Not all photos are equally valuable though and as you sort through the photos you might put them into three piles: the photos with people you recognise; the photos which have descriptions scrawled on the back; and the photos of unknown people and things. The photos of people you recognise are nice, and might provide a few new pieces of information (e.g. I didn’t know great uncle Mortimer rode a motorcycle!). The unlabeled photos are really not useful at all, unless you can take them to other relatives that might recognise the people and places depicted. The pile with the photos that also have descriptions are pure genealogy gold though! Below is a family photo of mine of a distant relative named Walter Wood. If the photo hadn’t been labeled with his name then nobody in my family could have identified him. The photo was much more useful than that though, since it confirmed a hypothesis I had about my family tree -  before I found this photo I wasn’t completely sure I was related to the Wood family!

W E L Wood

A written description on the back of a photograph would nowadays be called metadata. Indeed, any information about a photo apart from the image itself can be called metadata. Although the information can seem rather obvious and boring to the person who shot the photo, metadata can add considerable extra information and meaning.

In that spirit I have begun adding metadata to my own family history documents as well as all my digital camera photos too. There are three types of metadata that I have added to my photos:

• the date and time that the photo was taken
• the place the photo was taken
• who is in the photo? what are they doing?

I will talk about the third type of information first. It has become standard on modern computer systems that all image files support metadata and the easiest metadata to add to your photos are keywords. You can add keywords that indicate which people are in a photo and what they are doing. For example, the photo above might have the following keywords: Walter Wood; military; Australian Army; World War 1. I have added keywords to all my personal photos and most of my family history photos – and the payoff has been immediate!

The advent of digital cameras has meant that the average person takes hundreds of times more photos than they did 10 years ago. I myself take thousands of photos every year. In many ways it makes the photos less precious, and wading through them all in search of a particular photo can be tedious and time consuming. After I added keywords to my photos the process became instantaneous and it still puts a smile on my face every time I do it. If I want that photo of my cousin Jimmy at my grandmothers birthday party then I just search for photos with the keywords “Jimmy” and “birthday” and it brings up five photos. Amongst those five photos it’s easy to find the one I want.

An example: the tug-o-war team

I will demonstrate how I use metadata with an example. The picture below is of the tug-o-war team from the town of Merino, Victoria, Australia in 1932 (my great-grandfather was the anchorman).

1932 Merino tug-o-war team

The photo has keywords for all the team members whose names I know as well as “tug-o-war” and “Merino all stars” which was the name of their team. I also set the date of photo metadata value to 1932. The final piece of metadata I added was the location that the photo was taken. This is best added using the GPS coordinates. There are a number of programs that allow you to do this, but I won’t discuss that in this article. Once the coordinates have been added you should be able to see the location of the photo in google maps. The position can be accurately shown in google maps to within a few metres.

Photo location

So now my photos are encoded with lots of extra information about when, where and of what they were taken. It makes the photos more manageable for me to deal with and will make them much more useful to the people I pass them on to. Remember to have a good backup strategy though – I backup my computer at three different locations and have automated hourly backups at home. I have had hard-disk failures before and I don’t want all the work I have done adding metadata and scanning documents to be destroyed in an instant – so backup your data people!

I am somebody who is interested in both mathematics and genealogy and I decided to play with some more interesting ways to measure a family tree. In particular, I wanted to be able to calculate the genealogical average of any quantity. Some examples of interesting numbers to calculate are the completeness of a family tree or the average date your ancestors came to your country (if your family is not native to its area, like most people from Australia or the USA).

One important property of a family tree is its completeness. The two easiest ways to define the completeness of your family tree are:

• count the number of people or direct ancestors it contains
• count how many generations it goes back

These methods are fairly primitive though, so lets consider some better ones.

One thing I don’t like about these measures is that they become infinite as a tree approaches completeness. I would prefer a measure that approached 1 as a tree was completed so that you could define the completeness as a percentage. If we want that to happen then we’ll have to weight each generation so that the measure converges. The more mathematical details of all this are explained here.

I implemented my completeness measure in some computer code so that I was able to calculate these things for my own family tree¹. This measure defines my family tree to be 96.3% complete, while my parents family trees are 91.4% and 93.8% complete.

This type of maths can be applied to define quantities that are more interesting than mere completeness. We can define the genealogical average of almost anything. One thing that was of interest to me, as someone born in an immigrant society like Australia, was to define the average year my ancestors immigrated to Australia.

In order to calculate the average year of immigration, the average will only be calculated over immigrants. The weighting factor will be , corresponding to the ammount of blood shared with the ancestor. In my own family tree, I calculated the weighted average year of arrival in Australia is 1854.6. My parents averages are 1856.7 and 1852.5 and obviously mine is bang the middle of these two numbers. Once again, the details of these calculations can be found here.

Although I enjoyed programming all this myself it would be nice if something similar to a genealogical average were implemented in a common family tree program so that I could compare my numbers to some others!

To help you decide if you will enjoy reading my blog, here is a list of topics I am interested in:

• The many ways of formats for storing family trees, such as GEDCOM and the myriad of proprietary options
• The GEDCOM 5.5 standard that forms the foundation of computer storage of family trees
• Open standards in genealogy
• The best ways to digitally archive your old photos and documents
• Metadata! Wouldn’t it be great if all your old photos had the date and location information on them? Do your descendants a favour and add that useful information to the metadata of your digital photos
• How maths and statistics can teach you about your ancestors

I expect to be posting every few weeks, so stay tuned and comment if you are interested!

– Jack