Evidence rating and star coloring

Site Feedback Feedback: soliciting cmts
1

Dallan,

Knowing you allow master sources to be categorized it occurred to me it might be useful to categorize facts in evidences as being primary or secondary as well. Directness of the fact might be useful too, primarily when inferring events based on ages.

The thinking here being you could color code the stars in some way to indicate the quality or strength of the supporting evidence. A direct, primary fact in an original source is a green star. An indirect, primary fact in an original source is an orange star. A direct, secondary fact in an original source is a yellow star. Or maybe original sources items are bright colored and derivative dull colored with the other attributes determining the color.

I didn’t check the roadmap so maybe this has been suggested before or is already planned.

Thanks,
Chris

P.S. Update to this thought, you’d want an option to disable it for color blind people as well I suppose.

2

What about solid gold stars for primary evidence facts (the fact listed on the first evidence tab), and outlined stars for facts listed under the people on the fourth evidence tab?

3

Thinking more about this. You have the star next to the source/evidence item for each fact and then the star for each source in the profile header.

The problem as I see it is that you can not make a blanket assumption about the quality of each piece of information in the specific source record like that.

Consider a death certificate for instance, which is an original and not derived source. Some facts on the certificate are primary information such as the person’s name, date and place of death, name of informant, cause of death. Many certificates may also have burial location and date or cremation location and date and those too would be primary facts. Then there are the other secondary facts such as birth date and location and names of the parents. Then consider the case where there is not a birth date itself, which might be considered direct evidence, but an age which would imply a birth year and so be considered indirect evidence.

So my thinking would be to score each fact in the source record with a rubric of sorts:

type of source:      original=1   other=0
type of fact:        primary=1    secondary=0
directness of fact:  direct=1     indirect=0

For directness, the best example of indirect I can think of is deriving a year of birth, marriage, or death based on age. I am sure there are other examples but none come to mind at the moment, maybe someone else on the forum can think of some. So maybe you handle that one algorithmically and if deriving a fact from an age then you set it to indirect otherwise default to direct.

You let people identify the type of source today, there you would default to other unless they specifically selected original. So you would just need to let people optionally identify if the fact is primary or secondary. For that you could provide a check box and if checked it is primary otherwise the unchecked default is secondary.

So you score and then sum the values to get the index to the entry in a color table to use. You could then maybe provide a couple user selectable tables so some scheme can be derived that works for color blind people.

So maybe use something like this:

0	yellow-orange		brown
1	yellow				bronze
2	yellow-green		silver
3	green				gold

Notice if we default to 0, 0, 1 then we end up with yellow which is what you use today.

That color would then used next to the source/evidence entry for the given fact. Then for the profile header you could base the star color on the highest rated fact found in the source record.

Not sure if I explained that well, but that’s sort of where my head went with the idea if it makes any sense.

Thanks,
Chris

P.S. Also consider that some of this is information you might want to include in an evidence report in some manner.

4

This is the kind of complexity I want to avoid. We need a simpler solution. I’m open to suggestions.

5

Yeah, I understand that, the keep it simple principal is a good one.

You could scale down the concept and take a similar rubric approach using your suggestion in combination with whether the source record is tagged as an original or not. That would give you:

type of source:    original=1   other=0
location of fact:  main=1       additional=0

With main being what you termed the primary evidence fact and the additional being the others.

So your color map then might be:

0   yellow         bronze
1   yellow-green   silver
2   green          gold

And then for the stars in the profile header just color based on whether an original or derivative source.

Or of course go with your original suggestion. :slight_smile:

Thanks,
Chris

6

This seems better. Others’ thoughts?