The team found themselves at the center of controversy after winning first place
The winning team of Concordia’s ConUHacks event came under fire last weekend after accusations of violating competition rules were posted online.
ConUHacks, the event which the project was produced for, is a yearly competition during which teams must conceptualize and code an application in only 24 hours to impress judges and potentially win prizes. This “hackathon” is primarily organized by Concordia’s HackConcordia club.
NearByNow, the winning submission, was accused of being presented to judges under false pretence and omitting key lines of code to prove its authenticity. This led to an investigation into the validity of the winners’ application.
Although the accusations were proven false, the winning coders are now worried of online harassment.
“I was confused. We all were,” said Samuel Chuang, a fourth-year computer science student at Concordia and one member of the winning team. “Honestly, I had to think to myself, maybe we did do something wrong.”
Chuang said everything went well during the competition, leaving the judges impressed. Their hard work was rewarded with first place.
Six days later, Chuang found out about the controversy forming around their application.
A post on Concordia’s official Reddit page that called the team’s winning project fraudulent garnered the attention of many who had participated at the event.
Soon, Chuang said he’d seen a post by an anonymous user on Concordia’s official engineering and computer science Discord server, suggesting users should flood the team’s LinkedIn pages with negative comments.
“Hate on the project as much as you want but there’s a potential of ruining people’s reputation,”said Chuang.
Chuang and his team contacted Major League Hacking (MLH), who help run weekend-long hackathons like ConUHacks and partly oversee judgment.
“We found what we would expect to see from a hackathon project built in just 24 hours,” said Ryan Swift, a member of the MLH who reviewed the team’s project.
According to Swift, Chuang’s team was accused of faking the demonstration of their project. More specifically, by faking results given by the programmed A.I., which ran the application. They were also suspected of omitting the code which the A.I. had been programmed through- what’s known as a “neural network.”
NearByNow shows users information about a storefront or company in real time once given a logo. This feature relies on multiple application programming interfaces (API), for example one from Google Maps. The A.I. then communicates with the API to produce the desired results.
The neural network was not made public, which added to the accusers’ suspicions. Swift said his team verified its existence as well as its timestamps to confirm that it was coded during the hackathon.
The team would also “hard-code” data given by the application. Hard-coding data means that the results given by the program are directly put into the code rather than obtained by prompts. According to Swift, this was done for simplicity’s sake and the team had done nothing to break the competition’s rules.
“Because they are developed in just a single weekend, hackers don’t typically follow industry-best practices,” said Swift. “Their projects are often laden with bugs, and many features aren’t fully completed.”
Vatsa Shah, co-president of HackConcordia, the club responsible for organizing ConUHacks, said his team did not appreciate the public accusations of malicious intent towards the team. “Our team is always willing to investigate and review issues as they arise, but we prefer to do so in private specifically because of situations like this where public backlash can take over,” he said.
An official comment was written by the MLH under the original accusatory Reddit post stating that the investigation had been completed.
