Is the user "TreeBuilding Project" taking the tree forward or wasting time?
Answers
-
This thread is appearing as last updated Sept 19, and therefore is showing further down the recent discussions list than it should be. @Ashlee C. is this a known problem, maybe with big threads? This is still an active and in my view important thread, I am happy to split it if that's what you recommend.
0 -
Re GS9C-VTF.
The NUMIDENT source was added by USCensusProject on 18 September, and the Name. Birth, and Death tagged with it; but after that the name on the profile was still just 'John', and the Birth and Death information was still not present.
On 19 September CommunityCensusProject3 came along and set the Birth and Death dates correctly.
I have just set the Name to match the source, so no further work is required here by BYU RLL, but surely it can never be good practice to tag a source without setting the relevant profile information accordingly?
Also, why is this very much NUMIDENT focused work being done by 'Census' usernames and not by 'TreeBuilding'/'treehelper' ones?
1 -
Re the collections involved.
Numbers of interventions by the BYU RLL automated users or the 'treehelpers' in my analysis database of around 60,000 profiles, by collection, including only activities related to 'Census' or 'NUMIDENT' collections and still underway in 2024, and not necessarily right up to date:
user
collectiontitle
fromDate
toDate
number
treehelper
'England and Wales Census, 1911'
01/03/2024
28/05/2024
24
treehelper
'New York State Census, 1905'
09/09/2024
09/09/2024
2
treehelper
'Rhode Island State Census, 1915'
09/07/2024
09/07/2024
2
treehelper
'United States Census, 1870'
13/08/2024
21/08/2024
4
treehelper
'United States Census, 1880'
23/08/2024
23/08/2024
23
CommunityCensus Project
'United States Census, 1900'
17/08/2020
18/07/2024
358
USCensusProject
'United States Census, 1900'
06/08/2022
10/07/2024
205
treehelper
'United States Census, 1900'
02/03/2024
17/07/2024
46
treehelper
'United States Census, 1910'
27/02/2024
09/09/2024
7
treehelper
'United States Census, 1920'
13/11/2023
12/09/2024
77
TreeBuilding Project
'United States Census, 1920'
15/06/2024
17/06/2024
21
treehelper
'United States Census, 1930'
06/05/2024
16/09/2024
141
treehelper
'United States Census, 1940'
05/03/2024
02/09/2024
76
treehelper
'United States Census, 1950'
20/08/2024
09/09/2024
16
treehelper
'United States, Social Security Numerical Identification Files (NUMIDENT), 1936-2007'
08/09/2023
18/09/2024
1699
TreeBuilding Project
'United States, Social Security Numerical Identification Files (NUMIDENT), 1936-2007'
04/01/2024
19/09/2024
381
USCensusProject
'United States, Social Security Numerical Identification Files (NUMIDENT), 1936-2007'
16/08/2024
16/09/2024
103
CommunityCensus Project
'United States, Social Security Numerical Identification Files (NUMIDENT), 1936-2007'
15/08/2024
12/09/2024
55
-1 -
Just a warning that the thread linked to below has updates well after the 19 September date given in the discussions list.
0 -
@MandyShaw1 The same behavior of falling off the list of active threads was noted on one about the CensusProject(s). I have to think it may be done by design.
4 -
Hmmm.
0 -
I had to dig to find that old thread:
1 -
We need to escalate this … I asked FS EU Support for contact details for the FS Chief Information Officer, but they said there is no such individual & that anyway they wouldn't provide personal contact details (which isn't in my view what I had asked for). However they did say that if I provided more details of the issue they would see what they could do, so I am putting together an email (focused on the communication problem, in the first instance), which I will obviously paste here for any comments before I send it.
2 -
I have updated my summary at Numident summary.pdf to reflect where I think the discussion has got to.
In particular it now discusses BYU RLL's use of the standard FS duplicate algorithm, and the idea of showing the general public records rather than FT entries.
The document doesn't mention the cross-thread problems because in the end that was clearly a FT source linking problem that impacted BYU RLL along with everyone else.
As before I'd be very grateful for any comments.
0 -
I have posted on the original TreeBuilding Project thread suggesting any further discussion happens here instead.
0 -
Here's my proposed email to FS EU Support - please comment as wished and I'll send it off in a few days.
******
There has been much recent discussion on FamilySearch Community concerning the Census and NUMIDENT projects that have been undertaken over the last few years by BYU’s Record Linking Lab, and that remain active.
These projects use automation and teams of student and other volunteers to attach high volumes of sources to the Family Tree and to create large numbers of FT profiles.
The strong concern is that these projects have resulted in large numbers of FT integrity problems which have seriously impacted skilled and experienced FT users in their day to day work, thus also affecting perception and morale.
There appears to be no communication channel in place that would allow us to report these issues to FamilySearch to get them resolved.
Please advise how we can contact an appropriate team in FS to discuss this situation.
5 -
I think I'll take out the 'recent' at the beginning of the email text, the discussion here on Community goes back nearly as far as the problem, I think.
1 -
@MandyShaw1 - I will warn you that I am incapable of writing short explanations so feel free to ignore this comment but…
Would it be useful to mention that, as an example, some of the profiles created by BYURLL don't follow recommended best practice with respect to using maiden names versus married names, with the result that detection of possible duplicate profiles is impaired?
Firstly I think but could be wrong that the above is a fair comment. Secondly, feel free to say this is just one aspect so it is a bad idea to highlight just this as it's already covered by "large numbers of FT integrity problems".
1 -
Good point, I have added this to the end of paragraph 3 in my draft: 'Just one example of best practice not being followed is the frequent creation of profiles using women’s married surnames.'
2 -
As we have read, someone thinks that is a good thing.
2 -
Indeed - but if anyone else is to read this, it should be mentioned as a point of disagreement with presumably inevitable issues unless someone changes.
2 -
@MandyShaw1 We aren't seeing the issue with this discussion thread. In the Community we try to keep all relevant information together. Your new post has been merged into this discussion to keep all the relevant information in one place. We will keep an eye on this discussion for the issue you mentioned.
0 -
@Ashlee C. the comments from the 'split off' thread (which was quite active) are not showing here and your comment above didn't cause this thread to go to the top of the discussion list (and nor did this one). Could you put it all back (or in some other way make all the content visible, with update handling as normal) please?
0 -
According to the view that I have, it says
168 comments
Most recent by MandyShaw1 September 23
Family Tree
Mandy's last comment above is dated today (24 September) at 14:50. As the current time on my screen is 17:03, I can't see any likely way that it was last updated on the 23rd.
Please note @Ashlee C. and @MandyShaw1
2 -
I can still see some of my disappeared comments listed in my 'like reactions' view, but if I click on them it says 'No Comment Found'. I can't see any trace of them anywhere else, so it's not clear to me what is happening, but I would guess the forum software limits the number of comments in a thread, or the total size of a thread perhaps, and breaks in unpredictable and unhelpful ways when threads go over that limit.
Edit: The originating post in my 'split' thread is appearing here, but none of the comments I or anyone else added to it are present.
@ notifications do still seem to work in this thread, which is something.
0 -
Here, in case the missing comments don't reappear, is the latest Support email draft. If I don't receive any further comments I propose to send it on Thursday afternoon (UK time).
******
There has been much discussion on FamilySearch Community concerning the Census and NUMIDENT projects that have been undertaken over the last few years by BYU’s Record Linking Lab, and that remain active.
These projects use automation and teams of student and other volunteers to attach high volumes of sources to the Family Tree and to create large numbers of FT profiles.
The strong concern is that these projects have resulted in large numbers of FT integrity problems which have seriously impacted skilled and experienced FT users in their day to day work, thus also affecting perception and morale. Just one example of best practice not being followed is the frequent creation of profiles using women’s married surnames.
There appears to be no communication channel in place that would allow us to report these issues to FamilySearch to get them resolved.
Please advise how we can contact an appropriate team in FS to discuss this situation.
0 -
@Ashlee C. any news re getting this thread working correctly?
0 -
I've sent the email and will keep you all posted. Thanks a lot for the likes and comments (though we can no longer see them).
2 -
The missing comments are back! Thank you v much @Ashlee C.
0 -
https://community.familysearch.org/en/discussion/comment/570291#Comment_570291
This thread is still not updating correctly in the list of threads. Here is a screenshot from just now that says it was last updated on Sep 23, which is now 3 days ago, even though several comments have been left each day between now and then. @Ashlee C.
(screenshot seems to not want to render, so just in case, it is image sH1S3N4.png on imgur)
Edit: Now that I have made this post, the display shows "Most recent by JD Cowell September 23 [2024, 9:48 AM]" (timestamp information from tooltip shown on hover). That timestamp exactly matches @Adrian Bruce1's second-most recent comment, suggesting new posts are not registering for some reason, whether that is because of the thread merge or something else. But this thread is still on page 2, even with new activity. Why?
1 -
I got a 'following' email for @JD Cowell's post, so that aspect seems fixed also.
The timestamp/list positioning problem started before I split the thread, so I don't think it's anything to do with the merge.
0 -
Well I have had an answer. I am advised to use the Community Suggest an Idea to provide feedback which will be 'monitored by the engineers'. Once I have counted to ten a few times, and since they've asked, I will draft something and post it here first for review, unless anyone else wants to do it? I think I will also reply pointing out that the lack of a communication channel is really not the engineers' problem, it's management's, but that we will try their idea before requesting escalation (I'll post that text here for review also).
Edit: Suggest an Idea has a message indicating it's 'under construction' and that ideas won't appear yet, but the message includes a link to a new set of pages on the subject including an option to submit, and the latest idea listed is dated 18th September (I don't know whether it was submitted via the old or new interface, obviously). Does anyone know what's going on there?
1 -
Not wanting to be negative about their suggestion but haven't we been told that the Suggestion forum is suspended while they sort out a revised process?
1 -
I have no problem with being negative about their suggestion, personally …
0 -
Proposed Suggest an Idea (a lot of the text as before, bolding intentional, I'll add links to this thread and the USCensusProject thread when I post it):
******
There has been much discussion here on FamilySearch Community, in particular [here] and [here], concerning the Census and NUMIDENT projects that have been undertaken over the last few years by BYU’s Record Linking Lab, and that remain active.
These projects use automation and teams of student and other volunteers to attach high volumes of sources to the Family Tree and to create large numbers of FT profiles.
The strong concern is that these projects have resulted in large numbers of FT integrity problems which have seriously impacted skilled and experienced FT users in their day to day work, thus also affecting perception and morale. Just one example of best practice not being followed is the frequent creation of profiles using women’s married surnames.
There appears to be no communication channel in place that would allow us to report these issues to FamilySearch to get them resolved.
I have asked FS Support how we can contact an appropriate team in FS to discuss this situation, and they have advised posting here in Suggest an Idea, where the right people will apparently have visibility of our concerns.
Given the seriousness of these issues, we are hopeful that this will allow FS to take action in a timely manner to provide us with a suitable communication channel, but if this is not in fact the correct place to request this assistance please let me know.
0
