CONNECTIONS – Project Analysis

PROJECT SUMMARY

CONNECTIONS is a system for building online semantic relationships between people, places and things. It all begins with a person. Who were/are they? What did/do they do? Where did/do they live? What evidence can we discover about the life of this person?

Many existing genealogy applications require fees, or are stand-alone programs that work offline. Users are not encouraged to share, or think holistically about their data. Rather they create fragments of a wider network of people. Instead of building upon existing trees of information, users tend to repeat data by creating their own familial clusters.

CONNECTIONS aims to provide a free system to enable visitors to the website to quickly discover and navigate information that is relevant, reliable and engaging – something greater than a genealogy site. Users can visualise and generate connections between people, places and things. These people, places and things can then be further linked to other people, places and things in an ever expanding web of stories. In this way users can create multiple clusters of disparate hubs of information that can eventually be connected via user defined, associated relationships.

CONNECTIONS is a tool to allow today’s historians to leave breadcrumbs for future generations to discover their own history.

PROJECT ANALYSIS

CONNECTIONS meets the primary objectives of the project, to create:

  • a cost effective solution for online archiving of personal information
  • easy to share collections of familial information
  • solutions to improve the efficiency and effectiveness of genealogy research

CONNECTIONS is provided free for use. There are absolutely no fees for guests or registered users. Anyone can create a free account to gain access to the data. In future plans, development of the site will require a small budget to maintain hosting and offer new features. These costs may be offset by voluntary contributions from users – however the guiding principle of the site will be to provide a cost effective solution for online archiving of personal information.

Users can create multiple hubs of data that can be linked by relationships and locations. Users can link to material from collections held in external online repositories, such as The National Library of Australia’s digitised newspapers. Anyone can add and edit data, upload high resolution images and documents (up to a file size of 50MB) and connect these to personal records. All objects are stored in a central repository providing easy to share collections of familial information.

CONNECTIONS is a new tool for historians and genealogical enthusiasts to explore multiple avenues of associated data. With unrestricted search functionality, and the ability to connect data through many points of interest, this site provides solutions to improve the efficiency and effectiveness of genealogy research.

LESSONS LEARNED

Throughout Exploring Digital Heritage, we’ve manipulated a variety of datasets – utilising a number of existing applications in an effort to determine how digital tools can be used to enhance our understanding of cultural heritage.

During my research, I used various genealogy websites, such as www.ancestry.com, to collect information about my own family; however, I found that I couldn’t easily extract these digital assets. I had many paper based records along with the information I found online, which led me to believe that I would need to create my own digital dataset, in order to test and demonstrate my ideas.

To build a truly extensible repository, you need to work with a standardised set of data. In the case of genealogy, the recognised standard is GEDCOM – a specification for sharing data between genealogical programs. This format was developed by the Church of Jesus Christ of Latter-day Saints in 1996, and though there have been a few recommended updates, it’s still the industry standard.

I explored a few options to find an existing tool that had the features I required to meet the objectives of this project, such as:

  • import/export of GEDCOM files
  • the ability to attach images and documents to personal records
  • privacy filters to restrict the display of sensitive material to casual visitors
  • access control groups for administrators, users and family groups
  • Google maps display of geolocated data
  • A discussion forum to build an online community

In the end, I chose HuMo-Gen (www.humo-gen.com/humo-gen/cms_pages/11/). This program had most of the features I needed. Additionally, I implemented phpBB (www.phpbb.com/) as the community forum.
HuMo-Gen allows for storing a number of attributes for each member of the family tree, from the basic names and dates to locations, witnesses, and sources. You can also attach files to any family member. The program can generate a number of reports, including ancestors, descendants, timelines, and an outline view. Also featured in HuMo-gen are access control groups, allowing you to decide which information is public, and what additional information is revealed to any number of tiers of access.

HuMo-Gen is designed for individuals to add small family groups as users to edit their personal histories. Which is not exactly what I wanted, so there were a lot of changes that needed to be made to the display (PHP/HTML) and database (mySQL) code. Reverse engineering someone else’s code takes a long time – learning what the code means is all voodoo and black magic to me, so I had to have some help in understanding and modifying the code in order to get what I wanted from HuMo-Gen.

There were extensive modifications to the layout and display of both the front-facing site and the administration interface, as well as some reworking of the database tables. The Google Maps integration had to be completely pulled apart, as it didn’t work properly, in fact not at all with Chrome – and still doesn’t work properly due to several coding issues from the original source.

Many of the challenges in modifying the program came from the fact that the original author is Dutch, and while his English is excellent, his use of ‘interesting’ descriptions in his comments led to some confusion in interpreting functions within his code. This confusion can also be seen in some of the data entry interfaces – for example, the use of ‘Add’ and ‘Save’ are used interchangeably in some cases.

At this point in time, I have documented almost 200 individuals, and I’ve learnt that manual data entry is tedious at best. But now that I’ve done all this work, I know my efforts could be exported and used in multiple ways, by future historians. Equally, to expand this human network, I can easily import a GEDCOM file provided by another genealogist.

FUTURE PLANS

With this project, my aim was to bring information about people, places and things into one free for use research website – a single, ever expanding human tree, as opposed to a single family tree. But my ambition exceeded talent and time. My initial vision of this project attempted to solve too many problems at once. I didn’t understand the true time and effort required to bring the whole of this project to fruition.

CONNECTIONS is still a grandiose plan. There are many features that set it apart from other genealogy applications; however, some of the proposed key functions didn’t quite make it into this phase of development.
Some functionality that will be incorporated into future development will be:

A more user friendly data entry interface – The data entry screens are complicated to understand. A revised layout, with a more modern approach to data organisation would greatly benefit users. Also, there needs to be more help information, perhaps as hover actions when a user selects a field.

Liberate users from the search box – Provide entry points for users to browse the website, without knowing what data is available, such as enhancing the home page with a gallery of random images that link through to individual records.

Prevent users from creating duplicate records – Currently duplicate records can be identified via a manually run admin report. It would be more appropriate for the system to automatically notify a user that a similar or same record exists before the data is saved.

Simple download options for attachments – Images and documents need a ‘download’ button. One of the benefits of this site is to create a repository to allow users to share high resolution images. The ability to share information is key to the CONNECTIONS ethos.

Privacy filter for living persons – Currently, a privacy filter is applied to living persons so casual visitors to the site (guests) can’t view images and sensitive information. This filter is not in place for registered users. In future releases, this privacy filter will be in place for everyone; however, users will be able to lift restrictions to a group of ‘Friends’.

Separate events and places data – Events and places should have their own data entry interface to maintain information distinct from personal records.

Attach people, images and documents to events, places and addresses – For example, if I create the HMS Charlotte as a place or address, I should be able to add all the passengers. Or, if I create a church, I should be able to list the people married there.

Extra ‘relationship’ types – Need to define additional options for ‘relationships’. For example, ‘Friend’, ‘Business Partner’, ‘Group association’ (e.g. Freemasons).

Filter individuals by new categories – For example, ‘Relationship type’, ‘Lived at this address’, ‘Buried in this cemetery’, etc.

Geolocations of places and address – This functionality is integrated into HuMo-Gen, however I couldn’t get it to work properly.

Global user access – Currently users need to login separately for the website, the admin interface and the forum. There should be a single sign-on to access the entire site.

Fix Search functions – Some of the search functionality is a bit inconsistent. This probably needs a complete overhaul to ensure records are found consistently.

TESTING

  1. Go to: www.connections-hub.com
  2. Login with:
    Username = Test
    Password = test123

  3. Select Family tree
    Click on Photobook
    …Browse the images

  4. Select Family tree
    Click on Family tree index

  5. In the Search box on the right of the screen, type in: Povey
    Click Search

  6. Click on Capper, Grace Povey
    …My Great-aunt, Grace Povey Seccombe was a famous ceramist (trove.nla.gov.au/people/1486986?c=people)

  7. In the Search box on the top right corner of the screen, type in: Mehetabel
    Click Search

  8. Click on Crago, Mehetabel
  9. Hover over the report icon to the left of Mehetabel Crago
    Select Descendant chart from the popup menu

  10. Change Generations to 5
    …Ta-da! There I am at the bottom right of the screen: Lisa Beth Spencer

  11. Select Tools
    Click on Relationship calculator

  12. Individual 1 – First name, type in: Lisa
    Click Search
    Select Spencer, Lisa Beth (1969) from the search results

  13. Individual 2 – First name, type in: Jane
    Click search
    Select Poole, Jane (1768) from the search results

  14. Click Calculate relationships
    …Ta-da! I’m related to a first fleet convict. I knew there was a reason I had a tendency for larceny…

One last post

Access to data has been central to our journey throughout this unit…

One of the things I love best about my job is the contribution that I make to the NMA’s collections online – Collection Explorer. At any one time, about only 5% of the museum’s collection is on display. Collection Explorer gives access to 50% of the museum’s collection 24 hours a day. On the surface, Collection Explorer seems effortless, but the records and images displayed have taken years of development and are undergoing continuous improvement. In some cases, we can provide better and more in depth access to objects on line than in the gallery. In the gallery you can view the exterior of a car, online you can view the engine and the interiors. In gallery you can view one page of a sketch book, online you can read from beginning to end.

I am all about possibility and access. I believe in the power of collections to help people connect with past present and future, to inspire and promote discovery. This power is lost when collections are locked away in storage. At present, Collection Explorer does not have an API, but perhaps it should. I feel I should think that an API is a great idea because it fits with the ideology that I have just outlined. I love the idea that an API might give access that inspires new creation, new ways of looking at the collection, new discovery, new connection and for aboriginal communities re- connection and engagement. But I worry about misappropriation. The NMA releases records to Collection Explorer in an effort to provide access to the collection, but the majority of the content is not copyright cleared, nor have moral rights been applied. In good faith, we are leaving it up to our online visitors to decide what is appropriate in terms of use – but are they suitably qualified or informed enough to make these kinds of decisions. When DJ’s sample music, they can use a mere 8 seconds before royalties are due. What is the equivalent in data and images in terms of what is appropriate?

I also wonder about how we can capture new discoveries about the collection, how the museum can collaborate with those who engage with the collection. In an environment of dwindling resourcing it is essential that there is a two-way conversation about what is created from the data made available through an API. In relation to collections, the NMA is merely a custodian. We have no real desire to lock the collection away, but we do have responsibilities in terms of the collection and to ensure that museum operations are sustainable so we can continue into the future. I am left conflicted, but I am just one small cog – fortunately in this case, these decisions are bigger than me!

Week 12

Amazing! Today’s workshop was my very last class as an undergraduate… I hope!
It is interesting to consider that my first lesson was a classic anthropology lecture delivered by the legendary Brian Egloff. Things have really changed. In 2006, most museums had a website but engaging audiences digitally was a low priority. Today digital engagement is increasingly an essential way for cultural institutions to do business, so it seems fitting that my last undergrad class is Exploring Digital Heritage… Thanks Tim!

To begin with today, it was fun to seemingly hack ASIO – the black car that has been tailing me since is a bit scary! But all good.

I was surprised to hear that the redactions that Tim has been working on were created, not by an ASIO agent lurking in a dark alley, but at the Archives in a locked room by someone with a Top Secret security clearance. I am endlessly surprised by the variety of jobs in the GLAM sector.

The Head of the Library of Congress is a lifetime appointment – now that is amazing job security… The Thomas Padilla section of the library of congress’s symposium, “Collections as Data”, provides a great overview of the imperative to digitise collections, balancing this with an outline of the related challenges. Padilla refers to the digitised version of an artefact or book as a surrogate, further he says that each surrogate brings its own opportunity to make connections or to seat the artefact in a community. This is a very succinct way to describe the benefits of digitisation. I also identified with one of the concerns highlighted by Padilla – that viewing collections as data risks reducing lived experience to the equivalent of butterflies pinned in a cabinet of curiosities.

Allowing access to heritage collections online may appear a no brainer, but it is met with organisations grappling with moral rights and copyright. Is it really ok to say to the public, to quote Padilla, “Right assessment is your responsibility”.

Week 11

This was an interesting lesson. I am not sure why, but I am left with more questions than answers…

DIY web archiving

I really like the idea of playing arcade games from the 70’s and 80’s – so simple and quirky. As a teenager, access to arcade games was limited because it involved getting 20 cents out of our parents to waste in a machine! I was also so bad at playing, that the whole experience lasted about two minutes. These games and the movies are great examples of why we collect stuff. To evoke memories… and …wow 12 year old me is jumping for joy (I can play these games as much as I like for free!).

Here is what I do not understand …. Some of THE METs http://www.metmuseum.org/ collection is also posted in this archive. Why is this the case? THE MET has a perfectly good collection search portal that is not going anywhere. In this archive, you can collate similar objects together such as embroidery samplers from a range of institutions. However, if these records are changed or improved at their primary source, this is not reflected on this site. In contrast, check out THE METS Pinterest pages, it’s based on a live link so you at least know you have the latest info on any one thing. Additionally, you can make your own boards that consist of material from a variety of institutions. I can see that archiving is useful when a museum, for example, closes – but why archive material from sites that are alive and kicking, growing and changing?

On one of the blogs provided as part of this lesson, I read that most web pages have a life of only 100 days. I can see the point of archiving some pages and material released to the web. However, there are mountains and mountains of content released every day – how do we decide what to keep? It sounds great to keep everything, but we have already learnt that it is difficult to see the forest for the trees, or through the filter, as the case may be. Everything that is kept has a cost, even if it is digital – storage and back-ups need someone to manage the material. There needs to be a selection process. We need to be able to identify material that is unique and will disappear without attention and keep enough to paint a full picture but no more. Even Noah saved only a male and a female of every animal type, not five of everything…

Week 10

Week 10 – AR VR and 3D

This was a fun lesson…

I enjoyed exploring 123 catch. The 3d model of the Tardis was good as long as you do not look at the back. I can imagine playing with my kids and getting a bit obsessed with trying to render perfect models. I also think that this could be a great Public program at the museum. Kids could select something from the education handling collection and go about creating a 3D likeness… Speaking of kids what about free 3D colouring… http://www.quivervision.com/

Tim, like you I wonder about the application of AR and VR for Museums. The NMA has developed an AR trail relating to Kspace. You download an App, then look for markers throughout the museum, and are rewarded with a Kspace character when you aim your device at the marker. You can also download the markers and use the app at home – though, the site does not say much about using them at home. But in thinking about it now, maybe a game of hide and seek or a scavenger hunt would be cool! On the flip side, I’m not sure how this contributes to the museum’s mission. http://www.nma.gov.au/kspace/kspace_app.

I too find most of the cultural institutions on Google Institute a bit disembodied and lifeless. It is kind of fun that you can visit lots of institutions at any time of the day, in any part of the world, but what do we get out of this? Here is the Ted talk by Google’s Amit Sood, explaining the development and thinking around Google institute; https://www.ted.com/talks/amit_sood_building_a_museum_of_museums_on_the_web?language=en

While I am not that keen on visiting the cultural institutions via Google Institute, I really do think that being able to explore historic sites it useful. For sites such as Pompeii, or the pyramids of Giza, the technology used by Google institute allows detailed recording. Additionally, everyone can visit fragile sites without causing further degradation. Opportunity to visit such sites is a bit limited at present, I would like to see more. It would be great if these kind of sites were able to give google more access, allowing visitors to go behind the scenes and explore in more detail.
Lastly, I am not that keen on the idea of 3D printing artefacts, except in terms of giving access to people with visual impairment. Touching replicas and examples of materials from which the original object was made can aid in accurately imagining an artefact.

Perhaps for cultural institutions it’s all about horses for courses… Looking at where this kind of technology is appropriate or useful to adding extra layers of interesting interpretation.

Week 9

Redaction Art – I really want to know if the requirement to redact certain information created these blobs or if the emergence of a creature or object shaped the redaction (a chicken or egg question!). These images reminded me of the Rorschach test used by psychologists. I got side tracked trying to find out what inkblot tests are called and found this weird online inkblot test. http://theinkblot.com/ (do not be tempted. Once you start, you will have to finish! )

It was entertaining playing around with the Vintage Face Depot. I posted the following picture of my daughter. The face applied looks like an antique mask but check out the spooky little face on her arm.

elli-rabbit

It appears that Facebook, Google, IBM and probably many other companies are moving forward the development of facial recognition software at an alarming rate. It will not be long until concepts explored in the television series “Person of Interest” https://en.wikipedia.org/wiki/Person_of_Interest_(TV_series), are a reality. In this show, people are tracked using facial recognition via street surveillance cameras. This is a rather scary prospect, following the ideas presented in this show using cameras in public places and no doubt images included in social media our movements could be tracked most of the time. It is interesting to consider that France has just banned facial coverings https://en.wikipedia.org/wiki/French_ban_on_face_covering . This article talks about how as a society we rely on facial recognition to identify people. You have to wonder if some governments are already using computer facial recognition to stalk us.
As facial recognition software improves, it could be used to identify new imagery of historic people. For institutions with limited funding, it could be applied to previously undescribed images. However in historical terms there might be limited application, given that photography has a been widely available for only about the last 130 years. Perhaps, in time it will be possible to identify people in creative works.

As discussed in class, when the National Gallery of Australia was looking for a Digital Asset Management System (DAMS) to manage their images and other media assets, the tender documents describing requirements included facial recognition. As I understand it, they hoped that their DAMS would be able to automatically catalogue images of people via facial recognition, thereby reducing data entry. They have since implemented a system known as Diva (this might be an Oracle product). I am not sure if their notion of facial recognition has been successfully implemented.

See you later,
Lisa.

Week 7

I am probably the most incapable map-reader in the whole world (incorporating a Tom-Tom into the family probably saved my marriage!), so I was surprised by how much I enjoyed playing with maps this week.
Beginning with Google My Maps, I wanted to mark all kinds of spots. I can see how you could tell stories through plotting. It would be fun to mark-up family holidays and all kinds of stuff! Correcting the mistakes in the Australian National Heritage List so they would plot correctly became a bit of a google quest. I was able to map almost everything in the end by either replacing the ext with another detail or by removing some detail.
For example;

• Coranderrk – Change La to lane
• Koonlda Cave – Remove Cook
• Kingston and Arthurs Vale historic area – change ext to TAS

Moving on, I really like the way the Orbis map illustrates a story. It is a bit like a choose your own adventure book from the 80’s. Yes, I know many of you would be too young to remember those…

Many of the maps included in the Carto Gallery were very cool. My favourite was http://mortuarymapping.matrix.msu.edu/index.html. This site records the excavation of The Santa Clara Valley Medical Center (SCVMC) Historic Cemetery, providing a number of different filters. It includes what they have called a button map, which indicates people who were buried in hospital gowns. Burials took place in this cemetery in the early 20th century those buried in gowns were identified according to button patterns as the glass buttons survived. Very cool!

I also really enjoyed playing around with the Carto map populated with cycling accidents (not a very happy subject). Placing this data on the map really helps to make sense of information and makes it much more compelling than just looking at it in a black and white table. Using the variety of filters transformed the data in a different way

I am interested in trying to incorporate carto into the “Connections” web site that I am building. The limitation of 100 points per month could cause issues, but it is very useable. The other thing that I liked about the SCVMC Historic Cemetery site is the hand drawn burial map. Originally, I had thought that I might geolocate each burial, but it might be cleaner and more accurate just to geolocate the cemetery itself then provide a plan or map of the cemetery.

That’s all from me this session…

My Brain Hurts

I had to read Humanities Data: a necessary contradiction, by Miriam Posner a couple of times before I really understood her point – which probably makes me a bit thick. The title, of course, says it all – to boil humanities research data down seems at odds with the origin of that data. The analogy Posner uses is actually particularly poignant.

“Imagine that someone called your family photograph album a dataset. It’s not inaccurate per se, but it suggests that this person just fundamentally doesn’t understand why you value this artefact.”

Initially I felt that Posner was at odds with my view that preserving the context of data is essential, but in fact this article supports my view.

This article also has a link to the work of Trevor Munoz and Katie Rawson – Digital Heritage Curation http://www.dhcuration.org/. This is a fantastic site that includes numerous links to scholarly articles exploring issues and concepts relating to curating digital collections. It is well worth a look. You can follow @DHCuration on twitter.

Open Refine is a great tool… I am keen to test this program on some data exported from Emu – the National Museum of Australia’s (NMA) collection database. Like many institutions, the NMA is keen to publish as many collection records on the web as soon as practical. A barrier to publishing collection records to the web can be the quality of the data.

Till next week..

Lisa

Week 3 – Bits and Pieces

Hello.
I have to say, I like to tweet. I am naturally verbose, but I like the challenge of highlighting something or making a point in a limited number of words. Blogging feels strange, like talking to no-one or to myself…
Aiden, thanks for pointing out Mitchell Whitelaw’s Ted talk. I was lucky enough to attend a symposium at Canberra University a couple of years ago where Whitelaw spoke about “Generous Interfaces” vs empty search boxes. It transformed my thinking. Following the symposium, I was tasked with overseeing a project at the NMA that resulted in Collection Explorer (http://collectionsearch.nma.gov.au/) developed by a Canberra based web company Oxide Interactive (http://www.oxideinteractive.com.au).
I believe Collection Explorer includes many of Whitelaw’s concepts – inviting users in object type, place and even by material. Like Trove, Collection Explorer exemplifies a platform which continues to be a work in progress as NMA collections continue to be digitised. Also new ways to interact and traverse the collection are developed as time and money allows.
In terms of other great generous heritage interface my favourite continues to be the Rijksmuseum (https://www.rijksmuseum.nl/en/rijksstudio). Not only does this site entice you to explore, but much of the material offered is out of copyright and reuse is encouraged!
Regarding Google Algorithms and Filter Bubbles – it’s frightening to consider how our thoughts, actions and lives are being unconsciously directed and shaped without our consent. Masses of Information is added to the web on a daily basis, and with this ever increasing volume it’s difficult to see how in the future we’ll be able to navigate through the jungle/web to acquire a balanced diet of “vegetables and Desert”!

About me….

As a busy Mum of three it’s not very often about me!

You might be wondering about my username? It’s probably not what you think? For those not yet around in the 80’s, “Summer of Sixty Nine” https://www.youtube.com/watch?v=9f06QZCVUHg (sorry cut and paste :() …is a fab song by Bryan Adams. This is one good reason to have it as my username. The other is that it reflects my vintage. I am, only just, a child of the 60’s (a flower child). I was born the year man landed on the moon… a year that was a significant step forward for me and for mankind.

Many of you will already know that I work at the NMA sounds boring but I look after the Museum’s collection databases EMu, Piction and Collection Explorer.Perhaps in this company that’s not so boring? I love the endless possibilities that collections in digital form offer and while I have worked around data for a long time I can’t code or do more technical stuff myself …yet… that’s why I am here. I hope to learn new and meaningful ways to share my love of all things heritage.

Till next time
Lisa