WeRelate talk:Vision

Let's create a vision for WeRelate

  • What are our core values?
  • What is our vision - short-term and long-term?
  • What are we afraid of?
  • Personal deal-breakers?
  • What's most important about WeRelate to us? Why do we contribute?
  • Do we want to absorb more users? If so, what kind of users do we want to attract?
  • How important is quality vs quantity? Can we have both?

Topics


One person's wish list [20 February 2014]

Elsewhere I have responded to the survey about features in WeRelate where I do not see any value. I have also commented on where I think we need to add features to attract a broader user base.

I would like to directly respond here to Dallan's specific questions:

1) Q - Integration with living data?

  I think this is the main objection for a lot of potential users as I have said.

2) Q - Improving ease of use?

  Again, I think this is a hurdle that new users have to overcome.

3) Q - Adding DNA support?

  I am not certain what is the actual benefit.  DNA research can prove or disprove family relationships but it does not prove or disprove a specific ancestor (eg DNA could show that one of your ancestors was a close relative of Thomas Jefferson but would NOT show that you are a direct descendant of Thomas Jefferson; however DNA could show that you are NOT a descendant of Thomas Jefferson or any of his close relatives).

4) Q - Becoming multi-lingual?

  I think this would definitely attract more users.

--Jhamstra 02:51, 21 February 2014 (UTC)


Sorry - I did this backwards. Herewith my comments on the vision thing.

1) My vision for WeRelate is built around the synergy from opportunities for collaboration.

2) My biggest concern is that all the information I (and others) have contributed will be lost if WeRelate does not achieve the "critical mass" necessary to flourish.

3) Deal-breakers for me would be if the rich database content is watered-down, either for simplification or to merge with some other entity. Despite valid ease-of-use concerns, I think we have the right underlying database structure for the kind of information we are capturing. Let's not break what is one of the strongest arguments for this platform.

4) I think we need to attract more users to reach a "critical mass" where WeRelate will survive and flourish, as opposed to decaying into a "dead" database like several others that are out there.

5) It is inevitable that we will have a mixture of "quality" - that is certainly true of my own work where I have excellent sources for some pieces of the "puzzle" and very weak sources for other pieces. Nevertheless I have tried to create the most accurate and complete picture of the various families I am documenting (not just my own 8-). The only way to have an exclusively high-quality database is to restrict our coverage to those times and places that are well-documented. Otherwise we will have to acknowledge that a very large portion of Pando does not come from well-documented times and places and do the best that we can. Robust databases are designed to capture and tolerate information of varying quality, track the quality of the information, and have a process for improving content and correcting errors. The WeRelate platform qualifies and we should take maximum leverage from this feature. I think that this capability distinguishes us from the rest of what I have seen online.

--Jhamstra 03:11, 21 February 2014 (UTC)


Pando and growth [9 December 2014]

For me, the heart of the vision for WeRelate should be to be the number one place where people can collaborate to build a common publicly-available family tree. This concept of "pando" - collaborating on a single tree - can only be really meaningful if there is sufficient scale. There needs to be a good chance that one person coming to this site fresh will find information on here that has already been added about one of their ancestors.

This is currently the reality on Ancestry, for instance, but is not the case on WeRelate. If we are serious about being a pando, we need to have a clear desire to get that that scale and a clear understanding of what steps we need to take to get there.

Here's an overall perspective on size:

My Trees
Default (search)
pages: 0
107,600,000,000 Number of people who have ever lived [1]
21,100,000,000 Number of people who have been born since 1650
7,100,000,000 Number of people alive in 2013 [2]
4,000,000,000 Total profiles on ancestry.com (incl. Duplicates) [3]
1,000,000,000 MyHeritage [4]
780,000,000 GenesReunited [5]
640,000,000 Names on RootsWeb World Connect Project [6]
460,000,000 Community Indexed International Genealogical Index [7]
400,000,000 GeneaNet [8]
130,000,000 FreeBMD Births [9]
98,600,000 Geni [10]
60,000,000 FamilySearch Pedigree Resource Files [11]
12,493,344 FreeREG Baptisms [12]
6,815,024 WikiTree.com [13]
5,038,660 GEDCOMIndex Open Source library [14]
2,489,522 Number of people on WeRelate [15]
1,335,000 Thomson-Gale's Biography Resource Center [16]
800,000 Rodovid Wiki [17]
630,600 Biographies on English Wikipedia [18]
224,000 The Political Graveyard [19]
139,929 FamilyPedia Wikia [20]

In my mind we need to be into the hundreds of millions of people before the pando idea really starts working. To get there we would need to double every year for the next six years. That's do-able, if we set it as a target, but it would need a clear vision, a clear commitment and a clear path. Who's in? AndrewRT 20:12, 21 February 2014 (UTC)

You would never be able to accurately source every person who has ever lived, which is why 'pando' is an unreachable pipe dream. We could upload people all day, copy other people's trees and would never make a dent in those numbers. "Reliability" will always trump Pando. Daniel Maxwell 23:30, 22 February 2014 (UTC)
Andrew, thanks for very interesting post and table. Small correction: according to the page Rodovid Milestones passed 800000 persons milestone 7 July 2013.--Alexandre 21:30, 21 February 2014 (UTC)
According to the page Rodovid Milestones passed 900000 persons milestone November 2014.--Alexandre 22:20, 9 December 2014 (UTC)

My Two Cents [22 February 2014]

Here’s my two cents:

What are our core values?

Truth and inclusivity.

On truth. We will never know much of the truth, even about the fairly recent past. But we should respect the attempt to get closer to it. This means providing the best sources we can, and appealing to those sources, and reason, when deciding between competing interpretations. And honestly acknowledging doubt. And, (my personal pet peeve) not making $#@! up. (I.e., don’t invent birth dates or places without noting that they are guesses, and stating what the guesses are based on.)

On inclusivity. It is impossible to construct a pando (a unified world family tree), or even a substantial fraction of one. But that should still be our aim. (Oscar Wilde: A map of the world that does not include Utopia is not even worth glancing at…) This implies that we need to be open-minded about what counts as an “acceptable” source. And that we need to be reaching out to different communities and language groups.

Vision, Short Term.

To become the primary reasonable-quality pando on the Web. I figure that, genealogists’ time being limited, there is only room for one quality pando on the Web. This is because, as I see it, most genealogists will not contribute to constructing a pando and most of the rest will not contribute to more than one. And the one that they will contribute to will be the one with the best combination of visibility and reputation. That pando will ultimately wipe out the others.

Vision, Long Term.

Pointless, because none of us know what is in store, in the rapidly changing technological environment that we find ourselves in. That said, my fantasy is that at some point AI would greatly facilitate the construction of the pando.

What are we afraid of?

That Werelate becomes an also-ran. Putting work into a wiki is risky, because at some point someone may come along and subvert what you have done. That risk is not worth it, if the wiki is a minor wiki. Why not just construct a personal website, where at least you know your work is safe in the short term?

Personal deal-breakers?

There are many. I am only going to be putting my time into the best wiki out there. That is Werelate at the moment. But that could very easily change. I have been putting an effort into making many thousands of hand-edits on Werelate, because I want there to be a big-time, free, non-commercial, non-religiously affiliated pando in existence. It bugs me that there isn’t, yet. If Werelate decides it doesn’t aspire to be that pando, then I will be looking into constructing a half-decent personal website, and wait for the big-time wiki to come.

What’s most important about Werelate to us? Why do we contribute?

I touched on this above. Also important to me is the emphasis on providing sources. It was this seriousness which helped attract me to Werelate in the first place. (That was before I realized that the seriousness was mainly lip-service to an ideal, and not embodied in any kind of rule or structure.)

Do we want to absorb more users? If so, what kind of users do we want to attract?

Yes, we want to “absorb” more users. The pando vision is impossible, otherwise. We want to attract users who are committed to the same vision, and capable of contributing to it.

How important is quality vs quantity? Can we have both?

We need both, and if we can’t have both, then Werelate has no reason to exist. There is already quantity without quality, at Ancestry and elsewhere. And of course, there are a number of places where you can find high-quality work online. If we can find the right formula, the one that combines ease of collaboration with a commitment to minimum standards and an incremental improvement in quality, then I think we are off to the races. Easier said than done, I know.

A final thought. Any discussion of our values must, I think, start with an honest look at where we are now. There are many wonderful, well-sourced pages on Werelate. But, taken as a whole, Werelate is full of junk. We have to be honest about that before deciding that we would rather not take the risk of mixing with the swinish multitude.--Werebear 18:48, 22 February 2014 (UTC)

I completely agree with Werebear. I think it's important to recognize that family history sites, like other things (e.g., telephones, fax machines, social networking sites), are affected by network effects - all things being equal, the value of a tool increases exponentially with the number of people who use it. This generally leads to either one tool winning over others (VHS instead of Betamax is the best example) or to interoperability (e.g., telephones and fax machines). I think/hope that eventually network effects will lead to the emergence of one Creative Commons-based wiki family tree, and I want WeRelate to be it! I think that we need to do whatever we can to gather new members - we will figure out how to improve quality, improve newbie instruction, etc., but growing the user base is key. -- Jdfoote1 20:05, 22 February 2014 (UTC)

A few brief thoughts [25 February 2014]

As I mentioned in my comment on Werebear's topic, I think that we should do whatever we can to grow the user base. I think that one of the great lessons of large-scale collaborative projects is that quality will be pretty good, as long as there is a good revision history, and a large number of participants (e.g., Wikipedia, Firefox, Linux, etc.).

In order to get more users, I think that we need focus on improving ease-of-use, adding new features, etc. - things that require a developer community.

For me, one of the big questions is whether or not we need to figure out a way to include living people. I agree that this is one of the big roadblocks stopping new users from participating, and as I said, I think getting new users should be a primary goal. However, one of the things I love about WeRelate is the openness of all of the content - everything is Creative Commons, and nothing is "owned" by anyone. Including living people would require authentication, authorization, etc. Adding these leads to a much more complicated model (see WikiTree). I want my contributions to live on, and I think that Creative Commons content has a much better chance of surviving whatever happens. -- Jdfoote1 20:20, 22 February 2014 (UTC)


For me, living people are key. And that's because even when their vitals are hidden, living ancestors on site allow us to find out how We Relate. I can see that on Geni, and WikiTree, and that's why my trees are there, not here. WikiTree, not listed above, has 6,815,024 profiles today.
Other factors are that Geni and WikiTree have forums on their sites, which I think are easier to use than talk pages, and that I can download GEDCOM files for any part of the tree on WikiTree. I need that as a safeguard, for the simple reason that I do not know if sites survive.
--Enno 13:11, 23 February 2014 (UTC)
The whole point of this site is open collaboration. Having large portions of WR hidden to other users is not a step in the right direction (not to mention - you couldn't do much or any quality control on these entries); there is also the liability concerns with living people. Those sites also have tree ownership and there are many duplicates - and you have to ask yourself - which site is the best sourced and referenced? WR hands down. We have been in the midst of a long-term clean up of livings from the site; I have seen on more than one occasion a comment by the 'living' person himself demanding that they be removed from the site. So that isn't going to happen. However, there was a solution discussed awhile back about having 'livings' within a side program so the whole thing could be linked together, but that never happened. But I do like the idea (and one I had also thought of) being about to download portions of the site in a format of your choosing, such as GEDCOM, r some similar format, such as GRAMPS XML. I am not inherently opposed to forums, either, though I think in some cases other watchers of a page may not see a comment about a person unless it is at a talk page.Daniel Maxwell 18:21, 23 February 2014 (UTC)

Well, sources don't count much when they're attached to persons that I don't relate to, which is largely the case here. Geni has attracted lots of users from my part of the world, meaning that the number of reasonbly sourced profiles that they have for me is larger. Same for WikiTree. So, in other words, your best sourced and referenced is quite irrelevant to me.
And in order to change that, I think that it's important to attract users from Europe, who most probably use GeneaNet and My Heritage right now, and local sites.
--Enno 21:19, 23 February 2014 (UTC)
Well, then foreign users who care about sources will have to come here and start adding material. We have French and Dutch members who do just that. It's a chicken and egg problem. Then there is also the issue that we haven't very good multi-lingual support that probably scares off a few people. Daniel Maxwell 21:38, 23 February 2014 (UTC)
Well, yes, but the practical situation as I know it is that a Dutch user with a lot of time left with a fight, years ago, and started working on thousands of Dutch profiles on Geni. He still does that, and he's one of the driving forces that make that the Dutch section on Geni is better than anywhere else. I spend more time on WikiTree, for several reasons, one of which is GEDCOM download.
In other words, as I see it, the critical mass is there, not here, and I don't think it will change, unless WeRelate comes with features that make this site better than FamilySearch, Geni and WikiTree together. Without such features, more users will be attracted to WikiTree, because of a feature that many may not know: For all profiles on site, Geni and My Heritage provide matches against WikiTree. That means that WikiTree gets traffic almost automatically!
--Enno 10:55, 24 February 2014 (UTC)
The main problem there right now as I see it is the lack of multi-lingual support. Perhaps when (not if) this is done, it will help the situation. Also, maybe you could you could tell the viewers of this page some of the features that you wish Wikitree had that could possibly work better over here? Having no Dutch ancestry myself, I hadn't known the Dutch section over there is considered very good. Daniel Maxwell 11:02, 24 February 2014 (UTC)
I first found WeRelate from Google search for some of my Dutch ancestors. 3/4 of my ancestry comes from the Netherlands within the past few generations and almost as soon as I crossed back there in my research I started finding people on WeRelate. So a fair piece of my work was already in-place and I had (and still have) collaborators in fleshing it out. The other 1/4 of my ancestry goes back to 1630s settlers in New England and once I got back there I also found existing content on WeRelate, though with a lot of problems from the earlier "drive-by" uploads. So for me WeRelate actually added a lot of value. However I must say that now the coverage on the Dutch side has become better elsewhere and I find myself searching other web sites for clues that I can bring back to WeRelate. Yes they have a fair amount of duplication but the quality and consistency of what I find is quite good. On the Anglo side I personally am benefiting greatly from the dedicated efforts at cleaning-up early New England and earlier "old England" genealogy by other WeRelate contributors.
When working on my wife's ancestry which comes from Central and Eastern Europe I have found zero synergy on WeRelate. This particular part of the world has been through so many religious / political upheavals that much or most of the information one would like may never be recovered. So I am fortunate that my own ancestors came from places that are reasonably well-documented whereas my wife's ancestors are rather poorly documented (if at all) before immigration to the USA. The same goes for other family friends and relatives that I have investigated (I could cite many more examples but I think I have made the point 8-). Having seen the huge disparity of availability and quality of sources from different parts of Europe and North America is one reason that I seriously doubt that WeRelate can grow much beyond its current coverage with its current institutional biases. Even in North America a very large portion of the current populace are descended from immigration from places other than Britain or the Netherlands where present methods yield good results. To attract contributions from these people and places will require opening WeRelate to other approaches.
I really think we need to take heed to input from outside North America in this discussion. The situation is not the same everywhere.--Jhamstra 12:22, 24 February 2014 (UTC)
Well, the lack of synergy in Eastern Europe isn't a WR problem, but documentation problem from that part of Europe period. I know that for some people of Eastern Euro descent whose ancestry is flat out not possible for them to take their ancestry before the 19th century because of lost records. Of course, we'd love to have better coverage in that part of the world at some point, but I daresay that even Ancestry.com doesn't have much there, either. (depending on where you meant by Eastern ancestry of your wife). WR has, btw, some of the best and most accurate coverage of early New England families probably on the entire internet. I am one of the users who has seen to that, but there were several others who also contributed heavily to that effort. So in areas like New England, this site really shines. Daniel Maxwell 18:27, 24 February 2014 (UTC)
The foregoing comments also affect my views on the need to accommodate living persons in some manner. When I can trace my ancestors and their families back 6 or 12 generations, there is a fair amount of gain from capturing and publishing information about the deceased even if the living are excluded. When I can only trace my ancestors back 2 or 3 generations, then almost my entire known family consists of the living. In this case there is not much personal benefit in contributing to a wiki where I can capture very little of what I know about my family history. One of the huge advantages that FamilySearch now offers is a place to capture information (facts, photos, narrative) about the living as well as the deceased. The former might not be published but it is captured and can be shared.--Jhamstra 12:35, 24 February 2014 (UTC)
Right. And I think that we all know, or should know, that most newcomers have no more than those 2 or 3 generations. And depending on the area where they live, they may create a personal tree on-line, on Ancestry or My Heritage, or connect their desktop program to Ancestry (FTM), FamilySearch (most US made, except Gramps), or My Heritage (FTB). And when they use the latter, which is the most multi-lingual of all (except Gramps again) they will be confronted with matches on Geni and WikiTree. Chances that they choose WeRelate are low, I think, very low.
Language is an issue for many too, just like forums. Geni has a larger crowd of Dutch than any other site with a shared tree, and active multi-language forums, no talk pages. Geni's user interface is multi-lingual, just like My Heritage's, and the latter, being Geni's owner, tries to attract Geni members to its own (non shared) trees, and source database, by showing matches, just like Ancestry, and sell an extra subscription to them. That means, that I, being a paying Geni member, am daily nagged with record matches on My Heritage, which I can only access for an extra fee. I don't like that, but I do assume that it works for them.
The Dutch part of my ancestry is quite done, so I don't need Geni or any other site for that. I have trees on local sites, which some time leads to new contacts with cousins, but that's ad-hoc for me. I have a brick wall in Germany, which is quite hard to solve, and my main research area today is English nobility, before 1600.
For that, I sometimes use the FamilySearch tree, via RootsMagic, but that one is such a mess, with some people having 20 parents or more, that I find way better clues on Geni and WikiTree. And many times when I Google a name, I see results from those sites, and wikipedia, not WeRelate. That does not mean that WeRelate has no relevant English nobility, but for some reason, it does not seem to be found by search engines. And where I checked last weekend, for ancestors that I seem to share with the US president, it turned out that relevant profiles here refer to Ancestry trees as a source. And that's not much of a source, is it?
For that English nobility, Geni has the best clues, and when I find those, I try to cross check things with reputed medieval sites, and wikipedia. Trouble with Geni is that even as a paying member, I can't download any part of those trees, so for that I rely on WikiTree. They have an active project for European aristocrats, and more and more profiles are well sourced over there. And WikiTree, although based on MediaWiki, just like WeRelate, uses dedicated forum software, which IMO works much easier than talk pages over here. Members are invited to use tags, to vote on questions and answers, etc., which makes the site much easier to use for newcomers. And ... they can upload or enter living people, thanks to proper privacy controls. WikiTree is English only, but that works for me.
My main database is in Gramps, not on-line. I have no idea how long any site will survive, and many rely on investors or the personal efforts of people like Dallan here. And the only site that I trust to survive, FamilySearch has so much mess, that I'd rather not download much from there. And that's also because as far as there are sources over there, they are downloaded to RootsMagic in such a weird format, that I'd rather not import those in Gramps.
I'm writing all this, because it describes the factors that count for me. Those are communities for help on research issues, for which forum software is an advantage, I think, but most important is well sourced data, that can be exchanged two ways, which is an advantage for WikiTree right now, even though I know that WeRelate is more advanced, with family and source pages, which don't exist over there. And in my ideal world, there is direct communication between Gramps and WeRelate. Being able to download GEDCOMs for user selected parts of the WeRelate tree would be a nice start however.
Final comment for now: Geni and WikiTree have a relationship calculator, which can take private profiles into account. That may seem to look like a toy for many here, but I know that it works for me, to relate to fellow members and notable people alike, and because it encourages the review of profiles found on the paths connecting us.
--Enno 14:19, 24 February 2014 (UTC)

I have previously suggested a "layered database" approach. Having privacy controls does not mean you have hidden everything from everyone. For example the duplicate person / family checking could still show pages that are close matches. But you do need to hide private data from public searches. You could view private data as sort of like an unpublished sandbox. You can still share results with other collaborators and have a time-driven mechanism for moving private data into the public space. Also I suggested that Creative Commons licensing is not incompatible with private data. We can reflexively dig-in our heels and refuse to consider whether there might be a better way to do things or we can seriously examine other possibilities.
I would be interested to see whether WeRelate is continuing to attract more active contributors? Is there some way to report the trend over the past few years (eg quarterly) how many users actually made more than 100 edits / 1,000 edits / 10,000 edits per interval? If these numbers are increasing then we are on the right path. If they have leveled-off or are declining then we need to seriously ask whether we need to make some changes.
--Jhamstra 00:29, 24 February 2014 (UTC)
It is better left out for the reasons I already said - this debate was settled years ago, I should add, before either of us signed up. Alternatives that do not do 1) creating hidden person/family pages that are low in quality and not possible to maintain or perform quality control on 2) involve the creating of person pages on the website is another discussion. Perhaps this could be coupled with the GEDCOM/XML import/export option mentioned above - these people would remain inside a person's 'tree' in the 'view tree' option but are hidden from every one else but the user himself, and would exist only within the program, not creating person pages that come with liability/privacy problems that an OPEN wiki would never solve. This is basically what was agreed upon years and years ago but never implemented. Daniel Maxwell 00:37, 24 February 2014 (UTC)
Well with all due respect, if we are confined to exactly what was decided upon years ago, then there really is not much point in a discussion of vision, is there? The original vision cannot be extended or improved and there is no point in any change other than minor incremental adjustments. I would not claim that every new idea is a good one. Change can painful but it is sometimes necessary. Our policies and features should be driven by our vision, not the other way around. For me the point of this page is to discuss whether and/or how the vision should change. The answer to both of these questions may turn out to be NO - but that should be a decision taken after thorough discussion among the stake-holders, rather than an a priori assumption.--Jhamstra 00:51, 24 February 2014 (UTC)
That's a fair point. Let me say though that when I define for myself the 'Vision' on this page I had not in mind 'overturning', but you may disagree on the definition. You may well be right, anyway, though I do not believe on the 'livings'. The problem isn't that we don't have livings as they exist now (and since you've deleted hundreds if not thousands of them I think you'd agree?) but how to implement some form of them that would allow this *perhaps* blind spot in the site to be rectified. My participation here is not intended to be obfuscatory, but that the policy that were debated here years ago had good reason for implementing what they did. Daniel Maxwell 02:31, 24 February 2014 (UTC)
I have some ideas regarding how to manage the migration of data from GEDCOM uploads / initial data entry through private data to fully published data. We could borrow some ideas from industrial database practices (eg PLM) that would allow us to manage qualitative data progression in a cohesive way. We could get both more flexibility and also greater consistency in the quality of the released data. But before delving into the details of how to do this we first need to agree on what we are trying to accomplish 8-). --Jhamstra 05:52, 24 February 2014 (UTC)
Wow, sounds interesting. It would probably solve a few problems that I met here earlier. Care to elaborate on another page some time? --Enno 18:34, 25 February 2014 (UTC)

Another version/vision? [22 February 2014]

What am I afraid of?
That WeRelate will stagnate and die.

I’m old enough to be concerned about what will happen to my research after I’m gone, and don’t want later generations to have to repeat my research because it has been lost. Because of that concern, I also contribute a skelton-ized version of my research to Family Tree on Family Search, which has the institutional backing to survive, and has converted to a source-driven tree building model. Yes, there is a lot of cleaning up that needs to be done with existing data, but that is slowly beginning to happen, and is likely to continue. It just takes time. But there is a lot about Family Tree that I don't particularly care for.

What attracts me to WeRelate?
Flexibility in presentation of text and in linking various pages together; the potential for adding article pages; potential for linking speculative relationship pages. There is room for a lot of creativity and variety here.

Pages of persons who are associated with an ancestor, and who appear in a document with that ancestor, can be cross-linked and/or linked to an article page. That linkage can begin to create a sense of a community in the past, which speaks to a human perspective, has importance for historical research, and is also an aid in the process of genealogical proof.
The ability to indicate speculative and refuted relationships. One genealogist I know said that this established WeRelate as a “thinking person’s” site.
Text on a person or family page can be a proof (or dis-proof) narrative and/or a biographical statement. It can therefore appeal to both those interested in quality research and those interested in human interest family history.
Tree-building (Pando) is not unique to WeRelate, nor is collaboration. Other sites, however, do not include the kind of flexibility in linkage and in narrative that WeRelate does. Those sites that are “managed” also do not insist on/enforce the one-page per person rule, nor seem to have the same concern/support for source citation.

What would be a personal deal breaker?
The loss of flexibility, including an emphasis on “clean” pages without discussions or indications of conflicting/contradictory or speculative information.

More users, what kind?
Yes, active and passive.

We need more active users (people who contribute on a regular basis) AND passive users (people who primarily “watch”, or just visit without registering). We also need more volunteers.
The fear that an increased number of active users would result in drastically lowered quality is probably unfounded. First, the screening process for gedcom submission involves “work”, something many people seem unwilling to do. Second, many people are afraid of wikis, because they are not “secure”. As one man explained to me, he’d rather be wrong than have someone else edit his data.

Quantity versus Quality?
I think this is a false dichotomy.

Increased quantity (an increased number of users) does not necessarily involve a decrease in quality, particularly if we target the appropriate audiences. These might be the people who are actively associated with NEHGS, NGS, subscribers to TAG, people who attend the various Genealogy Institutes (Samford, Salt Lake, Pittsburgh, e.g.). NGS has a research trip scheduled to the Allen County Public Library at the end of August. Is there any way they could be informed about WeRelate? Part of the problem is just getting people to be aware of our existence.

Core Values?
Quality combined with collaboration.

With one page per person, ensuring quality should necessarily involve collaboration. If too many people who are concerned about quality forget that a bit of collaborative communication is useful, it will turn potential (and long-standing) contributors off. A lack of response by active users is an invitation for other users to ride roughshod over their material without necessarily improving quality.

Short Term Vision?
Giving the Genealogical Proof Standard a more prominent role as a standard for evaluating quality, and as a means of resolving contested/contradictory information.

The GPS is the professional standard, has been for some time, and does not rely on personal judgment. (And can now include “authored sources” as a third category of sources.)
Not all pages need to measure up – probably less than 5% of my pages do. But it could be potentially very useful in working out contentious disagreements, as well as helping contributors to understand what “quality research” really means.

Long Term Vision?
A respected site that attracts visitors both because of the quality of the information, and because of the quality of the narratives that accompany the basic data.

Not every visitor has to be a contributor, as long as the site continues to attract many visitors. Nor do all visitors necessarily need to be hard core genealogists. They could also be people only passively interested in their family history, or people interested in particular places or particular historical contexts, presented from a perspective different from that of Wikipedia.--GayelKnott 20:39, 22 February 2014 (UTC)

Thoughts on growth [27 February 2014]

It seems like a common theme in the discussion is the need to increase the number of users. That seems like a worthwhile goal. Should we focus on that this year? I started a discussion topic on this in WeRelate talk:Next Steps. I hope you'll contribute your thoughts there.--Dallan 06:37, 25 February 2014 (UTC)

A better subtitle for such a discussion would be 'Barriers against growth' - the above discussion has me very convinced that the multi-lingual support (the lack of it, rather) is really holding back growth in other European markets. Daniel Maxwell 06:39, 25 February 2014 (UTC)
I agree, Daniel ! We absolutely need such a development. Rodovid did it ! We can do it ! And WeRelate is much better, more serious. In May and June 2008, I hesitated between WR, WikiTree and Rodovid. I chose Rodovid mostly because this site was multilingual. But since that date Rodovid has not changed while WR has really progressed. Congratulations ! I left Rodovid in April 2011 because the french module has been settled in a few months by incompetence in history and genealogy, by stupidity and vulgarity, by dictatorial not respect of the basic rules of a wiki (traceability + debate before decision). Nevertheless, we currently are only 2 or 3 French speaking contributors who "worked" about families from/in France and Belgium, CTfrog, JeffreyRLehrer and I. The biggest obstacle for others potential contributors is the only interface in English. A lot of people in France set their datas on Geneanet (very good "tool" some years ago, when no wiki site existed ... but today fulled with countless duplicates and incredible and massive errors). 2 editors of genealogy software (GeneaTique and Heredis) extended their competition by setting up each a data site (with Gedcom import) for their respective clients. Amicalement - Marc ROUSSEL - --Markus3 07:38, 25 February 2014 (UTC)

Help me understand more about what you envision with multi-lingual.

  1. One possibility is to make the interface multi-lingual -- so that menu items, form labels, etc. would be multi-lingual, but the data would remain uni-lingual, in whatever language the page was written in.
  2. Another possibility is to go with Wikipedia route, with completely separate pages for each language.
  3. A third possibility is to allow the biographical text to differ between languages, but the names, dates, places, facts, and sources to be shared among all languages. Wikipedia is headed this route with their Wikidata initiative, but implementing this third possibility requires a tremendous amount of effort.

My personal preference is for the first option; is this what others are thinking?--Dallan 16:06, 25 February 2014 (UTC)

I think that (1) is mandatory. I think that (2) is actually undesirable as it mostly makes sense for notable persons who would be candidates for Wikipedia - so for multilingual content link to Wikipedia. I think that (3) might be desirable but I would rather not expend scarce WeRelate development resources trying to make it happen. If eventually Wikidata manages to develop a suitable platform for doing this we could come back later and look at the upgrade. --Jhamstra 17:13, 25 February 2014 (UTC)
One other thing to consider is being prepared to administer a multi-lingual web site even for (1). Are we prepared to offer Help / Support / etc in multiple languages? It may be that (2) or (3) is needed for at least the "general content" pages. --Jhamstra 17:22, 25 February 2014 (UTC)
In this case then, perhaps we could slowly offer one language at a time (starting with French and Dutch, our larger foreign userbase is already in those 2 languages)? I could see a major problem if we added all languages and then in a short time, we have a ton of Cyrillic using languages users (be it Russian, Ukrainian, Serbian, Bulgarian, etc) and no way to offer them support ! Daniel Maxwell 18:23, 25 February 2014 (UTC)
I agree that (1) is mandatory, (2) and (3) only for general content (help pages etc.) --Klaas 18:00, 25 February 2014 (UTC)
I would add to this standardization of dates and differing input. A user should be able to select a setting for their preferred date input (American vs. European/International) so that 5-8-1900 is changed to either 5 Aug 1900 or 8 May 1900 based on those preferences. Also, a user could input the month or month abbreviation in their language and it is also standardized (ie Mei>May). --Jennifer (JBS66) 18:07, 25 February 2014 (UTC)
I agree with Hamstra, we do not want it like Wikipedia with a different version for every language; ideally, logging in another language would simply change the date formats (per Jennifer), but since the sources themselves ought to be in the language they were written in, there is no need to have multiple versions of those. I also think multi pages would be a support nightmare. But for the date thing to happen, Dallan I would highly suggest you implement my earlier suggestion to have it so that 1) WR has a standard date format, depending on the language 2) the system can snuff out bad/no dates/dates that do not conform to WR standards. Otherwise, the changing of dates/places between languages wouldnt make alot of sense. Daniel Maxwell 18:12, 25 February 2014 (UTC)

Thanks. I'll start implementing (1) now, and standardized dates once we have a decent implementation of (1).--Dallan 16:47, 27 February 2014 (UTC)

I support this as a near-term direction. Multi-lingual support is important (though not to me, personally) and the cross-cultural date formats can be stumbling blocks. Presumably technically unsophisticated volunteers will be able to contribute to the translations for a language module. Internal date standardization is necessary for good date presentation. --robert.shaw 20:08, 27 February 2014 (UTC)

Users by number of edits over time [25 February 2014]

Year-month Users with > 5 edits Users with > 100 edits
2007-01 26 10
2007-02 48 27
2007-03 78 46
2007-04 145 93
2007-05 78 36
2007-06 107 54
2007-07 113 55
2007-08 102 48
2007-09 294 142
2007-10 126 54
2007-11 152 73
2007-12 118 67
2008-01 122 55
2008-02 91 42
2008-03 97 44
2008-04 90 41
2008-05 96 54
2008-06 96 50
2008-07 105 50
2008-08 120 63
2008-09 101 52
2008-10 102 54
2008-11 128 68
2008-12 130 75
2009-01 184 99
2009-02 142 73
2009-03 160 79
2009-04 116 54
2009-05 107 57
2009-06 143 58
2009-07 133 72
2009-08 141 59
2009-09 127 65
2009-10 122 61
2009-11 107 45
2009-12 93 44
2010-01 106 49
2010-02 87 36
2010-03 136 55
2010-04 114 46
2010-05 118 47
2010-06 103 45
2010-07 116 51
2010-08 161 62
2010-09 136 68
2010-10 121 59
2010-11 147 55
2010-12 134 55
2011-01 160 55
2011-02 144 48
2011-03 164 70
2011-04 206 80
2011-05 162 73
2011-06 187 73
2011-07 194 81
2011-08 237 89
2011-09 196 73
2011-10 215 82
2011-11 197 83
2011-12 236 107
2012-01 231 104
2012-02 235 98
2012-03 225 94
2012-04 228 88
2012-05 216 92
2012-06 198 86
2012-07 231 88
2012-08 221 94
2012-09 196 77
2012-10 182 76
2012-11 196 77
2012-12 191 75
2013-01 210 97
2013-02 197 84
2013-03 212 89
2013-04 198 89
2013-05 185 77
2013-06 181 80
2013-07 184 79
2013-08 192 88
2013-09 189 90
2013-10 210 96
2013-11 199 98
2013-12 187 82
2014-01 208 88
2014-02 170 83
--Dallan 16:50, 25 February 2014 (UTC)
Thanks Dallan for this info I requested. It seems to confirm my suspicion that although the WeRelate database continues to grow and the number of people who look at this database may also be growing, the number of active contributors has not really grown over the past few years. --Jhamstra 17:06, 25 February 2014 (UTC)
In the course of trying to get people to use WR, I often hear it's 'too hard' or that they want control over 'their' data (ie page ownership). That sentiment hasn't seemed to change over time. I also think we probably don't want it to be too easy either, because of the early days when this site was nothing but dump and go GEDCOMs. In some respect, we're probably always going to have a higher bar for editing than any other site, but is that really a bad thing? Daniel Maxwell 18:27, 25 February 2014 (UTC)
In a way, yes. I think that easy editing can encourage people to make changes on-line. I for one know that my sources need editing, because they were once created with a program that put everything in what is now the source title. In those days, sources were just three lines of text, without GEDCOM tags distinguishing author, title, page number, whatever, and I would love to improve those with a word processor like tool, so that I can split that title that holds everything into the right parts without having to play cut and paste between fields like I have to do now. I don't like that part in Gramps, and on-site it's even more difficult, because pages are organized by title, so changing that means that every reference to that source must be corrected too. That's plain awful, and hard to improve on MediaWiki too, I fear. --Enno 18:52, 25 February 2014 (UTC)

Whether or What to Change? [28 February 2014]

For the past 2+1/2 years we have plateaued at around 80 moderately heavy users (>100 edits per month). I do not know how much turnover there is among that base of contributors.

Two or three others argue that the current situation is just fine. WeRelate is in a premium niche in terms of quality and should stay there. If in fact we can survive that way it might not be a bad idea. Presumably there is no point in discussing other possibilities.

While I certainly do not wish to see the quality of the content on WeRelate deteriorate to what is found on other more popular genealogy sites, I do think it appropriate to ask if there are things we could be doing better, beyond obvious incremental enhancements like date standardization and support for different language interfaces.

In this regard the recent survey results tell us about present moderately heavy users of the site. Assuming there has been turnover in this crew, I think it would be interesting to survey formerly active users to find out why they have backed away. If we want to attract wider participation then I think it behooves us to ask why people who once contributed (seriously - not drive-by uploads) have wandered off.

Reading between the lines of the traffic over the past 4 years on the Support / Water Cooler / Suggestions pages, I have noticed two recurring themes - difficulty with GEDCOM upload and exclusion of living persons. The first does not concern me personally because I have never created a GEDCOM and do not intend to (except perhaps for backup from WeRelate). The second, if properly supported, would be a convenience for me but is not a show-stopper (otherwise I would not be here 8-). Nevertheless, based on the number of times these issues have surfaced I think we have anecdotal evidence that they are important to a fair number of users who have spent enough time and energy at WeRelate to bother to ask for help or offer comments or requests in these matters.

Elsewhere I have suggested that we could tackle both of these limitations in a coherent and synergistic manner.

Could you point me to that page? I'm very interested.--Dallan 16:54, 27 February 2014 (UTC)
I have suggested a couple of rather different approaches to this problem in Next Steps. One involves layered Private and Public spaces, the other involves borrowing ideas from industrial PLM databases that must manage information of varying quality and maturity at a fine granularity, while facilitating collaboration. Either could work. (It happens that I spent two years working on the problem of how to collaboratively capture and scrub incomplete and even erroneous engineering data in the industrial context.) Given the amount of contention here regarding this topic I think it would be more productive to have this discussion offline 8-). --Jhamstra 17:35, 27 February 2014 (UTC)

Alternatively we could take the stance that (intentional or inadvertent) barriers to participation are actually beneficial because they help to keep the "undesirables" away (reminiscent of the argument we used to make in favor of winter in Minnesota 8-). Every social organization has its guardians of traditions - should WeRelate be any different? If long-time users had to endure certain "rites of passage" why shouldn't the newbies?

I guess this depends on whether we want to share our benefits with a wider audience or whether we want to be an exclusive club?

--Jhamstra 05:19, 27 February 2014 (UTC)

I think you've asserted, wrongly, a kind of xenophobia among the more active or heavy users here like myself toward newbies. But take a look at other genealogical sites, and ask yourself what is different between them and us. I would characterize most of the others as circuses - sure they have alot of people, but is it matched up in quality? The genealogical world is nearly universally derisive toward 'unsourced GEDcoms', but that is in effect that the others sites are. We all want growth, and that there is zero debate or disagreement on. The real problem is how we get there. We have alot of ideas that are already agreed to, but not implemented (languages, date standardization). So we have a bit of a manpower/developer shortage. We have to make baby steps toward this process. The fact that so many people commented or are watching this page tells me that the interest and health of this place is fine, but it could get better. I have been on the ins and outs of the internet genealogical world for awhile, and the things I have had said about WR is not gedcoms or livings, but that it is 'too hard' or that they want total control of the people in their tree (page ownership - I have seen this lead to duplicates at sites like Wikitree and even FindAGrave). Ultimately, most people are just lazy. They copied the people in their trees a long time ago, and once that process is over they probably rarely change it except to try and expand the lines. This open model is something different from other sites, and it actually makes WR better than the others. That's why I disagreed with most of your solutions to the living problem (hidden pages and de facto page ownership) as step backwards. Please don't mistaken my cautious and conservative approach as meaning that I am resistant to change; there are many features and changes I would love to see made. Daniel Maxwell 05:35, 27 February 2014 (UTC)
There is an enormous body of evidence that says that if you want to maintain quality in a scalable information system you need to consciously design the system so that it guides and incentivizes the users to do the right things - ie it should be easier to use the system when you do the right thing than when you do the wrong thing. Almost every suggestion I have seen along these lines (not just my own) seems to be met with the objection "that's not the way we do things here". In fact one person actually suggested that I leave unless I agreed to do a certain thing "their way". I am not easily dissuaded and I have stayed here and built a fairly complete representation of my family history and am working on those of other families of interest to me. But most prospective users are not as persistent as I am. --Jhamstra 05:55, 27 February 2014 (UTC)
Daniel, your comment about unsourced GEDCOMs applies to WeRelate too. It may not be obvious to you, but in areas that I recently checked, i.e. my own English ancestry, I saw quite a few profiles that are either completely unsourced, or refer to a tree on Ancesty, which is the same as unsourced for me. And I guess that, when I discard all profiles that have no reference to a real source, the amount of reliable profiles left is less than 50 %. And that's an optimistic guess, I think. Now let's assume that on Geni.com, only 10 % of the profiles are reliable. With more than 75 Million profiles there, that 10 % is still 5 times as much as the amount of reliable profiles here. Xenophobia is a strong word, but I do think that your comment is elitist. I'm smart enough to see which profiles are reliable, and which are not, so I really don't need you to value things.
Jim, yes, the right thing, that's quite complicated. Most people agree that sources are the right thing, but the source pages on WeRelate are quite complicated. The pages that you can create in the source box on FamilySearch have about half the number of fields, and FamilySearch allows direct links between person pages in the tree and their own indexed records. And adding those links is very easy too. When you're logged on there, you can attach any source that you have on screen to a person in the tree with a few clicks. You are automatically guided to persons with matching names, and you can override the search process by typing a name or ID yourself. Very nice, and because most sources over there are the real thing, i.e. evidence, not citation, it can lead to a quick increase in quality. Ancestry, Geni, and My Heritage offer similar things for a fee, and that helps. FamilySearch will also add a source walker in a few months, which when I heard things right will help matching or creating tree persons for all persons in a record in one go. Can you imagine creating a whole sourced family from a census record? It may not be the most accurate, due to the nature of censuses, but it can speed up the sourcing process quite a bit. FamilySearch has more than 12 Million sources attached to persons right now.
There's another issue with that righ thing. I'm not likely to do that, when I can't download any improvement that I make on-line to my own tree in Gramps. I upload my tree to local sites like Genealogie Online, and GeneaNet, and have no idea whether any site, except FamilySearch will still exist in 10 years. That is a very short time in genealogy, and just because of that, I really want to have reliable data in Gramps, simply because I expect that Gramps XML will be readable much longer than that, and data uploaded to FamilySearch too. I have more trust in their vault than in any archive held by a company or government.
--Enno 00:48, 28 February 2014 (UTC)
Um, you do know, don't you, that WeRelate supports Gedcom download? (I wouldn't have started contributing without that.) It's available under Trees menu item ("Manage Trees" page), "Export a GEDCOM" column. The exported items are determined by one of your tree setups, and that's inconvenient at times, but one can get the data. --robert.shaw 22:10, 28 February 2014 (UTC)
Yes, that's my own tree, with on-line mods, and that would be quite a large job to merge back into Gramps, checking out what really changed, and what's just reformatting by the WeRelate exporter. So much work, that it's not really practical, I think. But anyway, that is not what the competition offers now: Download the tree of any person that I watch, not what I uploaded, hence already have. --Enno 00:35, 1 March 2014 (UTC)
You can get what you want by defining an additional "tree" and put new watches into that one. That requires hitting "Tree" instead of "Watch", then clicking the tree name and an "Update" button, so it's not the greatest interface. (Seems like a relatively easy enhancement to allow a user to declare a default tree name for simple Watches to go into.) This other tree wouldn't have your own tree, so download wouldn't include your uploaded individuals. --robert.shaw 01:17, 1 March 2014 (UTC)
Nice! It's been a long time since I actually used this site, for reasons better expressed elsewhere, that I completely missed this feature. I will do some more testing on it, but it may work better than the GEDCOM download on WikiTree.
'Elitist' is a strong term - 'picky' would be a better one, but I wasn't really thinking of any site in particular when I said that, nor was I intending to denigrate real work done elsewhere; just the massive number of them out there. Geni could very well have reliable sections; but I don't know the ins and outs of the site. The English sections seem poor, but you said the site has better Dutch areas. I don't think there is any need to take this debate personally. I wasn't even sure that you and I really disagree with each other on what should happen here? " I'm not likely to do that, when I can't download any improvement that I make on-line to my own tree in Gramps. " - this is something else I'd like to see, ie being able to grab sections of the site to a gedcom or gramps xml. Probably needs a formal suggestion because I am not sure it was discussed before. Daniel Maxwell 01:01, 28 February 2014 (UTC)
Elitist is a strong term indeed, and I used it, because I felt like you wanted to turn WeRelate into a sort of gated community, trying to keep the circus out. And that is something that I feel like leading to a further decline. Referring to others as circuses sounds like ignoring the circus over here, and that's not good, IMO. There are good profiles over there, and bad ones over here, and I think that the only route to growth is to embrace the bad ones, try to improve, and educate. And by doing that, we are in fact elite, in the sense of people that care, speak out, and act.
For me personally, this means that concentrating on my own medieval English ancestry, I try to pick the best of Geni, WikiTree, and WeRelate, and improve where I can. I mention sites in that order, because that reflects the number of reasonable quality profile that I can find on each. Medieval genealogy is not easy, but I try to value profiles on their reference to reputed medieval genealogies and sites, or wikipedia, both of which have errors, but are still better than a reference to Ancestry that I can't check at all. For modern genealogy, Geni and WikiTree are like a circus though, because they sort or promote the use of married names, which I find confusing, and very unprofessional.
All in all, I think that we do need trees, not keep them out, no matter how bad, provided that we can somehow separate uploaded data from the wiki part that concentrates on information that is checked for reliability, and completeness too. To me it makes no sense to have a profile of a Mormon pioneer on-site, with only one wife (of 11) and one child (of 57).
--Enno 14:46, 28 February 2014 (UTC)

Starting a Parallel Project? [8 March 2014]

For a large number of reasons - technical, managerial and political (at least) - I'm concluding that WeRelate - is dying. It's not a happy conclusion. I've worn myself ragged since '07 - trying to improve data quality within the context of this community and this technology. As my efforts have become increasingly desperate - I've encountered diminishing support and increasing resistance. The joy of working on the site has been pretty much replaced by an ominous slog - interrupted every now and then by unpleasant encounters with those who dwell under bridges and an oversight process that really isn't up to the task.

Mind you, I don't think the effort has been for naught. I think WeRelate has shown us a lot of what we really want to see in the next system. I'm also hopeful that the WR database would be the starting point for that system's content.

I just don't think that incremental steps - starting from here - are going to get us there in any reasonable time frame. Living people, for example, can't be designed in after the fact. We need to think about some sort of partitioned/distributed or layered database from the start. Bots that wander around the data validating dates and looking for inconsistencies need to be part of the environment from the start. Semi-automatic citation generation created directly out of key source materials such as census transcripts, need to be part of the model. The current community we have today is just not going to understand those sorts of activities - much less accept them.

Is anyone else interested in trying to look at this implementation starting fresh, from current infrastructure?

--jrm03063 01:01, 6 March 2014 (UTC)

Whoa! Back! I think the current endeavor still has a bright potential future if we can agree on where we want to go with it. And I don't think we need yet another "fresh start" at a genealogy wiki. I would suggest that if you and Dallan (and possibly some of the other "geeks" on this site) could collaborate, we might get to a better platform sooner than if you go off on a parallel endeavor. It might make sense to "fork" the code base into a shorter-term effort that incrementally adds a few minor enhancements, and a parallel effort that builds the next generation of the platform. The current platform may be in "beta" but that need not prevent us from beginning work on the next "alpha". I seriously doubt that Dallan has the time to do both. But if we can work together then we could probably achieve a stronger long-term solution without leaving the user base behind and trying to build-up yet another competing genealogy "brand".--Jhamstra 17:42, 6 March 2014 (UTC)
Yes. The thing holding us back is that Dallan is very busy and there is no one around alot of the time to do upgrades. Requests/suggestions go unanswered. There needs to be more delegation of power here so these things can happen. Daniel Maxwell 14:13, 7 March 2014 (UTC)
Please !... and the multilingual option ? Amicalement - Marc ROUSSEL - --Markus3 17:51, 6 March 2014 (UTC)
Dallan says he is starting on a multi-lingual interface now. See end of 'Thoughts on growth' above. --robert.shaw 22:04, 7 March 2014 (UTC)
Marc, si ce site était sérieux au sujet de ces choses, "the multilingual option" aurait été fait il y a des années. Si vous voulez plus de comprendre pourquoi il n'existe pas de telles choses, jetez un oeil au point 1 b, ici. Peut-être que je suis injuste, mais je suis de plus en plus pessimiste.--Werebear 13:17, 7 March 2014 (UTC)
I think there is consensus that multilingual needs to happen sooner rather than later. So I assume it is layered on top of the existing code rather than waiting for a new code base. But Dallan will have to make that call.--Jhamstra 18:27, 6 March 2014 (UTC)
I could be persuaded that a "next generation" effort isn't entirely outside the scope of WR. I seriously doubt that it can be reached by dragging the existing community along in painful increments. I agree that internationalization probably isn't out of the question in the current code base - after all - you're just providing a different data table for the messages. Big deal. But anything fundamental? I don't think of myself as old - but I'm not going to waste more of my life teaching recalcitrant genealogists principles of computer science.
There are user communities out there right now - who won't work on WeRelate either because they can't (living) or they have reference/editorial standards that can't be inflicted selectively beyond their narrow group. Consider the Massachusetts Vital Records project. I talked with the site administrator over there a year or so back. He's interested in the idea of being able to bring that information together with the genealogy tree representation (something like what I tried to do w/Savage). However, he's not willing to allow just anyone to work on it - which may very well be a good thing. I'm convinced that a monolithic wiki representation (along with monolithic editorial and other practices) is hurting WR in ways we havn't even begun to appreciate (far more than just livings). I shouldn't have to fight people to create the annotated savage transcript with back links - I should be able to create a data partition specific to this reference, with appropriate methods and standards. A cooperating cosmos of genealogy partitions should be able to intelligently interact and bring together whatever is present in the different domains for any given Person. We need to START with a distributed representation - and see what it takes to reproduce current WR capability - not tinker with what we've got while wondering whether we'll ever figure out how to split it up intelligently. --jrm03063 19:51, 6 March 2014 (UTC)
I'm interested in hearing more. The discussion in Next Steps, among other things, has led me to question whether I can justify to myself continuing to put any serious time into contributing here, but I am still interested in working collaboratively. I had been musing about forks. I don't know which way I am going to go, but in future, a priority for me will to be to find a place to work on where the decision-making structure is not completely opaque to me.--Werebear 12:53, 7 March 2014 (UTC)
I'm interested in hearing more. The discussion in Next Steps, among other things, has led me to question whether I can justify to myself continuing to put any serious time into contributing here, but I am still interested in working collaboratively. I had been musing about forks. I don't know which way I am going to go, but in future, a priority for me will to be to find a place to work on where the decision-making structure is not completely opaque to me.--Werebear 12:53, 7 March 2014 (UTC)
Me too. I agree that distributed is the way to go, and monolyths are dead ends, which implies that I consider other sites mentioned elsewhere as dead ends too. But if this is true for a wider audience than the people writing here, the other issue is: What next? Users can already connect to the FS tree, with loads of clients, and Ancestry and My Heritage do the same with their own client software.
I've seen discussions on the RootsDev mailing list and a proposal to the FHISO, about which you can read more here. Dallan's in the RootsDev group too. --Enno 14:22, 7 March 2014 (UTC)
Ridiculous. You realize that some of us have put thousands of hours into this site, and all you can say 'stick a fork in, it's done'. But I noticed that the discussion at WMF is already sounding like a parallel project, which makes me think you'd like to kill off WR so it has no competition. Or I am misunderstanding you; I wasn't aware of resistance on Savage, though Savage itself has alot of old inaccuracies. Becoming frustrated with resistance and then saying you want to 'start over' is not really the best plan. I'd also like to know why this Vital Records Project admin doesn't want to let 'just anyone' work on it, that seems poorly explained. Daniel Maxwell 14:07, 7 March 2014 (UTC)
You may not like it, but as far as I'm concerned, WeRelate is no competition now. Finding better ways to share data, saveguard its integrity, are necessary to survive, I think. And ending the monolithic approach is a necessity too. Why waste time on copying pages from wikipedia? What does that add? Why send GEDCOMs back and forth? I'm not wasting my time on that anymore. --Enno 14:27, 7 March 2014 (UTC)
I have no objection to a conversation about what is better for genealogical data in the big picture of things, but that really hasn't been settled yet and I think it is still too early to count WR out. Daniel Maxwell 14:47, 7 March 2014 (UTC)
My comment applies to other sites too. I think that many lack the critical mass to appeal to users, and a site like WikiTree is just as much a niche as WeRelate. People do like sharing data, but as far as I can check, they do most of that elsewhere. --Enno 14:56, 7 March 2014 (UTC)
Yes, but I think WR is probably the only one of these willing to say 'no, that's wrong' and have corrections. Most of those other sites have either wrong or old information. I've been keeping an eye on the Rootsweb discussion since it started, so I am mindful of talks on the future of genealogical data. Daniel Maxwell 15:00, 7 March 2014 (UTC)
Regardless of my comment about niche, I do see corrections on FamilySearch, Geni and WikiTree, no less than here, so WR is definitely not the only one. WR lost me as an active member several years ago, and right now, I'm still figuring out where to go. It'll probably be a place where it's both easy to exchange information, and select what I accept as good. --Enno 15:14, 7 March 2014 (UTC)

Interesting that the users that were at one point worried about WR's future are now seemingly planning it's funeral! Daniel Maxwell

I've been trying to drag this beast along for almost seven years. I don't think we have a contribution counter, but you'll be hard pressed to find anyone (on the user side of things) who has tried harder, spent more time, and taken more chances to get us to "critical mass". I'm just not seeing that happen - and I don't think incremental changes around the edges or just a few more data entries are going to get us there. At the same time - instead of increasing interest - I'm finding technical and organizational/cultural barriers to growth to be an increasing impediment. I am disappointed in the present - but deeply anxious about the future.
I'm not trying to kill WeRelate - and I don't see why it would die immediately anyway - but I came here to find a place where my genealogy would be safe and free to all long after I'm gone. WR proper - the site - doesn't have to be that answer. Software changes and is replaced and updated over time - that would be normal. I would expect that. What concerns me is that WR isn't growing out of it's niche in a way that makes survival of THE DATABASE a sure thing. I can't afford to wait until I'm a doddering old timer, unable to use my computer science skills, to come to this conclusion. --jrm03063 16:19, 7 March 2014 (UTC)
Ditto. I don't want to see Werelate die. An enormous amount of my work is stored here. (I have never uploaded a GEDCOM. I have made I don't how many hand edits. --About 10 000 a month, according to my dashboard.-- I have barely even opened my personal database in the last year. Werelate had practically replaced that.) But at some point, I have to ask myself whether it continues to be reasonable to invest so much effort here, when long-term survival is looking to me less assured. Until recently, I at least had the hope that Werelate would grow into The Pando. Now, it no longer even has that goal. Or so it seems to me.--Werebear 18:14, 7 March 2014 (UTC)
I believe the answer is we need to figure out how to integrate with living. I'm going to add some thoughts on the WeRelate_talk:Next Steps page.--Dallan 20:33, 7 March 2014 (UTC)
Could we also, please, leave the option to join WikiMedia open? This most recent round of discussion has left me, like several others, wondering whether it's worth my time to continue to contribute to WR, but there just isn't anywhere else to go. We do, however, need to be more inclusive in a number of ways if we are to survive and grow.--GayelKnott 20:48, 7 March 2014 (UTC)
If you've been following the discussion over at WMF, it already sounds like they are talking about a different project. Things like 'all sources' only seemed to be on the table. That would end up making the site tiny and even more niche than WR is. I know some people had their heart set on WMF as a survival plan, but what they are talking about and what we have here have seemingly already diverged quite a bit. Daniel Maxwell 07:00, 8 March 2014 (UTC)
Yes, I'll address that as well in my comments on next steps.--Dallan
After reading all of the comments above, I have to think that people have different perspectives on what the purpose of this site is. Some think it needs to be a preferred site for the masses. Others not so much. Does this need to be the much exalted Pando? Probably not. (Is it even possible?) Does it really need to be so good that the average armchair genealogist prefers this site over Ancestry.com? Probably not. I do see this as a good place where people can collaborate if they choose to. And it is a good place to put genealogical work without having to go out and create one's own website which may ultimately die when the researcher dies. I do see this as a site for placing some scholarly work on both the famous and not at all famous of prior generations so that people using other sites can reference the work done here. Therefore, I do not subscribe to a Chicken Little view. Now back to my work.... -Moverton 01:16, 8 March 2014 (UTC)
I recently became an Ancestry subscriber to access English records that I can't access elsewhere from The Netherlands, and after seeing how easy it is to attach source records to persons there, I'd say, yes, WeRelate needs to be good. And that's because there are other places where collaboration is much better for me, because of tree size (Geni, WikiTree), ease of use (Ancestry, Geni), and so forth. WeRelate was the first wiki like site that really appealed to me, but I moved on years ago, because of a couple of issues that are hard to address. Anyway, there are still a few areas where WeRelate can stand out, and that's why I'm writing here. --Enno 13:59, 8 March 2014 (UTC)