Posted by Ancestry Team on May 21, 2014 in Development, Mobile Development, Web

Last October Ancestry.com acquired a very exciting property called Find A Grave which focused on collecting content around the graves of family, loved ones and famous people. With the acquisition we wanted to take Find A Grave to the next level and provide the current users new and better experiences around consuming and contributing content. Find A Grave has been around for over 15 years and has done a tremendous job of organically growing the website and user base.

Around the time of the acquisition I was asked to lead the engineering efforts, manage a successful transition and focus on new experiences. But this was a challenge because Find A Grave serves millions of page views a day, was running on a different web technology stack, sitting in a set of physical rack mount servers in a separate data center at the time and contains over 100 million photos and over 116 million memorials.

Our objective was to transition all hardware and code to a modernized infrastructure within a short amount of time to provide the Find A Grave users with a better performing experience (quick page load times) and support future scale and growth over the years.

We also knew we needed to build a mobile app; it just makes sense. One of the primary goals of users for the site is to go out into cemeteries and take photos of grave sites. Before the Find A Grave app it was an arduous and quite disconnected workflow to take photos of a grave site, come back home, download all of the photos to a computer and then upload each photo to the site one by one, etc. As you can see, if you have 100 photos to process this can be quite a pain!

So our goal is to make mobile front and center for Find A Grave and we started with iOS.

Building the Team

The first step was to build the team and get moving, and I quickly hired two amazingly talented iOS developers to focus on the iOS app, John Mead and Shengzhe Chen. John came from the freelance world working on many different iOS applications for different companies and startups. And Shengzhe was doing mobile payment development before joining Ancestry.com.

We then hired a backend engineer to come in and bring some really awesome skills to the team, Prasanna Ramanujam. He is currently positioning Find A Grave to be a very API driven application which allows us to work with many external clients from the ground up. Prasanna comes from NodePrime as an engineer on their Node.js related datacenter infrastructure, and previously from VMware doing Node.js evangelism.

Next we hired another full stack engineer hero, Bob Dowling, to first focus on building a new, modern and flexible fronted to the application. We want to provide a more snappy and performant fronted to the website that ever before which includes optimizations throughout the whole stack. Bob comes from the freelance world as well focusing on both mobile and web applications.

And just recently we have an awesome new intern helping out, Shruti Joshi, who will be working with us to get the new website built.

And of course our product owner Mike Lawless, our designer Jonathan Rumella and the founder of Find A Grave, Jim Tipton, who have all been a tremendous help with all aspects of the team.

Infrastructure

Within a few months we were able to migrate all infrastructure, photos, database and content to the Ancestry data center. It was a very complex task as we needed to keep the existing site running live during this whole process. We also wanted to modernize the infrastructure and virtualize everything in our private virtual cloud. Part of this was using Go for continuous integration and Chef for deployment and provisioning to our virtual instances. This will allow us to scale to future growth beyond the acquisition.

iOS

As a team we worked closely with our product manager, designer, and the Find A Grave founder to work through what we thought would be an amazing 1.0 iOS application. We were able to work very closely using some extremely lean principals to build quickly and efficiently. From conception to release the two person team was able to build a 1.0 release that contained over 10 major features to get a great mobile experience into our users hands in just about four months. We released the first version to the public at the beginning of March 2014 and we’re excited that millions of people will be able to use this over Memorial Day weekend to remember their loved ones. You can check out the app here.

Backend

Find A Grave grew as a Perl based web application over the past 15 years, so we are looking to build things from the ground up using Node.js. It works perfectly for our JSON API layer, provided great real-time support for our iOS app and works great for some of the new NoSQL infrastructure we’re putting in.

Website

We also know that the Find A Grave website needs a new, fresh facelift. Utilizing our UX team we are beginning to envision what that new look and feel looks like and trust us, it’s going to be amazing. This is where we are going to harness the power of a very client-driven web application that will utilize the new backend API services heavily using Backbone.js and Handlebar templating engine.

Accomplishments

To recap, we’ve built an amazing Find A Grave engineering team. We migrated all infrastructure in-house and virtualized everything through proper continuous integration and delivery platforms. We’ve released an iOS app with some more great features coming out soon. We’re rebuilding the backend from the ground up. Next on our list is giving the website a new facelift to provide a better user experience. Whew, that’s a lot in just 6 months to get started on.

We have a lot more to come so stay tuned.

Comments

  1. Mary Herzog

    I’ve been an active contributor to FindAGrave for almost five years. I like the look and feel of the website but I find the number of clicks necessary to create a memorial or to add a photo to be excessive. Please keep a goal of fewer clicks in mind as you redesign it. I use my iPhone to take photos but I have been reluctant to use the app because when I am out in the cemetary taking a row of pictures I have no idea whether or not there is already a memorial posted. I don’t to contribute any duplicates. Thank you, Mary

    • Robert Schultz

      Hi Mary,

      Thank you for the feedback and it’s great to hear from a long time user as we really respect and value your feedback. Improving the user experience is one of our top priorities of the new design of the website as we want to make the overall process as simple as possible. I will communicate your feedback to the rest of the team.

      Regarding the iPhone version of the app, we are working hard to provide the best experience as possible. We just released a new 1.2 update which makes uploading photos easier than ever as they will upload in the backlog with some settings to minimize usage of your data plan. The next iteration of this will allow you to do more of a batch process so you take photos as quick as possible and come back later to attach them to the memorials.

      We appreciate your feedback and look forward to deliver the community more and exciting features in the future.

  2. Tessa Keough

    I notice this is for Apple/Mac. Any consideration being given to the many Android users or did I miss something? Thanks

    • Robert Schultz

      Hi Tessa,

      We are working towards supporting the Find A Grave experience on Android as soon as possible. As soon as we do we will make an announcement. Trust me, we are working hard to provide a great experience across all platforms.

  3. I have been a long time user of Find A Grave. I have been a member for over 7 years and submitted a few findings. But I have been a user for much longer that the 7 years.
    Find A Grave is one of the first places I look when I get a new person to track down.

    I would like a like to add the memorial url (not the search url) to FTM’s url list, but also be able to show these urls within a report and/or chart.

    Also there needs to be a way to submit where the same grave and even cemetery is listed more than once. The cemeteries may change names or be known as different names by different people, but they are one and the same.

  4. Bruce Reeves

    I have not played with the app much to this point, but I will. The one thing I noticed today, that would probably help others, is possible a map entry into findagrave – so given where I am – it would show me on the map the cemeteries in the immediate vicinity.

    I realize that not all cemeteries currently have GPS coordinates, but for those that do, that would be most helpful.

    Thanks.

  5. Jim Antley

    I am also interested in the API. I developed a findagrave family mapping web based application that no longer works.

  6. Jim Antley

    My mistake, the application is working again. I had not checked it lately. Still interested in the API.

  7. I’ve been an active member of Find A Grave for just over one year. Over all I have no complaints regarding the technical aspect of the site, rather the issues seem to arise from length of time before tech support emails to Find A Grave are answerd, and the occasional contributor who provides no way to contact them, and creates mildly arrogant and/or dictatorly paragraphs witin their bio page towards other contributors. Additionally, in some cases it seems like actions are being taken with certain memorials (Example: sponsored Find A Grave Memorial# 121464906 – add photo feature disabled – “Find A Grave has determined no additional photos needed for this memorial”) from time to time that can only originate from within Find A Grave, with the creator and/or manager of the page left to wonder what is happening.

  8. Nick

    My comment is more to the wish that the powers that be would lax of one of the rules for findagrave: using external links in a bio.

    Rather, many times there is more information that you would like to include about an individual or their family without writing a lengthy essay on the bio. Links let you keep the “meat” on the bio with links out for those that want more research info as well as indicating that other sites also have info on this individual or his/her family.

    NOTE #1: I have no desire have hyperlinks to pay sites, spam sites, or sites requireing membership or personal information from the viewer of a memorial.

    NOTE #2: I do understand that a link can break. In that case, by all means the memorial owner should have to go in and fix it if the link has moved, or remove it if the link is deactivated. However, if the link is on topic to a free URL, I believe it should be allowed.

    As an example of a link: There was a book about the history of a particular New England town that Google had scanned in and made it available FOR FREE on the internet. The hyperlink went to the exact page in that online book where it mentioned the ancestors of this individual.

    For another example, I made a link to a woman’s marriage license. The descendant of this woman belived her maiden name had an “s” at the end. That was not the case and these links showed it to be the case. This is important because the family where the surname was WITHOUT the “s” goes back to mid-1600 New England.

    Again, I can see where external links could be misused, but if the link is on topic (of genealogical interest for this person or their ancestry), is not to elicit money, membership, or information from the viewer of the memorial; I think it should be allowed.

Comments are closed.