Thursday, August 17, 2017

What are you doing, AncestryDNA?!

As some of you may have already seen, AncestryDNA is currently contacting some of its users and asking for feedback. I've filled out many such questionnaires over the years. Every chance I get to tell them how much I want full segment data, I take it. Part of me is still holding out hope that one day they'll actually listen.

Part of me knows it's probably a waste of time.

This time, however, was a little different.  Instead of various general questions about my experience, the questions focused on the cousin match list. It included a sample explanation for a new user, outlining what the cousin match list is and how it works.

I admit, I was pleasantly surprised. I wondered if AncestryDNA was finally addressing how much they oversimplify DNA testing in their advertisements. Much of the overconfidence and confusion the public has regarding ethnicity estimates is directly related to AncestryDNA marketing. And trust me, something has been rotten in the state of Denmark with these ethnicity estimates for a long time.

Meanwhile, their continued product development depends on not only a steady supply of new samples, but trees and family information for the algorithms to analyze. They've done very little to educate or incentivize new customers to create and share their trees with other users. I'd estimate that most new customers to AncestryDNA don't even understand how important those trees are. I doubt they realize the increased functionality they receive when they provide that information. The result is that many people dead-end at their ethnicity estimates, and have no direction for what their next step should be after that. And it's something that long-time customers have complained about to AncestryDNA for many years, without much of a response. (Myself included.)

All of this was on my mind as I took this survey. Then I came to a question that surprised me. No, not just surprised. It left me feeling really concerned about what AncestryDNA might have in store for its current and future customers. 

Here's a screenshot of the question:

Thinking back to my experience as a brand new AncestryDNA user, I would have preferred to be formally introduced to the cousin match list. I would have liked to be educated on how to maximize all of the features available to me. But that wasn't an option I could select.

I mean, I'm just a paying customer. What do I know?

It's that last option that really has me worried. "Do NOT show me my list of possible relatives"? "Do NOT attempt to find my possible relatives"? Is this really where we're going now? Where AncestryDNA is going to measure how many people click this option, and justify some crazy change by saying "it's what people said they wanted"?

So when they asked me why I answered the way I did, I told them exactly what I thought of their question:

The value of DNA testing for genealogical purposes relies on both information and collaboration. By allowing new customers to decide whether to engage with cousin matches, you're placing that decision into the hands of your least experienced customers. You'd be taking that opportunity away from existing customers and more experienced researchers. And as an existing customer, I'm not okay with that possibility. 
If people are not prepared for the information they are being given, or to participate in this collaboration, it's because they don't understand what DNA testing is. We cannot continue to perpetuate the myth that DNA results somehow exist in a vacuum, that you can sequester people from connection. If people have that expectation, it's because they've fundamentally been misinformed, either through word of mouth or advertising, about the potential and purpose of DNA testing. If you want to change that expectation and user behavior, you need to change the messages that you send to people. Because from where I'm sitting, as a customer of AncestryDNA since 2014, that messaging has become increasingly misleading as the service has grown. And it shows in the interactions I'm having with my relatives/your customers. 
If you give people the option to walk away from collaboration, when they don't even understand the decision they're making, you're not helping your consumers to make informed decisions. You're delivering a product to them at full-market price, without giving them any of the tools or education they need to use it fully. It was unacceptable when I took my test in 2014, and it's unacceptable now that AncestryDNA has 5 million customers. And I've been in contact with enough of your customers to tell you that they don't understand these choices. They don't understand the features and potential they're sacrificing. And once I've made some of them aware, and they have the desire to engage more fully, they find it exceedingly difficult to navigate the site to change their settings. This expectation that a user can change their settings any time vastly overestimates the technical ability of the average new AncestryDNA customer. 
For the sake of my own satisfaction of as an AncestryDNA customer, I go out of my way to make these choices available to your customers. I educate them because you don't. But I also don't have the tools to meet their needs, and it's frustrating to me as an existing customer. It keeps me from wanting to buy additional tests from your company. If you focus on giving me, the experienced user, the collaboration tools I need, it will give everyone else a better experience. The very thought that you might be considering doing the opposite is discouraging and disheartening.

I brought this up in my Family Tree DNA review, and I'll say it here in relation to AncestryDNA as well. None of these testing companies give users the tools they need to effectively communicate and collaborate with each other. This is not a problem that will be overcome by shrugging shoulders and saying, "Well, most customers don't want it anyway, so why bother to improve? Why not keep it simple?"

Because when you don't provide the functionality that users require for the task they've set out to do, that is the definition of making something harder. Not simpler. It takes information and collaboration to make family history discoveries using DNA. It takes work. It takes segment data. It takes traditional research. It takes source citations and original documents. It takes everything that good genealogical research requires. No amount of marketing speak about "responding to customer feedback" can change the fundamental nature of DNA testing. And every time AncestryDNA runs a commercial, I can't help but wonder if they've made the mistake of believing their own version of events.

People are not stupid. They know when they've been had. Consumers figure it out when what they were promised is not what they receive. They figure it out the moment they ask themselves what to do with their results, and they come up completely empty. They see it as they comb through page after page of matches, none of whom have trees. They realize that it isn't as easy as spitting into a tube and having the universe unfold before your eyes. And given that many people are pushing back against this problematic narrative of "spit in a tube and instantly discover your origins," this pattern of smoke and mirrors is unsustainable. 

Something needs to change. And if AncestryDNA thinks the change will be to simply hide the cousin match list, they're already looking way beyond the mark.

Monday, August 14, 2017

Getting Started: Interviewing Relatives

Genealogy comes in a lot of flavors. For many of us, it's a personal journey of self-discovery. We do it in pajamas, on couches, without a whole lot of human interaction. In my heart of introvert hearts, nothing makes me happier than digging through an unindexed image collection on FamilySearch by myself. Exactly because I don't have to see or talk to anyone.

You know, the finer things in life!

But here's the thing about researching your family history: it's their story, too! In order to tell it, you're going to need their help. Maybe you have questions only your grandparents can answer. Or your great aunt Mable, who you've only met twice, is the one with all the good photos. Or maybe you're an adoptee, and the only biological relatives you've ever encountered are the names on a DNA match list. Knowing how to conduct family interviews is an important skill to have in these moments. But when that doesn't come naturally to you, it can be daunting and intimidating!

I've had my share of good and bad experiences with family interviews. And even when you try to do everything right, they don't always turn out how you want them to. So let's talk about family interviews: the good, the bad, and everything in between. If your interview goes well, you'll find some great tips here to capture everything you need to know! If it doesn't, at least you'll have a great story to share in the comments!

How Do I Prepare?

This may seem odd, but the secret to a good interview is you. Do you know how to listen, without being distracted by what you're going to say next? Are you comfortable with silence? If you offend someone or stumble into an uncomfortable subject, do you know how to respond? The most uncomfortable way to discover negative answers to these questions is in the middle of an interview. Especially with someone you don't know well. 

If all you want to know is grandma's favorite color and what size shoe she wears, this may seem like some overly ambitious preparation. But I remember being a teenage genealogist, and asking my grandmother what her father's name was. To me, that seemed like a pretty innocent question. But for my paternal grandmother, it was a difficult and painful question to answer. She had grown up in a foster home, and never knew her father. And it took ten years of circling back around to that same question for me to hear the story she had to tell.

We are not in charge of deciding if, when, or how someone tells their story. That is not the purpose of an interview. Ten years of trying and failing with my grandmother taught me this. All I could do is create the safest space I could for her, and accept whatever story she tells me. If I ask her about her childhood in Canada, and she wants to talk about all of her doctor's appointments since the last time I spoke to her, guess who's actually in charge of the conversation?

That's right. The one with the information. Not the one asking the questions. Preparing to do an interview, first and foremost, has to mean letting go of our expectations of what we expect to get out of it. Only when people sense that sincere willingness to listen will they truly open up about their experiences. A good interview relies upon that sincere, meaningful connection.

Who Should I Talk To?

What do you want to know, and which relative has that information? It sounds so straightforward when you say it like that. But depending on what you need to know, who the person is, and how well you know them, you may be in for a real challenge. And finding the right person to talk to may be the hardest part!

You may come across the names of people in interviews with other relatives, in obituaries, or online trees. Social media is also a fantastic way to find distant relatives, especially in conjunction with DNA results. But sometimes, you'll stumble into relatives completely by accident. Such was the case when I wrote this blog post many years ago about meeting my father's half sister.

Because of my initial encounter with her, I got many clues that continue to influence my research. By the time our first phone call ended, I had in my mind a very rosy image of how things would go. Phone calls several times a year, Christmas cards, an ongoing conversation about new discoveries I was making. I invested a lot of hope into our follow-up meeting, especially for additional questions I wanted to ask.

Originally, she and I talked about meeting together at her home. But within a few days, she rescheduled to meet me by herself at a restaurant. I spent some time making up some pedigree charts and family group sheets to share. By time I showed up at the appointed place and time, I couldn't wait to meet her, to show her everything I'd found.

Minutes ticked away. The time we agreed to meet came and went. I glanced at my phone, over and over. After forty minutes, I got a phone call.

She wasn't coming. No, she didn't want to reschedule.

That was it. The one time I spoke to her on the phone was the only chance I would ever have. Had I known that, I would have prepared much differently for my original phone call with her. And when it comes to finding relatives, that is my most important advice. Never assume you'll have more time, or another chance to connect with someone. Treat every conversation like it might be your last. Because sometimes, no matter how good your intentions, people don't give you the chance to know them. But if you make the most of what opportunities you do have, without expecting more to follow, you never really have to be disappointed.

What Should I Ask?

When most of us think of interviews, coming up with good questions is where our minds go first. But what makes a good question? How many is your relative willing to answer? What questions will help them to talk about what interests them? Do they even need you to ask them questions to get them talking? Consider your research. What are the most important gaps you need to fill? All of these factors will shape the questions you come up with.

You can also search around online for inspiration. This list, as well as this one and that one and here, have some great places to start. Use any of these questions, or come up with your own.

Many people become frustrated with interviews when they are asked too many questions they don't know the answer to. It begins to feel like an interrogation, and may distract from the stories they might otherwise tell. While our research problems may require targeted questions, it's important to balance how many questions we ask with space for the person to speak freely. And whenever the person is talking, always be listening. You never know when the clue you need later might be in something your relative is saying right now.

One of my cousins I met through Facebook loves comic books. It would be easy for me to think that if I'm talking about comics, I'm not talking about genealogy. But nothing could be further from the truth for their family. Many of the stories I was interested in hearing about his father were also connected to comic books. Favorite heroes, favorite memories of going to buy comics together, how other family members would also participate. Those stories about comics have given me a much truer sense of what his father was like, more than any other question I might have come up with.

I may not understand a lot of what goes on in the world of comic books. But I now understand how important they were to my cousin and his father. That insight was truly a gift, one of my favorite series of interviews I've ever done. And I didn't need to be an expert. I just needed to make his interests an open-ended part of the conversation, and listen as he told his story.

What Should I Bring?

What are your goals for your interview? If you're looking to preserve a conversation, as well as the information it contains, much of what you need is probably in your pocket. Phones, tablets, computers, cameras, and audio recorders all have the capacity to record simple audio or video. How good you want that record to be will determine the equipment you need to use.

I personally enjoy making audio recordings. They tend to be less intrusive and intimidating to my family members. One set of recordings I treasure are from a lunch date with my mother, grandmother, my great aunt, and younger sister. I prepared a list of questions for my grandmother, and planned to record her answers with my cell phone.

Instead, my mother and great aunt jumped into the conversation as well. I left the recording running to capture it all. What I have instead is the banter between the women in my family. The sizzle of food cooking, the sound and volume of different voices, the laughing and shouting as conversations ebb and flow. Topics and stories I hadn't planned for. A verbal snapshot in time, as valuable as any picture.

From making it, however, I made an unfortunate discovery about smart phone audio. Many cell phones record in odd micro formats that aren't compatible with other audio players or devices. The format may also be low quality, and susceptible to data loss. For more on the basics of audio formats and preservation, see this guide from FamilySearch. And if you need something to play an audio file in one of these formats, VLC is great. I've yet to have a cell phone produce a weird file type that VLC couldn't play.

You may not be able to change the file type your phone uses when it records audio. If you have a smart phone, you can often access greater functionality through different apps. Searching your app store for audio recorders is one option. Searching for converters for ht audio files you already have is another. But don't forget that many of the major genealogy apps have added audio capabilities. Even if you can only upload the recordings you've already made, let this be part of your preservation process.

Interviews with family members are invaluable sources of new information. They provide richness and depth to the records we're creating about the people we meet. Whether they're our closest family members, or distant cousins we met on the internet, let interviews become the lifeblood of your genealogy. Nothing else you do will bring your relatives more to life!

What are your favorite tips for conducting family interviews? Who was your favorite relative to interview? Any horror stories you want to share? Let us know in comments!

Thursday, August 3, 2017

Uploading Your Tree to GEDmatch

GEDmatch has become an essential tool to everyone who is serious about genetic genealogy. I'm not sure how many matches you have there. But the last time I saved my match list as a PDF, it was 23 pages long. And the last time I checked, the last page of matches stopped at about 10 cM. As more people take these tests and upload their raw data to GEDmatch, making traceable connections to our distant cousins becomes easier every day.

But are you taking full advantage of everything GEDmatch has to offer you? If you don't have a tree uploaded, the answer is "No!"

Why You Should Have a Tree on GEDmatch

In order to take full advantage of what genetic genealogy has to offer, you need three types of information:

  1. Segment data
  2. A tree
  3. Shared matches
I like to compare this to a car. In order for a car to function, at the bare minimum it needs wheels, an engine, and a fuel source. Without any one of those elements, the car simply couldn't move. In genetic genealogy, these same three elements are THAT fundamental. Some testing companies create this data scarcity by not providing tools to convey this essential information to their users. And given that some of the newest testing companies are gravitating away from this bare minimum, it leaves some newcomers in our community at a disadvantage.

Whenever a testing company has refused to offer one of these necessities to its customers, GEDmatch has been the equalizer. AncestryDNA won't provide segment data? GEDmatch will do that. MyHeritage DNA won't provide segment data or shared matches? GEDmatch will do that. That same capacity for tree sharing is also available. The ability to cross reference people on all of the sites on which they've tested is there. But how many of us are taking advantage of it?

I've used trees on GEDmatch to match users there to their original testing companies. Once I know I have the right person on AncestryDNA, their shared matches become a powerful finding tool. I compare those shared matches to what I know about the segment data. A much clearer image of the relationships at work in my DNA begins to emerge. It makes my job on GEDmatch AND AncestryDNA easier. And when I invite these shared matches to upload to GEDmatch, and they accept that invitation, documenting and preserving this record becomes possible.

How to Upload a Tree to GEDmatch

Now that I've been focused on linking GEDmatch users to their original testing companies, I'm noticing a trend. Many who provide trees at their original testing companies haven't done so on GEDmatch. Some of these users may not realize the capability to add a tree exists, or how to do it. So I've made a video for our YouTube channel to explain the process.

The trees they produce are not fancy, by any stretch of the imagination. But they provide needed information to the discoveries we're all trying to make. If you haven't uploaded a gedcom to your GEDmatch kits, be sure to add one today!