Suggestion to improve collaboration and document quality of record in Family Tree
Comments
-
joe martel said: An UI affordance is something like an icon a user can click on to activate something - like the hazard tape icon/affordance activating a popup.0
-
m said: I had never understood why people kept adding this wrong middle name to this man over and over until today when someone added the wrong middle name and the wrong wife, so I was able to look the couple up and LO AND BEHOLD a real couple.
Just 2 men with similar names.
I have now figured it out for everyone---so it would be great to preserve the knowledge for the benefit of everyone---with some sort of warning.0 -
m said: The merge one looks really good. Plenty of room on the merge page for anything.0
-
m said: yellow black "refuted parents" click to see yellow black and then when click and you see what is entered into the template parents so and so are refuted by his marriage record in which he stated who his parents are.0
-
m said: yellow black "2 people often confused click to see " yellow black and when you click to see what is entered into the template "so and so b 1602 m to such and such often confused with other person (similar name) b 1639 m to other person"0
-
Juli said: I like the caution tape idea, but it should be vaguely square to match the other "flags":
0 -
Jeff Wiseman said: I really hope that FS is not seriously considering adding user generated "stickup" warnings all over the place.
After all the discussions in this forum about Notes and how they are being diffused through so many alternate places where the same exact things can be documented, nobody knows where to put specific things or where to go to find specific things. Right now in addition to the Notes feature, we put exactly the same things into Life Sketches, Discussions, the schizophrenic "Reason for Change"/"Reason this info is correct" boxes, Custom Facts, and as a result, the normal Notes function is being totally diluted and ignored by many folks including FS engineering. Chaos in the area of note handling is continuing to increase significantly in the FSFT.
And now we're discussing yet ANOTHER way for anyone that wants to, to create as many "stickup and shove into your face" type notes that they want, thus diluting the concept of organized and centralized notes even further? Really?
If FS truly considers doing this, then they MUST provide a switch so that it can be optionally TURNED OFF COMPLETELY if a person wants. System warnings about something you are doing have always been perfectly acceptable because they are based on fixed operations of the system. But what makes somebody else's personal OPINION so important that I am forced to read it and have it totally clutter up what are already screens that are far too "busy" and confusing for many? I have seen way too many "Do Not Merge" notes in discussions or life sketches that were just plain wrong. The person putting them in there was so positive that they were correct, but they were not affected when faced with the sources and other evidence available.
And then we will have yet one more area of contention between folks with opposing opinions. Will FS put all of these user submitted "warnings" (a.k.a. notes) under change history logging so that when people change things on the system, there will be a history that they must account for? Can I delete all of the warnings that get in my way?
This will be another feature that is easily abused. Do we really want to give even more power to individuals who don't feel like "playing by the rules"?
Anyway, I believe that this is an overkill mechanism for people to advertise their own opinions in a "billboard" type manner. We already have other mechanisms in place that can be and are being abused enough to achieve the same thing at present (e.g., Life Sketches, Discussions, etc.). Please, let's not add yet another level of entropy to the web site!0 -
Juli said: I wonder if we could kill two birds with one stone: rename the "collaborate" tab to "Notes" (because that's what it is, really), add tagging to that tab (same mechanism as on Sources), and move the Life Sketch to the top of that tab. Even without a "caution tape" icon, if relevant Notes showed up whenever you went to edit a vital or merge two profiles, people might actually notice them...
On consideration, I do think Jeff is right: user-added flagging needs controls. Perhaps it would work to have a single place (on the Notes tab) for a user-entered Warning/Caution text, with its existence indicated by a single icon-and-label at the top right (with the research suggestions, data problems, and record hints)?
0 -
Jeff Wiseman said: Yes, that's the type of thing I've been talking about for a while. Put Notes into the Notes, and then create useful mechanisms to make them visible in other places (e.g., tagging to vitals). In fact, just have a default note with the title "Life Sketch". After all, it is just ANOTHER NOTE.
Even without a "caution tape" icon, if relevant Notes showed up whenever you went to edit a vital or merge two profiles, people might actually notice them
Yep. A warning could be nothing more than a carefully titled note with the details of the warning inside. So when a person opens the edit on the Birth vital, he would see the list of Notes (some possible "warning notes") that have been tagged to it all listed by title right there. Just like sources, you have the titles to give you the summary of the contents, with the details being the contents of the source.
Again, vital and person record notes (e.g., NOT change history notes on the reasons a change was made--that is a totally different part of the system) all belong in a central list for the person record. Everything there. no need to search all over creation to find reasons for vitals and person records.
The system just needs to make those notes easily available to areas that they may be needed in (e.g., via tagging). Designing a bunch of custom note handling features, all different, in general makes no good engineering sense. When I need to make a note of some kind for myself and others, why should there be 6 completely different mechanisms to do it with when a single optimized system would work just fine.
BTW, the GEDCOM X standard basically states that (paraphrasing here) anything (e.g., a sources, proof, or note) that is associated with a vital, must also be automatically associated with the person record as well. That's just another reason why each vital should not have it's own separate list of notes. They should all be attached to the Person record (as notes are right now) and from THERE associated to the individual and pertinent vitals for that person.
(of course this also has ramifications of relationship entities each having their own private and separate list of sources and notes independent of the persons that they relate together--but the tagging of sources to couple relationships and such is another separate subject that seems to be getting ignored)0 -
ATP said: Jeff,
As usual, you hit the nail(s) on the head, and as you say, if FS implements this "stickup" warning, it should provide a SWITCH OFF. If such were not able to be switched off, my whole work in progress on my family lines, alone, would be full of those signals! Hurts my eyes to even think about it!0 -
joe martel said: I mocked those things up trying to illustrate what I think was being asked for. I'm not a fan of those mockups. It is not pretty, it is busy and clouds the information. THere has to be a better way for a warning but not be overwhelmed by it. Please feel to mock up - I did these with screen shots and powerpoint.0
-
Adrian Bruce said: Warnings that can be turned off seem profoundly illogical.0
-
Adrian Bruce said: The purpose of mockups is to provoke comment and Joe did that, thanks! :-)
I like Juli's suggestion of putting a flag about the existence of Caution notes into the Research Help. That's where we find warnings now, so it would make more sense to use that rather than yet another method.0 -
Adrian Bruce said: "the GEDCOM X standard basically states that (paraphrasing here) anything (e.g., a sources, proof, or note) that is associated with a vital, must also be automatically associated with the person record as well."
Ooooh. That's a bit odd. That's a fundamentally totally redundant association! It might make sense to present it like that in the user interface, because I quite like the idea of a list of notes that can be tagged to appear elsewhere. But in the underlying data model? It may be that Jeff's paraphrase has oversimplified things??0 -
Juli said: (This is Juli's sister, the occasional DB admin, temporarily stealing her account to chime in) If it uses a relational database model, then the transitive association is an inherent part of the design - there's nothing extra that needs to be done, and in fact there's no way to enter the data *without* the association being transitive. (By "transitive" I mean that if table B is linked to table A, and table C is linked to table B, then table C is also, ipso facto, linked to table A.)0
-
Adrian Bruce said: My understanding is that FS FamilyTree isn't based on an RDBMS - however, my feeling has always been that it made sense to construct a logical model using at least the basics of that sort of entity relationship modelling, and that therefore at the point of doing a logical analysis of the entities, then the transitive (crikey, I'd forgotten that word) nature of two relationships like that applied. In fairness, once you get to the physical, all bets might be off, but instinctively it still seems a little odd to me to have it in the interface spec'n. Hence my reaction.
NB 1 - I don't remember what phrase describes FS FT's "DBMS"
NB 2 - when I say "at least the basics of that sort of entity relationship modelling" I only refer to the first 3 of Codd's Rules, 1st thru 3rd Normal Form.0 -
joe martel said: My answer has no bearing on how it should work or what can be done. Much of FSFT is based on a distributed noSQL cassandra database. That is to support the scaleability and performance needed to support the projected 10-100x growth. In 2017 it was handling over 100 million transactions per hour.
- BTW: I loved Adrian's saying Joe's desire to "provoke comment." It was
0 -
m said: If warnings work for Werelate and Wikitree, why WOULDN'T they work for FS?0
-
m said: How would this idea work with the Pieter Classen examples on Werelate & Wikitree, for example?
https://www.werelate.org/wiki/Person:...
https://www.wikitree.com/wiki/Claesz-4
" Refuted Father?: Claes Van Schouw (1) "
"disproven claim is that Pieter's father was Claes Corneliss (van Schouwen) "
"See posts and see G2G and G2G discussion, What is the origin of the surname Wyckoff and who were Pieter Claesen Wyckoff's parents?"0 -
Jeff Wiseman said: Adrian and Juli’s sister,
Yes, all entities are related to all other entities (just as everyone on the earth is related). But these are mostly all indirect relationships (i.e., via transitive relationships as Julie’s sister pointed out). A vital is essentially an attribute of the Person record. I.e., their relationship can be viewed as the vital being “contained” by the person record. Therefore, a source that supports a vital conclusion also supports the person record that contains that vital.
I think that the GEDCOM X requirements has to do with a requirement of the actual manifestation of that indirect association. So in the example given by Juli’s sister, A is associated with C. But that is only via all of the intermediate rules associated with each of their associations with B.
I.e., if the relationships between A and B are limited by the rules 1 and 2, and if the relationships between B and C are limited by rules 3, 4, and 5, then any manifestation of the association between A and C are limited to all permutations of the rules 1, 2, 3, 4, and 5.
Anyway, Although I would need to study it more, the GEDCOM X reference seems to be implying that the “transitive” relationship that exists between a person record and (say) a source or note supporting one of the person record’s vital conclusions is important and should be explicitly provided in an actual implementation. This is how sources are currently supported in the FSFT. Notes are only partially supported at present as their actual relationship to various vitals are not explicitly supported (e.g., a note is not tag-able to a vital the way sources are).0 -
Jeff Wiseman said: Adrian,
BTW, the concept of a logical vs. a physical model was introduced in the early 1970s by Ed Yourdon in his structured analysis and design methodologies that form at least part of the basis of nearly all CASE (Computer Aided Software Engineering) tools out there.
However, that terminology frequently created confusion. Engineers using the methodologies were (incorrectly) refusing to put physical components into their logical models. However, as most software engineers are aware, requirements (i.e., the logical model) for enhancements to legacy systems ALWAYS contain physical components around their contexts. I.e., when updating or enhancing a pre-existing system (which is the larges portion of all software projects these days), requirements CANNOT IGNORE assumptions and facts about these physical constraints.
So in the early 1980s, Yourdon renamed the logical and physical models as the essential and implementation models since those were far better descriptions of what they truly were. Those terms have stuck for the past 4 decades as they work so well with the essential model being the requirements and the implementation model being the design.0 -
Jeff Wiseman said: Adrian,
What if the "warnings" are not even true? What if they are nothing more than someone wanting to force their opinion onto the site pages so that everyone sees it? So if you consider this, your comment might be phrased as:
"Opinions that can be turned off seem profoundly illogical"
...and in my opinion, that statement is illogical :-)
We need to be able to train people to review all notes regarding any change they are going to make and decide for themselves. To me, opinion based warnings are based on the assumption that nobody else is as smart and can interpret notes and sources as well as the person posting the unblockable warning.0 -
Adrian Bruce said: But I would suggest that there are two concepts -
1 "I'm so smart that I don't need to read anyone else's opinion, so I'll turn off the warnings from these little people" (A deliberately tongue in cheek provocative way of putting it!)
2 "No actually, your belief is wrong because I've got this source and that source and they all show that John Doe did cross the Atlantic.... And here is all that evidence."
I don't have any problem with Case 2 turning off a warning - in fact, the ability to ensure such a suppression of warnings is essential.
The ability to do a Case 1 suppression is what is illogical.
But, but, but.... How can the FS software distinguish the 2 cases - you got me there!0 -
Adrian Bruce said: Interesting Jeff - "essential" and "implementation" models don't ring any bells with me. My assumption would have been that the preexisting stuff was in the Constraints. Obviously my employer never bought any courses from a company who'd read 1980s Yourdon.0
-
Adrian Bruce said: Jeff - re GEDCOM-X explicitly documenting the transitive relationship even though it's logically redundant at first glance - yes, there might have been some documented and justifiable reason. On the other hand, having seen how the stuff was developed, I remain to be convinced!0
-
Pioneer42 said: Not going to happen! You know why? Because the USERS DO NOT READ ANYTHING, NOR DO THEY CARE anything about there ancestors (they treat them like they are just a peice of candy, eat it and throw down the wrapper, and say NEXT!). There job is to go in and destroy and not clean up. The only way forward is a 3rd party arbitration army to control the data AFTER THE PERSON HAS BEEN MERGED FIXED AND SOURCED AND LOCKED! And NO MORE GEDCOM UPLOADS ALLOWED! If this doesn't happen then all that you stated is useless. Trust me coming from a seasoned veteran that has 6000 hours + on this program. Plus the church has now just changed the LATEST CHANGES screen so that now its is a giant mess, don't even know where to begin. Corrections now are the biggest pain it has ever been.0
-
m said: Today it has been extremely difficult to make correction.
Something about the Changlog is difficult to read. Format? Font? Blue/pink distracting? Location of user that made changes? Having to search so hard for previous entries? It is just much harder to read.0 -
Jeff Wiseman said: It's not like the guy invented the words :-) They were already there and he (and others like him) just started using them because they were more appropriate. Check out the dictionary.
Now there are equivalents when dealing in the IT field. For database requirements and design, the equivalent to the "Essential model" and "Implementation model" would be the Information model and the Data Model respectively. Those terms are more common in the IT world. However, there are models that Business analysts use where the "essential" and "logical" terms are used. Unfortunately, their evolution doesn't always mesh well with the general dictionary meanings. So to each their own.0 -
ATP said: Pioneer 42,
If you are not familiar with the seeding of the current version of FSFT, the following summarizes the history before and after of that seeding and the corresponding pitfalls that were inherited and continue even to the present time. And, even if you are familiar with this history, as the author says, it's worth a review. He follows the review with how best to perform clean-ups and presented in the following link posted yesterday, 26 May 2020.
https://genealogysstar.blogspot.com/2...
It's my understanding, he is not a close kinsman of Ron Tanner, though they share the same surname.0 -
m said: So a warning can be added to any location and it will be seen by people visiting from their cellphone, from Ancestry, from their desktop genealogy program?0
This discussion has been closed.