Post by Uncle Buddy on Dec 17, 2019 19:03:11 GMT -8
Most genieware is conclusion-based. The user enters people, places, events and attributes as if they were facts. Some of the genieware even uses this word "facts", probably because it's an easy word to use.
There are also "evidence-based" genieware products, wherein the user enters a source and the information on the source, and somewhere down the road I guess this is supposed to lead to a conclusion. I haven't put much effort into using any of the evidence-based programs because I don't think users want to be told how or when to enter their data, and moreso because the interfaces I tried to navigate turned me off right away, which has nothing to do with whether they're evidence- or conclusion-based. So the actual capabilities of the evidence-based geniewares might have escaped me. I've written reviews of many of the software programs but I'm afraid to post them because someone might think I'm a negative person. Actually I'm a-goin' blind kinda person so if I can't change background colors to dark and font colors to light, and if I can't increase font sizes, I just stick to the two programs I've actually bought and paid for--which can't do all that stuff either--and spend more time writing my own software than trying to input the huge piles of data that I've accumulated on many families.
In order to create better genieware, the basic premises have to be examined. The first fact that has to be taken into account is that in genealogy there are few absolutely provable facts. The more obscure the person/place/event, then the fewer facts there will be. For the most part we have to form conclusions based on relatively scant evidence, but for absolute proof there's rarely going to be any unless you knew the person yourself. So I've broken "facts" down into more accurate terms like entities, assertions, and conclusions.
Entities in genealogy (according to my definition) are things that exist whether genealogy knows about them or not, such as persons, places, and sources.
An assertion is something that can be claimed or theorized about an entity.
A conclusion is something that can be said to be definitely true about an entity.
In Treebard at this point, my plan is to separate facts into assertions and conclusions and keep them separate. For casual genealogists, this means they can still make unsourced trees by entering everything as a conclusion. Conclusions are displayed on the front page and the assertions functionality can be ignored.
More evidence-oriented genealogists like to compile assertions and gradually watch them build into conclusion-worthy data. Doing this in existing software is klunky because there's no structure to hold it. So in Treebard we have assertions, and we have conclusions. Each assertion is linked to exactly one source, but one source can be linked to many assertions. If a family of ten lives together in a certain place and time according to a source, then you can extract ten residence assertions, possibly several occupation assertions, some relationship assertions, etc. In another source about the same people, there will be some overlap, some unique data, and some contradiction, so separate assertions based on that source can be created separately and never tainted with information from any other source.
On the front page, which shows the relationships, events, and attributes for the current person, there will be an events table which shows conclusions. For example there will be a row for the current person's marriage event. At the end of the row will be a button with a number on it to indicate how many sources/assertions were combined into that single conclusion. The user can click a source button to be shown the exact assertions--information entered directly from a source--which became that conclusion. For a user who enters conclusions directly, the number on the button will be 0.
In this way I hope to design the first software that is equally usable by all kinds of users from experienced, skeptical genealogists who have to have at least one or two sources to form a conclusion, to first-day novices who tend to jump to conclusions on first sight of some data, or even copy other peoples' trees.
I'm trying to create genieware that works for everybody. It also has to work for the nearly blind like me because there's a lot of us senior citizens doing genealogy and using computers doesn't get any easier as we get older. I realize that Windows has so-called access features but I don't like the way they work. I've tried changing my Windows to a dark theme and the result was ugly and sometimes unusable. I believe the software designer should build these features into his own software so that the user doesn't have to depend on what Windows has to offer.
In a way it would be better to make a web app instead of a standalone app but I have no interest in tackling the problem of connecting an app to the internet. This is too technical for my tastes and involves all kinds of security issues that don't exist with a standalone app. Treebard is here to show the way to developers who are sometimes in too big a hurry to get a product to market, but Treebard is not here to be all things to all people.
There are also "evidence-based" genieware products, wherein the user enters a source and the information on the source, and somewhere down the road I guess this is supposed to lead to a conclusion. I haven't put much effort into using any of the evidence-based programs because I don't think users want to be told how or when to enter their data, and moreso because the interfaces I tried to navigate turned me off right away, which has nothing to do with whether they're evidence- or conclusion-based. So the actual capabilities of the evidence-based geniewares might have escaped me. I've written reviews of many of the software programs but I'm afraid to post them because someone might think I'm a negative person. Actually I'm a-goin' blind kinda person so if I can't change background colors to dark and font colors to light, and if I can't increase font sizes, I just stick to the two programs I've actually bought and paid for--which can't do all that stuff either--and spend more time writing my own software than trying to input the huge piles of data that I've accumulated on many families.
In order to create better genieware, the basic premises have to be examined. The first fact that has to be taken into account is that in genealogy there are few absolutely provable facts. The more obscure the person/place/event, then the fewer facts there will be. For the most part we have to form conclusions based on relatively scant evidence, but for absolute proof there's rarely going to be any unless you knew the person yourself. So I've broken "facts" down into more accurate terms like entities, assertions, and conclusions.
Entities in genealogy (according to my definition) are things that exist whether genealogy knows about them or not, such as persons, places, and sources.
An assertion is something that can be claimed or theorized about an entity.
A conclusion is something that can be said to be definitely true about an entity.
In Treebard at this point, my plan is to separate facts into assertions and conclusions and keep them separate. For casual genealogists, this means they can still make unsourced trees by entering everything as a conclusion. Conclusions are displayed on the front page and the assertions functionality can be ignored.
More evidence-oriented genealogists like to compile assertions and gradually watch them build into conclusion-worthy data. Doing this in existing software is klunky because there's no structure to hold it. So in Treebard we have assertions, and we have conclusions. Each assertion is linked to exactly one source, but one source can be linked to many assertions. If a family of ten lives together in a certain place and time according to a source, then you can extract ten residence assertions, possibly several occupation assertions, some relationship assertions, etc. In another source about the same people, there will be some overlap, some unique data, and some contradiction, so separate assertions based on that source can be created separately and never tainted with information from any other source.
On the front page, which shows the relationships, events, and attributes for the current person, there will be an events table which shows conclusions. For example there will be a row for the current person's marriage event. At the end of the row will be a button with a number on it to indicate how many sources/assertions were combined into that single conclusion. The user can click a source button to be shown the exact assertions--information entered directly from a source--which became that conclusion. For a user who enters conclusions directly, the number on the button will be 0.
In this way I hope to design the first software that is equally usable by all kinds of users from experienced, skeptical genealogists who have to have at least one or two sources to form a conclusion, to first-day novices who tend to jump to conclusions on first sight of some data, or even copy other peoples' trees.
I'm trying to create genieware that works for everybody. It also has to work for the nearly blind like me because there's a lot of us senior citizens doing genealogy and using computers doesn't get any easier as we get older. I realize that Windows has so-called access features but I don't like the way they work. I've tried changing my Windows to a dark theme and the result was ugly and sometimes unusable. I believe the software designer should build these features into his own software so that the user doesn't have to depend on what Windows has to offer.
In a way it would be better to make a web app instead of a standalone app but I have no interest in tackling the problem of connecting an app to the internet. This is too technical for my tastes and involves all kinds of security issues that don't exist with a standalone app. Treebard is here to show the way to developers who are sometimes in too big a hurry to get a product to market, but Treebard is not here to be all things to all people.