Our project focused on Japan’s fixed internet speed. Japan used to be a frontrunner in the internet infrastructure. But, recently the connectivity has worsened dramatically because more and more people watch video streaming service on the internet. Japan Internet Provider Association frequently receives claims from customers and we heard that some people had difficulty to receive an email in the evening peak. We suspected that the congestion of the internet in Japan had reached a critical level.
The heavy traffic happens only in the evening between 8 pm to 11 pm and Japanese telecom giant NTT who owns broadband infrastructure hesitates to invest in telecom equipment. Due to the lack of investment, Japan’s internet speed has slowed down since 2015 and now is the 23rd among 36 OECD countries. Japan’s night time internet speed is as slow as some emerging economies like Russia.
By revealing the inconvenient truth, we encouraged NTT to invest more on the telecom infrastructure and called for the government to rethink telecom industry strategies.
What makes this project innovative?
This project revealed the bottleneck of Japanese telecom infrastructure. People have long believed that Japan's internet speed is among the fastest in the world. But, due to the lack of investment, Japan's broadband speed lags behind the world in fact. Also, we showed the heat map of hourly internet speed for selected countries. This might help readers to understand that the situation is critical especially in the evening.
What was the impact of your project? How did you measure it?
More than 100 thousands readers accessed our article within 3 days after publication and we received comments from our readers. Some of them think that our telecom infrastructure needs to be updated immediately to compete in the 5G era.
Source and methodology
We used open data from Measurement Lab led by Google, Princeton University and other academics. We collected hourly average internet download speed data for every month since 2010 and compared them among 199 countries.
Our project was mainly built in Python. We used Jupyter notebook for collecting data through APIs and data cleaning.
Mao Kawano (telecom reporter), Shohei Yasuda (designer), Kenichi Yamada (Seoul correspondent)