RootsTech 2026 'Tree Integrity' session - does anyone have the slides/syllabus?
This session (https://www.familysearch.org/en/rootstech/session/tree-integrity-strengthening-the-familysearch-tree) wasn't available online and doesn't seem to have been recorded, and I can't even find the slides/syllabus.
This is disappointing, especially as the brilliant 2025 data quality session was fully available to online attendees.
This material would, I suspect, be very useful to regular posters here on Community.
Does anyone know where at least the slides/syllabus might be found?
Thanks.
Answers
-
I attended that session in person. There were no slides available on the RootsTech website or app. That is disappointing.
0 -
@Alan E. Brown thanks. Were there any key or hot-off-the-press points that you feel able to share?
0 -
There are 2025 videos:
https://www.familysearch.org/en/rootstech/session/improving-familysearch-family-tree-integrity-understanding-the-data-quality-score?cid=rt_copy
https://www.familysearch.org/en/rootstech/session/tree-integrity-protecting-your-research-in-familytreePerhaps the 2026 version(s) will eventually be put online.
4 -
In the 2026 RootsTech session, there was a lot of discussion of merges:
- As we know, bad merges are the source of many problems with tree integrity.
- FamilySearch is continuing to work on improving the merge experience by detecting factors that could indicate a bad merge, and presenting them to the user so that they will be discouraged from doing a bad merge.
- Undoing bad merges (after additional edits have been made) is a thorny problem, but continues to be a significant area of focus. It will still take some time to get this production-ready.
- Work is ongoing to improve how the Change Log shows the results of merges so that you don't, for example, see ”Relationship Deleted" without a clear connection to the merge.
Another area of focus is preventing bad edits. This is leveraging the Data Quality Score technology. If a proposed edit would lower the DQS (because the new value would not match the values in the tagged sources as well as the existing value), then the user will be discouraged from saving that edit.
4 -
@Alan E. Brown thank you very much for this.
0 -
Yes, thank you very much for sharing this with us.
The creation of conflated profiles through bad merges is a constant concern of mine. Firstly, because of the fact that so many individuals that have been diligently added to Family Tree have then been effectively treated (at least as far as the tree is concerned) as though they never existed! But, secondly, because of the extraordinary amount of time and effort that can be involved in restoring the respective profiles to their original state. For example, I'm sure there are many inexperienced users who do not realise that restoring a profile does not mean that the sources and facts attached earlier to the "surviving" profile (in the merge) then have to be manually removed from that profile - otherwise, they remain attached to both of the profiles involved.
It is pleasing to know the developers are actively seeking to improve this aspect of Family Tree, by focusing on steps that might help diminish the widespread problem of users making incorrect merges, as well as (hopefully) making the effort of reversing the process rather less arduous.
3 -
@Alan E. Brown - likewise, thanks for the insights.
One aspect of bad merges has probably occurred to many of us, but maybe not all - unravelling a bad merge may not simply be a matter of reviving a merge-deleted profile and reverting the source record attachments to their previous state.
Quite recently I had to unravel a bad merge and, having done so, I was double checking all source record attachments. I found that one of the attachments, which was now in its "correct" pre-merger state, was, in fact, attached to the wrong profile. (Same name but two or three parishes distant).
Thinking about it, this attachment, done years before the bad merge, had, in fact, set the scene for that bad merge because it had attached a source record for the wrong parish. This had made the bad merge much more geographically plausible. The root cause of the bad merge, therefore, was arguably not the merge itself but a poor source attachment years before. I'm not sure where any improved checking needed to be for that one. Everywhere probably!
5 -
I would have liked an update on DQS strategy beyond the 'preventing bad edits' piece (which I was delighted to hear about from @Alan E. Brown's notes above). In particular it would be good to understand how the DQS will work over time with the 3rd party products (at present I don't think the APIs are DQS-aware at all).
0 -
Robby Parker gave a shorter version of this information in the Global Tech Forum (https://www.familysearch.org/en/rootstech/session/familysearch-global-and-tech-forum-2026) starting at 18:58.
2 -
Yes, but it was /very/ short (and iirc Robby recommended people attend the session that is the subject of this thread - perhaps not knowing that only in-person attendees would have access).
0 -
I raised the lack of online materials for this specific session on my RootsTech feedback survey.
This is not the only in-person session with such problems but it's the only one I have encountered that doesn't even make the syllabus available online.
I also raised the complete lack of any 2026 session about indexing or Get Involved - the AI-related sessions I had access to were uniformly excellent (Steve Little, in particular, is an exceptional presenter), but CAI was not mentioned by anyone.
0 -
The slides for this session are now available on the RootsTech site: https://www.familysearch.org/en/rootstech/session/tree-integrity-strengthening-the-familysearch-tree
1 -
@vanwagenenl That link doesn't go anywhere useful — there is no such session on the RootsTech site.
0 -
@Alan E. Brown I accessed it a few moments after vanwagennenl's comment, and I was able to download the slides. I see that the presentation is no longer available.
0 -
Apologies. The slide decks and syllabi for all of the in-person sessions are being made available here: https://www.familysearch.org/en/rootstech/events/syllabi.
A direct link to this slide deck is here: https://cms-z-assets.familysearch.org/c7/78/1506743d40f2a2420f45130d6d29/slides-for-tree-integrity-strengthening-the-familysearch-tree.pdf2 -
@vanwagenenl thank you very much for this.
0
