* Advisability of changing Census Roles from default options

[Gary_G]

In porting my data to FH7, I am going over my census "roles".
It appears that FH7 comes with the option to label the head as "Head of household".
However; the other options are "Resident" or "Witness".
There must have been a good reason for making the default, "Witness".
I'd like to create roles to match what is used in the census; Head, Wife, Son, Daughter, Boarder etc.
I had these in RM9, but I also had to adjust the sentence logic so that the sentence read correctly (adding gender pronouns etc.)
I haven't yet seen enough of the reports or addons to know if changing these role names will cause problems later.
Has anyone any advice on this?

[jbtapscott]

Just to clarify, does this mean that you are using FH Witness Roles for each person on a census record as opposed to giving each Individual their own Census fact?.

[tatewise]

In my FH v7.0.19, the standard Census event only has Head of Household and Resident as Fact Witness Roles.
Anyway, there is no concept of a default Role.

Users who have tried defining a Role for every relationship find that they either have to keep inventing new Roles or need a catch-all Role for all the unusual relationships. Some have decided to use Fact Witnesses to enhance Narrative Reports as well as having Census events for every household member.

See FHUG Knowledge Base Recording from a Census Record that explains the pros & cons of Census events for each household member versus Fact Witnesses. Although, the Lookup Missing Census Facts plugin now does handle Fact Witnesses on Census events.

[Gary_G]

"jbtapscott" and Mike;

RootsMagic has the concept of shared "facts", which are quite commonly used for a census entry. I'm not exactly sure how to they ought come across into FH7. It seems that these are imported to FH7 as separate facts for individuals, but are linked to each other. When I double-click on the census fact in any participants list of facts, it brings up a panel showing the people involved in the event as witnesses and allows me to go to their properties box. Based upon the displayed sentence, it appears to know which person was the head of house. When I say that "Witness" appears to be the default, it is because that is what is showing for all witnesses. The import must have carried over the default description used in RM9, not my custom ones.

My reason for wondering about custom witness roles was that they tend to be used in the stock report sentences in RM9. From the few test reports I've run in FH7, I didn't see it use "Witness" anywhere. So; I'm really not sure what effect its presence could have. I'll have to read the article Mike noted, because it currently appears that FH7 imports RM9 census events as one event for each household member and uses associated witnesses as well. It's still a bit confusing. Need to see what the article says about the subject.

[tatewise]

Gary, I believe you are misinterpreting several features and concepts.

RM9 Shared Facts are conceptually the same as FH Fact Witnesses and use the same _SHAR and ROLE tags when applied to GEDCOM. They are indicated by the 3-blue-ball icon in the More column of the Facts tab. It is perhaps unfortunate that FH calls them Fact Witnesses when most other products use Shared Facts, as it just leads to confusion.

My import of the RM Sample project maps Shared Facts onto Fact Witnesses for Census events and does NOT create a Census event for each household member. I suspect what you are seeing are the Fact Witness entries indicated by a blue arrow on the left in the Facts tab. They are NOT local Census events which is confirmed on the All tab by there being no corresponding Census item. Local Individual facts will have a blue bullet on the left. Family facts have a red bullet and also do not appear in the All tab. That is all explained in the Help for the Property Box dialogue.

The Role listed for each Fact Witness is whatever was imported from RM.

Such Roles are not necessarily defined within FH. If you check Tools > Fact Types... for the Census event then it only has Roles for Head of Household and Resident. The sentence you are seeing for the Witness Role is the default that FH uses when Tools > Fact Types... has no Role definition.

[Gary_G]

Mike;

***I think we posted at the same time, so please read the following with that in mind. I'm digesting your recent post now. ***

The article you referenced seems to apply to RM9 imports as well as the already noted TMG and Legacy. What is described in "Census Witness Roles" is largely what I am seeing in the imported data.

I was hoping that the article would be a bit more direct in its recommendations regarding the use of witnesses. If it were me, I'd likely summarize after the discussion to leave the reader with a clear vision of the "standard" approach to using FH7. Sometimes I find presenting all the options just clouds what appears to be "best practises". However; the essence of the article appears to be that one should adopt splitting and then it is unnecessary to specify the witness roles (even though one can). I'm fine with that approach, since it is clear that just using the imported RM9 file is not going to be the optimum route for me in FH7. Before I redo things, though, did I correctly understand the essence of the article?

[NickWalker]

There are thousands of users of FH and probably many different methods used so there isn't one way of doing it. However, Ancestral Sources doesn't support the use of witnesses for census entries as I wouldn't recommend that way of working and I've never heard a convincing argument for doing so.

[Gary_G]

Mike;

Thanks for the clear explanation. It helped.

I can see that the "Witness" role and "{individual} appeared in the household of {principal} in the census {date} <at {address}> {place}." sentence was transferred from RM9. I note that the role is never actually included as a tag in the FH7 sentences. However; it is clear from the text of the sentence when the person is the head of the house or just a resident.

[Gary_G]

Nick;

Thank you for chiming in. It's good to know what is done in AS and that one can get by without the additional work of tagging roles. I assume that FH7 is able to formulate reports that make sense without one explicitly entering the roles.

[NickWalker]

Using AS, each of the individuals in a census household in my FH project have a census fact (occupation too, etc.) with a citation back to a source that has a transcription of the census entry and a link to the census image. So if I need to know their role in a census entry I can look at either of those things. I don't see their role as being particularly significant enough to record that in a fact but if I did and e.g. wanted to have the role recorded in the census fact's local note then AS could accommodate that with a change to the census templates.

[tatewise]

I note that the role is never actually included as a tag in the FH7 sentences.

On the contrary, the sentence code {role name} can and does appear in FH7 sentences.

I assume that FH7 is able to formulate reports that make sense without one explicitly entering the roles.

That is a somewhat vague assumption.
Yes, FH7 does formulate Reports that make sense without explicitly entering Fact Witness Roles.
However, the Reports won't identify the other members of the household in the fact descriptions or sentences.
Those relationships will appear in the Sources section of Reports but only if the Text From Source option is enabled.

To get household relationships to appear in Narrative Reports then you must use Fact Witnesses with Roles and associated Sentence templates.

[Gary_G]

Gents; I'm hearing two apparently different views on the subject of witnesses from two very experienced users.

On the one hand, AS doesn't typically include the roles and Nick explained why. On the other, based on Mikes response, it sounds as if some people do use them. I only wish that I had a local user-group that I could have a conversation about the pro and cons of each approach.

Yes, FH7 does formulate Reports that make sense without explicitly entering Fact Witness Roles.
However, the Reports won't identify the other members of the household in the fact descriptions or sentences.
Those relationships will appear in the Sources section of Reports but only if the Text From Source option is enabled.

As an observation, sparked by the above (and this is not any criticism of a quite valid point)...

[Before I'm beaten up for openly expressing myself; keep in mind that I was explicitly asked, at one point, to record my "new user" impressions of the available documentation. So; consider this an instalment :>)]

When I see caveats like the above; as a new user, it nicely highlights why it is so difficult to get started with FH7. No new user can wade through the whole knowledge-base and figure out how FH7 "hangs together", or even really knows what questions to ask. On their own, they WILL miss important details and WILL make serious errors in approach to or configuration of FH7. There just doesn't seem to be any single, easy to find, written guide (possibly one for a splitter and one for a lumper) that gets an average user up-and-running in a reasonable direction... one that's unlikely to get them into difficulties from which they can't later recover. The work of authoring a guide would likely substantially reduce the need, on the part of experienced users, to explain the nuances of FH7. At this point in time, like many newbies, I'm just keen just to see myself making measurable progress towards my goal of finishing porting and continuing to enter data.

[ColeValleyGirl]

Gary,

Part of the 'problem' is that there are many valid ways to achieve the same end-result using FH; which is 'best' depends upon the priorities of the person asking the question, and how much they're willing to configure (and in some case, 'torture' the configuration of) FH, and whether they want to retain their existing workflow or are happy to change it.

There is a Making the Most of FH6 paperback, which you can still buy via the Family Historian Store, which might help, although it doesn't (obviously) cover the new features introduced in V7 including Source Templates. There isn't an electronic version available, and we have no idea if or when there will be a V7 version (although we hope there will be).

Personally, I don't use Witnesses as they're not included in the Individual Summary Reports [ISR] that I use to generate my website. I don't use Narrative reports at all which are the only standard reports that they include Witness Facts. (No doubt there's some convoluted scheme (mis)using apparently redundant fields and writing complex expressions that could shoehorn the information into an ISR, but I'm not interested -- the less I have to customise to achieve what I want, the better.

[NickWalker]

On the one hand, AS doesn't typically include the roles and Nick explained why. On the other, based on Mikes response, it sounds as if some people do use them.

You make it sound as if Mike and I are saying different things. I only wrote two sentences, and the first of those says that there are many different methods used and there isn't only one way of doing things, so of course some people do things the way Mike says. But AS doesn't support this and I can't see any reason for doing so. I think it would be confusing to have someone appear twice for the same census as witness and as principal.

There just doesn't seem to be any single, easy to find, written guide (possibly one for a splitter and one for a lumper) that gets an average user up-and-running in a reasonable direction...

I designed AS to help me to enter census records (initially) and I would have found it extremely slow and tedious to enter the data directly into FH. So I would argue that I have developed "a single, easy to find" system for adding census entries and I've produced at least 2 relevant videos showing how to use it for that purpose (one for beginners using Generic Sources and the other showing the use of Templated Sources). I wouldn't recommend using method 2 (even though it is supported) so I've not made a video for that. So there's certainly no ambiguity in terms of what I recommend. I think there will be many hundreds of users who would feel that AS has helped them "to get up-and-running in a reasonable direction". Once they become familiar with how AS generates their census records and perhaps use AS for the other supported source types, they can then take those lessons learned and use them in FH for the sources that AS doesn't support.

The issues you are highlighting are also the strengths of Family Historian. You would like to be told the correct way to do things, but FH doesn't restrict you to doing things in one way and the users of FH all do things in slightly different ways so you're never going to find a single guide that tells you the correct path to take.

[tatewise]

Personally, I don't use Witnesses as they're not included in the Individual Summary Reports [ISR] that I use to generate my website.

I need to clarify what I think Helen is saying.
She is correct that if a Principal fact has Fact Witnesses then they are only included in Narrative Report sentences.
i.e. If Principal A's fact has witnesses B & C then in a Narrative Report the sentence can say the fact for A had witnesses B & C.
However, all Reports include the Fact Witnesses, so the Individual Summary Report for B includes their Role as Fact Witness in A's fact. Similarly, the Report for C includes their Role as Fact Witness.

As others have said, FH offers a variety of methods that are all perfectly valid with different pros & cons.
I was simply highlighting that Reports present information in various ways and you must experiment with say the Family Historian Sample Project to see what works for you.

[ColeValleyGirl]

Is it worth including that clarification in the KB article you wrote about Standard Report Types, Mike?

[Gary_G]

Nick;

You would like to be told the correct way to do things, ...

Now; I didn't actually say "correct", did I? I purposely stayed away from that word suspecting what the response would likely be. Of course there is no single "correct" way, but there is definitely at least one way that will keep most people out of "trouble" until they develop their own style based on experience with the program.

With FH7; I almost feel like I've been presented with a bucket of miscellaneous parts and been told what you need is in there somewhere. So; I think I'll just start using it with my old RootsMagic workflow. Yes; it will definitely get me into trouble in FH7, but that's OK as there is no "correct" way...

[Gary_G]

Thanks for the added explanation Mike.
The tendency is for me to think in terms of how my old program did things and that is not always the right thing to do.
I think part of my current issue is that the Import process has brought over and highlighted a lot of "idiosyncrasies" of RM9.
That makes it really difficult to figure out what's going on and to find a way to fix it in FH7.
I'm really considering starting from scratch. Then, if I have and issue, I know it isn't one caused by "legacy data".

[Jean001]

Gary,

I think you are asking for something that cannot exist.

There is not just one, or a few, ways of using FH. That is why I moved to, and continue to use, FH. I use it to aid how I want to store, display, analyse, report and export my genealogical/family history information, rather than my having to follow a standard pattern laid down by the software.

Over time I have made changes to how I handle the information that I gather from my researches, that has happened since Day 1 (then using my own paper-based system), long before I came to use FH. It has never caused me difficulties, other than, perhaps, occasionally some extra time to make some changes to earlier records.

With FH I started off pretty much using the default settings (although I had a lot of customisations in TMG, my previous program) and, as I increased my knowledge of the capabilities of the program, I've developed all sorts of customisations that work for me.

A recent change for me is using 'Witnesses'. I used these in TMG but I transferred to FH before it had that capability. (Mike Tate's 'Give Witnesses Their Own Facts' Plugin meant that no facts were 'lost' when I moved to FH.) When FH did introduce Witnesses I was rather reluctant to start again.

With regard to Census Facts, in certain circumstances, I do use Witness Roles. I spend most of my FH time using the Property Box Individual Facts Tab and I like to see the information there recorded in a particular way so I have customised the Census Fact to accommodate my wishes.

(In TMG I used the default, and created my own, Source Templates. Again, they weren't available in FH when I transferred over. As things stand today I am using what are now termed in FH 'Generic Sources' and intend to continue to do so but, perhaps, one day I might have a change of opinion on the matter.)

In some areas of FH, using's Helen's excellent description, I probably 'torture' the configuration of FH. Some parts I use rarely.

[LornaCraig]

[... I was explicitly asked, at one point, to record my "new user" impressions of the available documentation. So; consider this an instalment :>)]

Gary, I think the problems encountered by new users depend very much on whether they are new to any genealogy software or new to FH having come from other genealogy software. Many of your questions have arisen because you are in the second camp.

Someone, like you, who is moving from different software will understandably want to know whether some of the things they could do before can be done in FH, and if so how. Sometimes the goal can be achieved but it requires a lot of customisation in FH. In these cases the user needs to decide whether to go down the customisation route or adopt an approach which is more straightforward in FH but may mean re-organising or even re-entering data.

A new user who has never used genealogy software before is more likely to start by using FH ‘out of the box’ to see what it does with its default settings. They are less likely to have a lot of detailed pre-existing requirements and don’t have any preconceived ideas about what it might be able to do if they tinker with it. Their questions are likely to have more straightforward answers for which the existing documentation was designed. It would be useful to hear feedback from ‘brand new’ users for comparison.

[Gary_G]

Jean;

I really appreciate you sharing your experiences in starting to use FH. Before I started using RootsMagic, I also used TMG prior its demise.

It's good to hear that the "out-of-the-box" settings didn't appear to have resulted in massive rework for you.
That makes me feel MUCH better. I was sincerely concerned that not setting the correct "switches" would make later changes impossible. Consequently, I think making sure I have all the needed fields in any templates is one of the more critical things at the moment. I can see that not getting that right from the start would likely mean substantial rework later on.

I have created my own source templates. My ones from RootsMagic were "Lumper-style", because it tends to encourage that. While FH7 can accommodate both styles, I saw that I needed to adopt a Splitter style in order to more closely follow my preferred Evidence Explained citation style. To keep the repetitive data entry to a minimum, I also heavily leverage storing invariant data in the relevant repository. For me; setting that up to work consistently has been a major effort at this point in the learning curve.

My current situation is that I'm seeing a lot of issues due to importing data from RM9 and the way it represented information. So; I "think" it may be safest if I kept the tree structure and re-entered all my facts "a la FH7" using my custom EE-Style templates. I'm not really sure that there is an another way to ensure that old RM9 data doesn't impact the way FH7 behaves and sends me "chasing ghosts".

Like you; I think I feel more comfortable entering data from the properties box, facts tab. Part of this is that I'm actually always mentally doing analysis on the data as I enter it and like to have a pseudo-timeline in front of me as I do so. Because I do a lot of French research, I also do a lot of transcriptions and translations prior to entering fact data. I don't currently do a lot of extractions into forms, so auto-text has a limited value in my workflow. Again; your approach seems to be a bit more intuitive (for me). Thank you for sharing your experiences. It helps.

[Gary_G]

Lorna;

You must read minds.

Yes; I am trying to see just how much I have to change to leverage the new (to me) features of FH7, while not totally destroying a workflow that has become comfortable for me. As noted to Jean, I am willing to change and even re-enter my data. My only concern is that, in doing so, I'm not making issues for my self and incurring the risk of even more rework.

[tatewise]

Is it worth including that clarification in the KB article you wrote about Standard Report Types, Mike?

Article updated.

[RS3100]

I would echo Helen's suggestion re the Getting the most from FH6 paperback. I started using FH with version 6, but at the time only the similar FH5 guide book was available. The differences between the book and the software were fairly straightforward to negotiate for me, as a fairly competent PC user, but by no means conversant with programming or similar intricacies.

Nevertheless, I spent between 2 - 3 months working my way through the tutorials in the book, using the Sample Project, before doing a single thing with my own tree or creating my personal project. That taught me a great deal, including the various possible ways of doing things, and enabled me to work out how I wanted to proceed and the benefits of disadvantages to me of various options and methods. I would highly recommend it, as it will almost certainly give you a good all round feel for FH, from which you can target any further queries as necessary.

[Gary_G]

"RS3100";

I always meant to ask if anyone had heard whether CP intended to publish an updated book for FH7.
The FH6 one may cover some aspects, but I understood that FH7 made some important changes.