Skip to content

Discoverability Project

An open data initiative to increase the online visibility of arts events.

Project Report

Introduction

“the characteristic of being able to be found or discovered, (specifically) on the internet”

The “Centre For Arts Tape – Discoverability Project” began in 2019 as a response to our perception that the evolving technological landscape of artist-run-centres in Canada was changing. We noticed that information flow and dissemination was no longer primarily supported through conventional website technologies. The new world order was replete with new mobile initiatives and applications. At that moment, we responded by envisioning the development of a new process or application that would help us:

communicate with existing supporters

establish new and wider audience bases

foster collaboration between arts partners

allow art centres to embrace their digital futures

With the support of the Canada Council Digital Strategy Fund, we partnered with The Film and Video Arts Society of Alberta (FAVA) and embarked on this ambitious multi-phase project. An important aspect of the project was an extensive consultation prior to the development phase. The purpose of this was to ensure the project proceeded from a perspective of sector-wide needs, with a strong vision for knowledge-based planning. We envisioned four phases of exploration:

Phase 1

Research and Evaluation

The CFAT Discoverability Project began with a research component that was designed to take the pulse of the Canadian arts community in terms of creating a public-facing mobile application that would improve or increase our organizational ability to reach out to our respective communities with event and activity data. 

Led by CFAT coordinator Patricia Boushel, the Phase 1 enquiry helped us to learn how organisations use technology for administration, marketing, and artistic practices. Information collected through this process guided our development in the second phase of our project. Data collected here was also intended as a resource for other stakeholders in the arts and culture sector wishing to engage in their own projects or initiatives.

In March of 2019, as part of a national review, we canvassed 94 individual agencies in 8 provinces. Most of these agencies were artist-run centres, or media arts centres as well a number of other arts service organisations. Participants from both urban and rural areas were selected, and there was representation from all five Canada Council administrative regions:

  • Eastern Canada (NB/NS/PE/NL)
  • Quebec & Ontario
  • Central Canada
  • Western Canada (AB/BC)
  • Northern Canada (NU, YT, NT)

This narrative consultation was used to determine the art sector’s interest in supporting sector-wide technological development and to generate a clearer understanding of:

  • the past technological engagements of other arts organisations,
  • previous projects of similar nature/scope that had been attempted, 
  • the level of success achieved by such initiatives and the circumstances that contributed to these outcomes. 

The survey aimed to take an intensive look at the structure of artist-driven organisations across the country, including their communication needs, funding situations, annual budgets, staffing, volunteer, and technological resources.

Results

The main objectives and aspirations articulated by the study group were as follows:

  • A strong desire to reach new audiences;
  • The ability to market in new digital formats;
  • To be listed in a public-facing platform alongside other, possibly better-known Canadian independent institutions;
  • The ability of the project to encompass institutions that are not officially recognized and/or that are farther afield, geographically;
  • A way to quickly and easily access other centres’ programming, events, and activities.

Some centres were skeptical of the project’s intention to consolidate Canada-wide programming data into one app. The difficulty expressed was that it was already hard to “Stand Out” in a sea of potential cultural offerings. A mobile app that addressed this issue successfully would have to do what no current technology had done so far.

We concluded that there was a cautious interest in the project from organisations across Canada. And while most respondents were excited about the App’s potential, they also expressed concerns about the attendant “Management and Infrastructure Costs”. 

Phase 2

Development of Event Application Solutions

With the survey’s background of information and a wish list of “Desires and Wants,” we continued our exploration of “Discoverability” by enlisting the resources of The Film and Video Arts Society of Alberta and their powerful arts management application “AMS-Network”.

When CFAT began this journey to make event data for our community more widely available on the web, we understood the problem in decidedly Web 2.0 terms and syntax. “Could we design a device or a program that would drive potential audiences to our events in more effective ways than we had previously?” We explored:

• The Event App

We researched developing an event app that would let users know about events happening around them and inform them about performances or offerings that they might attend. The problem with this approach was that you were preaching to the converted. If this individual is connected well enough to the art scene to download an app that tells them about upcoming events, then they are likely the kind of person who doesn’t really need the heads-up. And furthermore, how do you get uptake for the App in a larger public context? We discovered that this could be very difficult.

• Geo Fencing Technology

The idea of this App technology is that it uses the proximal data of your phone to that of a participating organisation’s physical location to give you a heads-up about art activities you might wish to participate in. The problem with Geo fencing is actually two problems. 

  • How do you get participating organisations involved in managing their Geo presence? The web transmission side of event monitoring is complex and requires endless care and feeding. 
  • And again, how do you create user uptake in the general populous? This just turns out to be another, but more complex version, of the Event App idea. People must buy into the idea and download technology before the system can be realistically used. 

So, our problem remained the same. Our event ideas were not discoverable and required too much effort on the part of the potential audience user to be practical.

Knowledge Graphs and the Semantic Web

While we were immersed deeply in exploring the possibilities of event app technology and its limits, the world wide web was changing. And here, we apologize, but we must take a step sideways to discuss two new and uniquely interesting technological fields of development, “The “Semantic Web” and one of its many components, the “Knowledge Graph”.

First, let’s talk about the Semantic Web or Web 3.0 as it is sometimes called. The fundamental difference between Web 2.0 (the web we all know and love) and Web 3.0 (the current and future web) is that: Web 2.0 was managed by humans. Web 3.0 will be and is managed by artificial intelligence. And the way that AI communicates, whether it be in your computer or on mainframes around the world, is by using its own semantic language constructions. Thus, the adjective “semantic”. In the future, data discoverability will be based on preparing data so that it can be communicated on the Semantic Web. Good to know. 

This leads us to the next ephemeral construction, “The Knowledge Graph”. A knowledge graph is composed of coherent stacks of indexed or assembled metadata that can be found by browsers and other Internet search engines. For example: “restaurants near me”. Unfortunately, at the moment, there are not many knowledge graphs on the web that specifically assemble shareable data about the arts community. 

So, that’s a lot to take in! 

But it is the epiphany at the heart of this project. These two ideas would cause us to shift our focus from the development of an event app to the development of technology that could assemble discoverable data on the Semantic Web. This would become our pursuit in Phase 3. ant “Management and Infrastructure Costs”. 

Phase 3

Revised Development Solutions

A Change In Understanding

Our findings in Phase 2 caused a complete review of our ideas around audience aggrégation. Instead of trying to push audiences to our websites, maybe we would be better served by pushing our event data to the web where it could be found “by Anyone” as open data. This idea is fundamental to the future development of information technology on the web. 

This is both an exciting and frightening idea, because, in this case, “Anyone” includes web-based artificial intelligence. Exciting because artificial intelligence helps us understand and port data to those most likely to respond to it. Frightening, for the same reason. The same invasive technology that sends you the clickbait links you follow down rabbit holes, could also be used to inform our public about the boundless artistic opportunities in the country. “We could use the Semantic Web for good, and not for evil”

Event Data Expressed As Open Data

The first problem is how to express an event using meta-data, that is sharable and presented with the correct syntax, so it can be discovered on the Semantic Web. Fortunately for us, a number of organisations were already working on this issue. Culture Creates had adopted and refined the proposed data sets created by Scheme.org, and they were using these sets with their Semantic Web aggregator Artsdata.ca. 

Although the syntax of these defining data points is weighted somewhat towards describing performance arts events, we found them entirely usable to describe most other arts events as well. The data set adopted by CFAT and our development partner FAVA, has twenty-one discoverable data points that define any given arts event. They include:

Name of Event

Type of Event

Description of Event

Photo of Event

Instructors/Facilitators etc.

Performers

Location of Event

Start Date

End Date

Price of Event

Member Price of Event

Early Bird Discount

Attendance Type

Max Participants

Status

Archived Status

Presenting Organisation

Partner or Contributor

Finder

Relevant URL

Target Audience

AWS Open Data Server

The second issue was to establish where events described with this data set would live on the internet. After considerable consultation and feedback, particularly around privacy and security, we took the approach to set up a separate AWS Neptune server that would host our open data. For these reasons, only selected and public data, such as the event data described above, would be pushed to this server. This would allow for a balance of control and privacy while allowing organisations to participate in the CFAT arts open data initiative. 

Our intention was to create a SparQL endpoint on an AWS Neptune server to receive and disseminate data. SparQL is the standard query language and protocol used for LOD (Linked Open Data) and RDF (Resource Description Framework) databases. More simply stated, the SparQL endpoint functions as a conduit for moving open data between participating arts organisations, the knowledge graph, and the event aggregator. After some experimentation, we were able to abandon the AWS Neptune server in favour of creating our own SparQL endpoint on Triply.cc, which was much more tailored to the specific requirements of our project.

Fortunately, because we had adopted the artsdata.ca and schema.org data syntax, our project data could be easily accessed and read by other aggregators looking for the same information. Many conversations were conducted around the future partnering and exploration of open data. We found this to be very exciting and full of opportunities. It was also very humbling when we looked at the scale and scope of the work required to integrate these ideas across multiple organisations, each with very different needs and resources. 

By the end of Phase 3, we understood how to construct the necessary flow of data from the AMS User to the Arts events Aggregator. It looked like this:

Phase 4

A Proof Of Concept

The last phase of our project was to create functional technology that could aggregate and present event open data, as a proof of concept. For this purpose, FAVA created and developed an aggregating instance where programming event data could be showcased and shared from any participating organisations in the AMS-Network. In addition, Artsevents.ca could be used as an example for any event aggregator or application interested in displaying event data from the arts community.

The beta test site for the project is hosted here on artsevents.ca.

The Process

In its simplest form, the process looks like this:

  1. Programmer creates an event or activity in AMS
  2. AMS simultaneously prepares data for discovery by the Semantic Web
  3. AMS pushes data to the SparQL endpoint 
  4. The SparQL endpoint makes this event data publicly available as open data
  5. Aggregator collects open data from the SparQL endpoint and displays data as Event Data on their websites.

Conceptually, if an organization manages its programmed events through AMS-Network, this private event data is automatically prepared for presentation on the Semantic Web as open data. Here it can be discovered by any service or aggregator that utilizes open data.

And because all of this activity is happening in the back end of AMS  it eliminated the laborious administrative burdens of manual data entry and updating for programming staff. Once the event is created there is no effort required to display it. And because this capacity is built right into the AMS-Network Application there are no additional costs. 

This eliminates two of the largest worries expressed by our initial survey group: 

  • No additional required staff time, 
  • No additional management costs. 

At the moment, any organization currently using AMS-network to manage their programming affairs can enable the discoverability option in their instance. With a click of a permissions button, their programming information can be made instantly available on artsevents.ca and instantly available to any open-source data aggregator displaying event data for the arts. 

Glossary

Below are some definitions of useful terms we came across during the exploration and development of the Discoverability Project.

Discoverability – the characteristic of being able to be found or discovered on the internet.

Index – the ability of certain software “such as footlight” to create connections between metadata and compile them in pages so that they can be discovered in a variety of ways.

Knowledge Graph – a coherent stacks of indexed or assembled metadata that can be found by browsers and other internet search engines. For example: “restaurants near me”. At the moment, there are few, if any, knowledge graphs on the web that specifically assemble shareable data about the arts community. Our partners, “Culture Creates,” are in the process of creating a knowledge graph for the arts.

MetaData – the smallest piece of data that describes or gives information about other data. For example, a creation date (metadata) might be the metadata for a photograph (the data).

RDF – Data Resource Description Framework

Triples – smallest string of metadata presentation, information presented in the form of (Subject – Predicate- Object Expression). For example, “Bob is 35”, or “Bob knows John”.

OWL – Ontological Web Language – Semantic Web Language

Shareable – The quality of metadata that can be found and shared on the internet. In this case, metadata with the proper syntax.

SparQL Endpoint – is the standard semantic query language and protocol used for LOD (Linked Open Data) and RDF (Resource Description Framework) databases. Simply stated, the SPARQL endpoint functions as a conduit for moving open data between participating arts organisations, the knowledge graph, and the event aggregator.

The Semantic Web 3.0 – The fundamental difference between Web 2.0 (the web we all know and love) and Web 3.0 (the future web) is that: Web 2.0 is managed by humans. Web 3.0 will be managed by artificial intelligence. AI will communicate using its own semantic or language constructions. Thus, the Semantic Web. The attribute of discoverability is based on preparing data so that it can be communicated on the Semantic Web.

Syntax – The form, order, or rules by which metadata must be coded, stored, or presented so that it can be shared on the internet

The Future of Discoverability

We have learned a lot, and we want to take the next step in harnessing the Semantic Web and its capacity to make arts events discoverable. Tell us how an arts events aggregator might help complement your programming outreach and audience acquisition aspirations.