Alexander Turnbull Library: Harvested Twitter data relating to Ihumātao

Twitter crawl conducted by staff at the Alexander Turnbull Library relating to the Ihumātao protest.

This dataset contains 54,868 tweets collected between 24 July - 19 August 2019 from the public Twitter API using Twarc.

The dataset contains harvested Tweets (Twitter JSON data) and associated image files. Also includes Tweet IDs, and a ReadMe text file that details the process to capture the Tweets and the tools used. Twarc retrieves content going back several days, so some tweets prior to July 24 will have been captured.

In 2014 Auckland City, using the Special Housing Areas Act, designated 32 hectares adjacent to the Ōtuataua Stonefields Historic Reserve as a special housing area. This land, known as Puketāpapa, was confiscated under the New Zealand Settlements Act in 1863. 151 years later, the land was sold to Fletcher Residential Limited for the purpose of building a low-density subdivision with 480 high-price dwellings on the unique heritage landscape. In 2016, Pania Newton along with five cousins and other supporters, formed the group Save Our Unique Landscape (SOUL) to protest the development of land at Ihumātao in south Auckland. Since November 2016, kaitiaki peacefully occupied the land whilst campaigning to #protectihumatao.

On Tuesday 23 July, New Zealand Police arrived at Ihumātao to issue eviction notices. As a result, thousands of kaitiaki/protectors came from across Aotearoa and the world to support the struggle to reclaim the whenua, both in person at Ihumātao, as well as via social media and online activism.

Quantity: 1 data set(s). 1885 digital image(s). 5 Electronic document(s).

Processing information: The data from the crawls was combined and deduplicated to reveal a total of 54,868 Tweets over the crawl period. Images in the dataset were harvested separately by the Library using a Python script. The Library also created CSV access copies from the JSON files.

The Library unshortened shortened URLs in the Tweets, and conducted a WARC crawl to captured HTML pages referred to in the Tweets.

JSON dataset requires computational methods for access (eg. Python and script editor). Access copies for some material are available as text and csv files.