Is there consensus about the use of the forms plugin?
Since I do have a number sources that would easily fit into the use of a form I thought to try that. And yes it makes data entry easier.
BUT
since all data are stored under the attributes of the person/reference (and not as data under the source citation) it is difficult to use the entry as a source elsewhere.
Win11
Gramps AIO64-5.2.2-r1-f905d14
Form Gramplet 2.0.40
Best answer is “opinion is divided”
I use it a lot but that is just the way I work but it does not suit all.
So work out what is more important to you and sort out any work rounds
you have to come up with before you get too deep in
phil
And that’s one of the reasons why I don’t use it, and don’t plan to either, unless the data is stored in a place that’s easier to access, and in a way that does not depend on the actual form. Sites like Ancestry and FamilySearch seem to have a universal format for the person data that they extract from sources, and I’d love to see that in Gramps too.
Another reason is that, when indexed data are available on the web, on the big international sites, and on local ones, it feels like an enormous waste of time to type those into a form, as new data, where one could also copy the data from the site to a note, if we’d had support for formatted (html) text in notes, where formatting includes simple tables. GEDCOM 7 allows mime/html for notes, but minimum demands for that do not include tables.
I echo what Enno said. Because they use attributes the data is not included in a gedcom export. So that is a hard no for me. Also, I have no interest or reason to capture all the info in a Census record. I do copy into a citation note birth, death, Baptism, Burial and some other records that I believe are important. These notes are included in a gedcom export, so I’m happy. I see no reason to enter data using a form when I can just copy/paste the info into Gramps.
There are a couple of reasons why people like to transcribe all the data into a structured format:
Although some information may not seem relevant immediately, it may be useful in the future and form part of a bigger picture.
When the data is structured it is possible to list a given attribute over time. For example, I like to see how the place of birth varies over time in census records. Sometimes people list a nearby town instead of the village that they were born in. I’ve also seen an incorrect spelling recorded that gives a clue to the pronunciation of a place name.
Another benefit of a structured format is that it can easily be transformed using a script. If you wanted to export the information into a Gedcom DATA tag, it would be easy to write a script to do so. We also maintain the flexibility to migrate the data to a different format at some point in the future.
The next version of Forms the @RenTheRoot is in the process of writing will provide even more features. These will include a highly configurable mechanism for creating conclusion objects from the transcribed data.
This new method of source based data entry will always be optional. Users will always be free to create conclusion objects directly.
Thank you all,
I have decided tot refrain from using forms.
Also am already in the process writing a helper program (something I already use for entering certificates)
Entering data in the program, following the layout from the source I have found. Then click and paste the info in the way I want under event>source citation.
Yet another days work, but the result will be precise what I expect and need. Everytime.
The very first census addon started off in exactly the same way. It was just a helper script that automated my data entry process. Then I decided to share it, so that other people could use it if they wanted.
Nick, yes, it might be useful in the future, however I currently have the potential to add 10,000 persons to our family tree and my time is limited, so I need to add the census events as quickly as possible. If I could click a button and it automatically copy the info in, I might reconsider my workflow. It is so boring adding these events.
I’m only using them to track where they lived at the time, which can be shown on the map. I do check to see if the family members are in sync too, but I don’t record that info. I keep the ULR, so I can access the record again if necessary. If the occupation is shown, I will enter that. The Census data fields change from year to year and are different per the country which makes things even harder. Why there is a different form for each country/year. Analyzing any Census data would be problematic.
Hey! I am Renee, I am working on an updated version of the form addon which I am hopeful will be completed before the end of summer- I’m hoping to release a beta version within the next 1.5 months or so.
I’m just going to go ahead and @ all the people in this discussion who seem to have specific features that if included in forms would help them out. I have a post where I periodically update with how the new version is coming along and ask for feedback. I would love if you guys could head over there and drop a list of features you would like to see included. It’s a long thread so don’t worry about having to read through it all- If you accidentally include a feature that has already been suggested its not a big deal. I read every reply and consider all suggestions. I have gotten some very helpful feedback so far and I think you all probably have even more insights that could be helpful.
Here is a link to the thread:
I think that a separate thread is reasonable. Moving the goallines while code has begun development is a bad practice. Refining based on feedback loops is necessary though.
Hi Renee, thanks for mentioning us, and that thread. I tried to follow it, but I’m not much interested in forms, for a couple of reasons. One is that we didn’t have many censuses here, in The Netherlands, and another is, that I have no interest in typing data into forms, because most of the data for my country is already there in indexes on local sites, and also on FamilySearch, so typing that from screen to forms looks like a pure waste of time.
Another thing is that the way that data is stored now is quite obscure, and differs for every form, so it’s not as universal as the data that I can find on those sites, where all persons have the same elements, regardless of the source they were extracted from. Not all data fields are filled of course, but for me as a software engineer, there is a kind of unity that I can’t find in the forms data structures as they are used in Gramps.
The idea that it’s quite silly to type things seen in one window into another is also the reason why I never bothered about so-called evidence based programs like Clooz or Evidentia, or the Dutch made Centurial, which is a real disaster, TBH.
I also inherited a lot of Word documents from my father, from which I’d really like to extract names in an automated manner.
It would be way more appealing, if we can find a way to extract data from notes, maybe with semantic mark-up, as they are already avaiable in Evernote, Gramps, Joplin, Word, Zotero, and whatever other tool there is that creates data in a known format.
That’s true I hadn’t quite considered that. I figure if some suggestions are just too far away from what I am trying to do I can keep a level of discretion to filter those out.
A few years ago I wrote an editor that tagged information in notes against people. You could add people to a list in a sidebar, then by selecting a person and selecting some text, a piece of information could be recorded against the person. The information could be given a type, and selecting a previously tagged item would highlight the text in the note.
This editor was intended for source based data entry from unstructured documents. The process was manual though.
I see, this makes a lot of sense! So many countries don’t have robust indexes for information. I personally am US based and my genealogical background encompasses most countries in Europe with my origins not being more than 20% max for any individual country. I’ve found that most of the records I am looking at are unindexed, I have heard good things about the digitization and indexing of Dutch records though.
I agree, but it is still interesting to read the comments. They may give you ideas for future development.
It would probably be best to focus on enhancements that benefit existing users first.
That would be an advantage, if it were structured, but what we have now does not look like a structure at all, for a couple of reasons, like being stored in attributes, being user defined, with column names that may be meaningless to users speaking other languages, and an absolute lack of uniformity. To me, it looks as chaotic as the hundreds of different templates that one can find in Evidence Explained, and the way they were implemented in programs like RootsMagic.
All sites that I know, and those are the big commercial ones, FamilySearch, and all national and regional archives that I’ve visited over time seem to use universal data structures, regardless of the type of source, so that they can use universal screens for searching and connecting/copying source data to user trees, where applicable. Such designs make software much easier to use, write, and maintain, as you can see when you use any of those sites.
And to paraphrase a well known TV program: We have the technology to work with texts, like it already exists for the semantic web. Such tags can easily be included in html/xml based texts, stored in notes in Gramps, and already exist on a couple of sites. You may also add these yourself with the right software, that works in a similar way as a word processor in which you select text and apply a style. Marking a piece as a person’s name, or part of it, a date, or place can be done in the same way, and it shouldn’t be too difficult to add mark-up that defines a higher order structure like a row in a table as a person with a specific role, while leaving the user’s preferred lay-out intact. It’s all been done elsewhere already, and I’ve seen parts of these tricks work in an old version of Family Tree Maker.
I just checked the source of a page on a well known Dutch archive portal, made by an entrepreneur that I know, and the semantic data are all there.
Right, that’s what I saw in that old version of Family Tree Maker too.
Whilst respecting everyone’s version of what they do that is the beauty and the beast of GRAMPS
“The idea that it’s quite silly to type things seen in one window into another is also the reason why I never bothered about so-called evidence based programs like Clooz or Evidentia, or the Dutch made Centurial, which is a real disaster, TBH.”
My fundamental reason for doing this is the sheer number of transcription errors that occur on all the commercial and free sites, means that I cannot possibly just copy and paste or use the transcription I always refer to the original image.
So rather than mess with transcriptions and index’s (also prone to transcription error), I would rather someone looked at integrating an AI that can interpret the originals from images into a series of standard formats which then might only need minor manual correction. Now that would be worth changing my work pattern for.
phil
I would be interested in reading which they are, because I know that indexed data does exist for other western European countries, like England, France, and Germany, in databases held by national and regional archives, and by sites like Ancestry, FamilySearch, Find-My-Past, Geneanet, and My Heritage. I also know that a lot of indexed data are available for the US on Ancestry and FamilySearch, and on Ancestry, I can even find quite recent indexed records for relatives in the US. Some censuses are free, and I often add data from those to my tree, where citations are automatically added too.
Ancestry is not cheap, and Find-My-Past isn’t either, but AFAIK, they cover all censuses known in the UK and North America.
My college major is Machine Learning and Artificial Intelligence, so rest assured I have some more “out there” ideas for how that can be used in addons. Stay tuned for next year’s summer for more on those haha. And feel free to send me messages with your most interesting ideas. I keep a local record of the suggestions people send me for inspiration on future development projects. I’m only 24 so I have many years of contributions I can plan and look forward to!