must be available, that is an assumption you made and it would be illogical to insist on something like that.

Not only do we need to verify secondary sources when we suspect an error, but we should always verify secondary sources with primary sources.(when they are available goes without saying)

---

@Louis, I didn't assume anything. 

It's short sighted to only see value in primary sources. Adding a secondary source is better than having no source at all, and therefore should not be discouraged. It's misleading to view primary sources as infallible. I've got secondary sources that are correct and a primary source for the same fact is incorrect. If I attempt to verify the secondary sources with primaries, I'm going to be very wrong. It's ironic that this needs to be pointed out to accuracy advocates.

---

You just assumed I only see value in primary sources. I didn't say primary sources are infallible - again an assumption by you- but they are way more often right than wrong, compared to secondary sources which are made from primary sources often subject to the accuracy and interpretation of the author of the secondary source. I also would not discourage the use of secondary sources unless primary sources are available. 

You absolutely missed missed the value of the proposal and confused it with the general use of sources on WikiTree compared to a thon where the use of only primary sources would encourage members to get a better understanding of the value of primary sources, espesially to those members who  would generally not use them.

---

It's rude to keep telling me what I assumed, when you have no idea what I assumed.

Your words were "Now a thon where only primary records for Birth, Marriage and Death are added, would in my opinion add the most value to WikiTree."

I don't see the value in spending time in searching for sources, only to ignore the secondary sources and not add them. I'm not assuming that you said only primary records are to be added. It's right there.

---

Dear Leandra, thank you for your reply.

Answer 19

I do think examining connection paths and looking for weak links there would be a good way to improve accuracy on WikiTree. That said, I have a few reservations about making that the subject of 72-hour thon.

- I'm aware of a few "weak links" - links so weak that I'm convinced they should be disconnected. (I already know that corroborating sources don't seem to exist). Fixing those definitely can't be done over a week-end. It would involve collaboration and communication, probably over several weeks. And there are several of these weak links that I know I won't ever touch, because probably all hell would break loose if I tried. It's hard to imagine doing this as a competition.

- W Robertson is surely right that a lot of iffy connections would involve 17th or perhaps 18th century colonists in America. Unfortunately, many contributors can't help with those. And in the event that a disconnection is required, we run into the same problem as above.

- I might be an isolated case (not being a descendant either of migrants or of nobility) but my connections always run through the same people. Would that be a problem for this concept or is the way it should be?

I'd like to see this idea used in weekly or monthly activities with the focus being someone's connections every month, perhaps?

Comments

We're thinking of doing a 52 Week weakest link challenge in 2021 to go along with 52 Ancestors/52 Photos.

What if for the purpose of the Correct-a-Thon the instructions were to improve the weakest link you feel the most comfortable working on?

Answer 20

I think that it is an excellent idea for us to check these connection paths,  though not an easy one . I wonder if it will work well as a 'one off' 'competition'.  

I have to admit that I often don't look at the connection finder.It is depressing. When I do check I always find lots of problems,. There are frequently profiles with  fictitious or unsourced links between countries This week a connection  went through a man who died in 1930s England 'married' to a woman living in 18th C America (now disconnected).

I wonder how far this experience of  the connection finder being  not worth the 'paper' it's written on,  is universal. For myself,  I find that even if the Notable was born in the UK, my  trail will  snake to N America and back.Sometimes  it visits Australia on the way. It is at the junction between countries that I often find the links are weak. No sources, occasionally quite preposterous as above.   Working  on these connections may end up with quite a lot of potential disconnections; something that shouldn't happen in a speed competition. Neverthess, getting rid of spurious links would enhance the Tree's  integrity.  

Aside from cross country links, I often see unsourced or poorly sourced profiles along the trails. These need work but that's where another problem lies; they are likely to be in a country  where I have little knowledge and expertise.

So in conclusion, yes I think investigating these connection trails and rectifying them is a really good idea. I think it might work  to have a weekly connection finder 'thread' where people are encouraged to  report their findings of one trail  and  could get help and support on fixing them. There could be a competition element but  not necessarily. Points don't motivate me so I leave that open.

Comments

I have just - after considerable preparatory work - disconnected an annoying connection that constantly appeared in my path when it had to go through royal profiles. This was part of a fabricated line which was disproven more than a hundred years ago. (There still a bit of mopping up to do before the little cluster of fabrications can be announced as Disproven existence).

Finding dubious stuff is one of the things the connection finder is good for - but you are right: if it's not within my area of competence I cannot do much about it. May not even recognize the weakness.

Interesting to hear that your trail also snakes back and forth to North America. There are active Swedish WikiTreers where my connection path crosses to America and back to Sweden. I would have thought that England was more centrally connected than that.
Edited: typo

---

Helen -

Your post both shows why this exercise is valuable, and also why it is difficult to do in a "competition" setting. :-)

Thank you for your thoughtful response!