Connecting with communities: citizen generated content strengthening relationships and enriching collections:
an Online Cenotaph case study
Michaela O’Donovan and Victoria Passau, Auckland War Memorial Museum. This article updates a presentation at the 2015 LIANZA Conference in Wellington.
Museums and libraries rely increasingly on social media interaction with customers for engagement, promotion and crowd sourcing labour for particular projects, but are we thinking in terms of purposeful connecting, meaningful relationships and enriching collections?
Auckland Museum’s Online Cenotaph, developed in association with memory institutions across the country and launched in late January 2015, was actively designed for digital collection development using social media tools — to harvest the information and digital versions of objects in the community and harness community interest in the commemoration of the people who have served for New Zealand Aotearoa — while still maintaining the integrity of the Museum-provided information in the core database.
We’ve taken a look back over the last couple of years to determine how effective this approach has been. Has the public taken it seriously? What about misinformation or mischief-making? How does the User Generated Content fit with the authoritative data set? How has the strategy of no moderation fared and what design compromises had to be made? What are the outcomes so far — including the types of engagement from the community, challenges encountered in the first 2 years online and implications for resourcing.
A bit of background
Online Cenotaph, is a national resource which provides a record of individual New Zealanders’ service in international conflicts and a lasting legacy of the WWI Centenary commemorations. Auckland Museum redeveloped the original Cenotaph database and the user experience in collaboration with New Zealand’s Ministry for Culture and Heritage, memory institutions and community groups, large and small, across New Zealand and internationally to be the online gathering point for the personal and official memory of people who served for New Zealand Aotearoa.
Auckland Museum regards the content in Online Cenotaph as digital collection, with its own scope, rights, retention and preservation considerations.
Value proposition, aspiration and challenge
The Museum’s aspiration was to develop the service records in Cenotaph into a modern, attractive resource, that is a recognised collective authority on the experience of New Zealanders at war, and provide this resource with functionality to allow the community to augment and expand the records and overall content.
To deliver this, our two aims were to:
· Present a credible core of information in each record in a form that encouraged and excited both researchers and the general public to add content. That is, to provide a starting point that people saw had sufficient merit, substance and longevity, to warrant them spending their time adding content
· Make available tools that were easy to use and allow those engaged users to add content.
The credible core of information came primarily from the original Cenotaph database. A biographical database, whose content had been gathered over 17 years prior to the redevelopment, Cenotaph had pulled together information relating to individual service people from a huge range of sources such as embarkation rolls, military service and other government records, newspapers and families..
The Museum’s intention is for Online Cenotaph to eventually provide a record for every person who has served our country in international and national conflicts. We wanted it to be an engaging social space as well as the comprehensive online hub for the stories of New Zealand service personnel.
When developing the business plan for Online Cenotaph, we were aware, from the Ministry for Culture and Heritage’s research into New Zealanders’ connection to WWI, that many New Zealanders felt they did not have a personal connection with that conflict (http://ww100.govt.nz/benchmark-survey). We also knew that the number of service personnel who went overseas was in the order of 1 in 10 of New Zealand’s entire population at that time and we felt sure there was a lot of untapped information still in the community. Finding ways that people could connect with and share what they have and what they know was a fundamental challenge for the project.
To do this, the existing Cenotaph database had to become much more than a standard data repository which serves up pre-selected and verified information — it needed to transform into an interactive, intuitive experience for users, while remaining respectful and honouring the Museum’s commemoration remit.
Our thinking was that an amalgam of the type of social media features people routinely use in experiences like Wikipedia and Facebook, but all relating to those who have served, would support both objectives of engagement and information ‘harvesting’.
So, our aspiration was for the new look Online Cenotaph to be:
· a continual work in progress, while providing an increasingly rich biographical database of New Zealand service personnel
· a partnering experience for end users which is both interactive and empowering, by making it possible for them to contribute information, documents and photos, as well as the personal stories of the experiences of New Zealand service men and women
· about sharing and not giving away — there is no requirement for the original documents and photos to be passed to the Museum, but the Museum’s aim is that what is shared online will be retained and taken care of digitally in perpetuity
· a living online social space for enthusiasts, family and researchers alike — where people can contact each other without the Museum needing to be the intermediary
· linked to other information available on the web to join the dots for every service person
· a New Zealand research node, linking with other related sites globally
· a useful social history resource about New Zealand’s experiences in international conflicts.
The starting point
So, how to make this happen? From here on, we’ll use the words customers, the public, people and users interchangeably to signify the citizen researchers or members of the public who both consume and enrich the information in Online Cenotaph.
It’s interesting to see the expectations of people in this ubiquitously digital age. People routinely think they can find everything they need online. In this profession we know that often they can — maybe it’s in multiple sources and maybe it’s collated — but often it’s just not possible.
Information about New Zealand service people is much like this, so it’s useful to understand at this point how the database content originated. The original Cenotaph records had been created manually, by transcribing the New Zealand Expeditionary Force Embarkation Rolls into individual biographical pages. These Rolls are the handwritten lists of New Zealand Army soldiers who boarded troopships in WWI, and contain the information these men and some women provided when they signed up — with the inevitable inaccuracies, omissions (such as Iwi and Hapū connections, which were not asked for at the time of enlistment — fundamental information for Māori customers in particular and in fact all researchers), and variations on names and ages that public records can contain.
Then the records had been enriched as time and resourcing permitted, and also expanded to include other service groups like nurses and masseuses (the forerunners to physiotherapists) using information from newspaper articles, published military and unit histories, the service records held by Archives New Zealand, death notices and obituaries together with online resources like Births Deaths and Marriages registers and Ancestry.com. Information also came from family members who had contacted the Museum. It was a painstaking, slow and labour intensive process for which the Museum is immensely grateful to its team of dedicated volunteers, many of whom were retired service people and their relatives, who assisted in this process.
As well as creating the starter records, the Museum team needed to verify the information that was coming its way — resulting in huge backlogs of material having been submitted by people and waiting for Museum staff to check and process into the database. Resourcing for this created a massive bottleneck and frustration for people who wanted their relative’s record updated.
So, we had a situation of people wanting to contribute information, resource hungry database processes, bottlenecks and an unsustainable way of processing and publishing this material — in the lead-up to the commemoration of New Zealand’s involvement in WWI and also WWII. As part of the upgrade we needed to be realistic about what we could and couldn’t resource and find practical but credible ways to deliver on the aspirations for Online Cenotaph.
The solution: data and design
Put simply, the approach was to facilitate and collaborate with people to enrich the records, by taking the Museum as intermediary out of the process as much as practicable.
The Museum still has to verify a service person and initiate a record, but from there on it’s a collaborative effort with users. In fact, users frequently advise us of a ‘missing person’, for the team to check and initiate a record for.
So, the decision was to empower people to add information, but partition off the information which had been verified and entered by the Museum team. The social media features have enabled this to happen. The site makes it clear where the information has been contributed from or by whom, by displaying the source in every field. Several design treatments for this were tested, with the resulting design judged to be most effective.
We anticipated that the new Online Cenotaph needed to have places where people could add information, comments, photographs, documents and captions. This meant the underlying data structure needed to support this and the design of the user interface needed to be radically different from the ‘snapshot’ one-screen views that users were accustomed to.
The underlying data structure was completely reviewed and rebuilt in the source Vernon database to allow for more granular information to be entered against each person. For example the Vernon interface now enables staff to link related pieces of information regarding embarkation or enlistment — Rank / Occupation; Event; Body; Unit; Vessel; Place and Date. In turn, the source of information for each linked piece of data can be uniquely referenced.
To do this the team needed to redevelop all the hierarchies such as rank, unit, medals and honours and map the existing data into the appropriate fields. There was the inevitable data cleaning to be done. Both are huge exercises and we’re still working on them, but they are considerably improved and serviceable.
We adopted the Iwi Hapū Names List[1], with the addition of an ‘other’ option which recognises the potential for self-identification, in place of the Museum-developed list previously in place. This took a significant effort to map into Vernon.
For the user interface, the biographical screens needed to accommodate greater amounts and more complex information. The design solution for this was to have expandable sections and views which allow people to only call up and see specific fields.
We also wanted it to be possible for people to add to, provide alternative views or comment on information that is in the Museum’s core records in the source database, without compromising these records. Our solution was to present multiple views of the same fields, with clear attribution of the source of each view.
The user interface requires people to provide credential information — name, relationship to the person and email address, in order to be entitled to add content. So far we are not aware of any spurious or false credentialing and we trust the social media process to winkle these out. That said, our research shows that a significant proportion of people who try to add information, drop off during this process, so we’ll be trying to make this a much better experience for users over the coming 12 months.
Users also have the option of having their email addresses published. More than two thirds of contributors have agreed to this and we are seeing both multiple contributors and cumulative discussions between contributors on some records. This too, adds to the length and complexity of the records.
We know that a significant proportion of users are accessing the Museum’s online resources from mobile devices. So a fundamental principle was that the design and social media features needed to make sense visually and work effectively on smartphones, tablets and other mobile devices as well as desktop scenarios — adding even more length to the individual biographies.
This caused some consternation and a short flurry of comments from people who were used to the previous compact screens[2]. The new layout seemed counterintuitive to people accustomed to working on the previous database (e.g. genealogists), particularly when the new screens have more white real estate. Responding to this feedback, we prioritised a more compact print view in a second phase of development and this issue seems to have quietened down.
The decision was made to physically partition the user generated content from the Museum’s source data, for both technical and security reasons. The source data is held in the Museum’s collection management system, Vernon, whereas the user generated content is stored within Kentico, the Museum’s online content management system.
The source information is pulled out of Vernon into a Linked Open Data middleware layer (the Digital Content Platform), which the user interface searches across. While this looks seamless to users, the interface in fact brings together data from the 2 different sources and displays them as one record. While this helps protect the integrity of the Museum’s source data, the drawback is that the user generated content is not searchable at present.
Further, the Museum’s other collection management information housed in the Library Management System and the Museum objects information in Vernon also feed into the Digital Content Platform, with the result that all items relating to one person can be retrieved in a single integrated search on Collections Online. For example, a person’s Online Cenotaph record, a photographic collection, and a uniform will all be found through one search on Collections Online.
Storing the user generated data in Kentico as opposed to the Digital Content Platform, means that it will not be available to future applications that use the Linked Open Data as their key data source. It’s very early days for using the Museum’s Linked Open Data and this approach may need to be revisited in the future.
Verification, rights and moderation
Principles were developed in three relevant areas — verification, rights and moderation.
Firstly we determined how far the Museum was going to verify the material submitted as user generated content as opposed to the Museum-created records. Our decision was to not verify at all, which seemed practical given our aspiration of self-regulation and the huge volumes of material being added.
However, there have been instances when we have needed to revise this after correspondence with users. For example, we will verify content when a member of the public contacts us directly, through the Online Cenotaph email, about a date of birth or death or alternative spelling of their name. This is because the dates of birth or death when added by the user do not populate the biographical section at the top of the page, which has caused some frustration. Also, as user generated content is not searchable, alternative names are added into the source data to enable family or researchers to discover records with previously incorrect spellings.
Secondly, we needed to make users aware of the rights implications of uploading information and documents. This took many discussions to work through, resulting in the following rights statements on the site.
Thirdly, the user generated content is un-moderated. It is, however monitored for mischief and inappropriate language. That said, eventually we hope to be in a position to add some of the user generated content (particularly birthdates and Iwi Hapū names) back into the source Vernon database. This will be a staff-assisted process of checking and mapping the information, and will then have the benefit of being findable and usable.
The user generated content is published in real time. In our view, users now expect to see their content as soon as they’ve entered it. This in turn fosters engagement.
And finally, in this section, a note on the implications for the overall website, as Online Cenotaph is an integral part of the Museum’s online offerings rather than a standalone product. The Museum’s website style guide was originally developed for information ‘pushing’. During the build/test phase, the nature of users’ interactions highlighted areas where the overall website design needed to be modified to make it work for interactive transactions. The Museum’s next online offering, Collections Online which went live in June 2015, was able to build on the initial experience from Online Cenotaph.
Results so far
The social media features have transformed the site from being a static ‘push’ information tool, where the Museum had to add every piece of data, to a lively, busy, international social space. The engagement with Online Cenotaph was immediate and so far has been sustained, with highlights from the first 2 years below.
The redevelopment makes it possible for people in the community to enrich this national resource themselves. New features include:
· a personal/individual call to action: greater opportunities for public contributions to the database, including uploading photos and scanned documents and filling in missing information (e.g. only 5% of records had date of birth and a mere 370 records held Iwi/Hapū information)
· a group/community call to action: an opportunity for community groups (eg alumni, schools, sports clubs) to honour the service people associated with them by completing the records for every soldier/service person, particularly with information from their archives
· the ability to pay tribute and honour individuals by laying an online poppy each year
· the ability to include details of current/retired service people, where formerly Cenotaph only published records for those the Museum could confirm had passed away — these records are viewable as ‘Preliminary Records’.
· a showcase function on the home page, to honour service people in line with anniversaries of particular battles, events or contributions.
We are satisfied that the two primary aims for the project are being met. Our first aim was to present a credible core of information that people saw had sufficient merit, substance and longevity, to warrant them spending their time adding content. The second aim was to make available tools that were easy to use and allowing those engaged users to add content.
We are also pleased with the volume and range of material that has been added, as it’s unlikely this material would have been made available to Auckland Museum any other way and the resourcing required to include it would not have been sustainable. Users generating this content has freed up our own team to complete the digitisation and mapping to create new starter records for some 80,000 WWII army personnel this year alone.
From a qualitative perspective, we’ve noticed.
The quality and range of the user generated content has been impressive, so much so that it has lead us to think that many people had already done this work out of interest or as a way of honouring their family members, and are delighted to have somewhere to share it with pride. It’s unlikely that the Museum would ever have had access to this material otherwise.
This is especially true for those soldiers who served in the Australian Imperial Forces. For example Peter uploaded five fully transcribed diaries on to Ralph Dorschel Doughty’s record. Ralph, a member of the AIF Field Artillery, 9th Battery, died 25 July 1917.
A family member of Ella Cooke has uploaded documents into her service record.
The range of content added spans the whole continuum from a single date or nickname through to deeply felt tributes and updates on how families have taken the servicemen’s memory forward, through to lovingly transcribed letters and diaries.
At times we’ve been surprised at how enterprising people been with the user generated content features — one small regional Museum began touting for documentary heritage related to servicemen from their area, on their local men’s records.
Many people have wanted to connect with others who have a common interest in a particular service person. There are many, many instances of distant family members connecting through a common relative. In this record the two Anns were able to connect as they both allowed their emails to be made available:
There are also records where individuals are able to remember their relatives on behalf of a deceased family member.
A note from an Italian man who met a young New Zealand solider during the Second World War was added:
During World War II, in the year 1944 in the town of Forlì, Italy, my grandfather was 14 years old, and met a young and friendly soldier from New Zealand.
My grandfather tells me that this soldier, whom he had never met before, asked to be invited for dinner at my grandfather’s home, since he wanted to share some corned beef that he had brought from home with my grandfather’s family. Attracted by this delicious food, my grandfather decided to take this soldier home.
Initially the presence of an unknown person was a cause of concern for the other members of the family — my grandfather describes him as a tall, strong man — but subsequently everyone became attached to this young soldier, who was really a good guy. This first dinner was followed by many others.
One night the soldier brought a cake he had received from his “mama” in New Zealand, and he shared it with the Italian family.
My grandfather still remembers this tall young boy crying like a child when eating his mama’s cake…
Almost 6 months later the family contacted Online Cenotaph and asked for the email of the contributor. The Italian contributor in turn responded that her grandfather was ‘extremely moved’ to hear that they had been in contact.
The features have also helped people ‘do their bit’. A heart-warming example was of a user who had found a set of dog tags in a second hand bag and was able to return them to the delighted family now living overseas.
The Museum library team has increased its network with researchers and family members through the collegial relationship engendered by the social media tools. One researcher added 1500 separate pieces of data from her own personal research in the first 6 months, and built a sound relationship with the team in the process.
So far, malicious or misinformation seems to be minimal, with people self-regulating particular records. The onus is on users to let us know if anomalies exist and they are! People have let us know where they feel there are two records related to the same person, most supplying detailed sources for verification — we are happy to work with people to correct these anomalies.
Other organisations including regional museums, schools and cemeteries are proactively adding and linking material from their honour rolls, plot maps and archives into individuals’ records.
An exciting development now, is that community groups who have been working on their own WWI projects are beginning to send us their data in a format that can be readily uploaded into Online Cenotaph — this really is the power of many at work. For instance, the Panmure Branch of the New Zealand Society of Genealogists had researched and cross-referenced the names of those men who signed a flag created to commemorate Cyril Bassett V.C. Our Soldiers Flag is held within the Auckland War Memorial Collection and the Museum team was readily able to match the image of the flag to over 345 records.
So, what’s not going so well?
Responding to expectations of perfection is one of our biggest challenges. Some people are upset if there is an error on a family member’s page — most, but admittedly not all, of which are as they appear in the source documents. We do not change the information from source documents, but do add the alternative information provided by users.
As anticipated, not all members of the public were so enthusiastic about the new look or feel. We became aware of a ‘future shock’ effect happening for some long-time, practised Cenotaph users who were accustomed to using Cenotaph as a ‘push’-only information resource — where they often printed out records and referred back to them.
These people sometimes mentioned they were unfamiliar with social media tools, did not necessarily see the benefit of user contributions and found the ability to contribute challenging and disempowering. This was unfortunate and we redoubled our efforts to work with these people and help them over that particular learning curve.
Others simply liked and valued the old look and feel, with its engaging sepia-toned historical design, which we understand.
Then of course, we have encountered the inevitable mischief makers who add malicious or obscene content — surprisingly few however. The Museum’s normal security and bad words indices do pick up most of the malignant and mischievous traffic, although they are not fool proof — from time to time the team has had to email users with surnames such as Cockburn, and explain why the system has not allowed them to add content on their first attempt!
Overwhelmingly among the vast majority of users we have observed a strong sense of pride and people playing their part in filling out their relatives records –in some people we’ve corresponded with, it appears to have become almost a matter of honour.
What we’ve learnt along the way?
Auckland Museum Library’s preferred approach is to collect original versions of manuscripts, photographs and archives. However, our experience is that using a user generated content approach to augmenting ‘digital collection’ resources and collection knowledge/context is absolutely worth pursuing in a targeted way, for specific heritage collection areas. While we have encountered a range of challenges and conundrums and are sure there will be others coming our way, using this approach has resulted in a superior product, sooner and at a lower ongoing resourcing cost to the Museum.
We know that what people see as ‘official’ information is not necessarily accurate or correct in all instances — we are all aware of historical accuracy being influenced by any number of factors — including who documents the story. But by presenting as many views of this information as come to light, we hope Online Cenotaph will continue to augment New Zealand’s documented history.
It is also apparent to us that people have been willing to share their information and content when they are not put in a position of deciding to give away their items — we believe this is a watershed moment in our understanding of what Online Cenotaph and connecting digitally could be — and are excited at the possibilities.
At this stage, we’re not envisaging opening up the collections for more user generated content. Instead, further uses of social media tools are likely to be selective and in controlled instances.
And finally, we’ve begun to consider what the Online Cenotaph experience could mean for our collection development practice and our roadmap for Collections Online. An example of how this could play out, to illustrate the point only at this stage, would be at the time of acquisition. Imagine the power of inviting donors and whānau to participate in an Online Cenotaph-like experience, where people could (if they chose) each contribute their relationship with the document or taonga and the stories of that item, in their own voice and language and unfiltered. This practice would be based on the notion that a person (or a taonga), as a ‘node’ on a network, is in part accorded its mana through its relationships.
Wrapping up
While not perfect, Online Cenotaph is meeting the aspirations in its business plan to be a national resource and online gathering point for the personal and official memory of people who served for New Zealand Aotearoa.
Our first aim was to present a credible core of information that people saw had sufficient merit, substance and longevity, to warrant them spending their time adding content.
The second aim was to make available tools that were easy to use and allowed those engaged users to add content.
We’re satisfied those aims are being met. Auckland Museum Library’s experience of using social media tools for collection development has resulted in a rich resource and a dynamic and rewarding outcome for the Museum and, we believe, customers.
[1] Source: Iwi-Hapū Names List used with the permission of the Māori Subject Headings Governance Group
[2] Order of scale, 30 ‘brickbats’ compared with over 1300 ‘bouquets’
