[Air-L] DiscoverText, TrustDefender, Twitter Data, and the Future

[Shulman, Stu]

The free academic text analytics and machine learning platform
DiscoverText, and the spin off TrustDefender, are now Canadian software
projects running on fast new hardware in a Vancouver, BC high-speed
co-location site. Software operations in the United States after 20 years
in universities, start-up incubators, the commercial cloud, and finally my
basement for the last 6 or so years, are officially over. Canadian
researchers with requirements to keep sensitive research data in Canada
please take note.

For 15 years, with the exception of about three months, we have had access
to Twitter data. That access (currently a paid subscription) ends this
month. If you want one last dataset of up 500,000 Tweets from the last 12
months, set up a meeting and I can get that for you. There are costs for
data access and labor to access the chunks of up to 20,000 Tweets per
download.

We have migrated about 1,000 projects with 300,000,000 curated Tweets to
preserve these idiosyncratic archival records for future research studies
by academics interested in the topics like the first three days of #metoo,
#BLM, COVID-19, and too many things related to American elections. If you
have any historical Twitter in JSON format, you can load it for
collaborative research purposes and see the (live) Tweets in our platform
using the authentic Twitter display.

I am excited to be returning to school to start a Masters program in
library science at UBC this fall. My goal is to focus on trust
in information systems. At the 2025 Annual Meeting of the American
Political Science Association, to be held in Vancouver this year, I will be
offering a 4-hour pre-conference workshop September 10, 2025 9:00 AM - 1:00
PM titled "Teaching and Researching Trust with Large Scale Twitter Datasets"

https://connect.apsanet.org/apsa2025/pre-conference-short-courses/

If you want to be a part of the workshop, I could use a few assistant
presenters in Vancouver to help me set up and execute what could be one of
the larger workshops I have had to date.

I am going to try and keep these tools free for academics as long as
possible. Some of the core annotation, measurement, and adjudication
features designed in 2007 are highly functional and efficient scientific
instruments.

https://tinyurl.com/DTCitations

~Stu


-- 
Dr. Stuart W. Shulman
Founder and CEO, Texifter
Editor Emeritus, *Journal of Information Technology & Politics*