Black Lives Matter. Please consider donating to Black Girls Code today.

Plotly, Open Science and Reproducibility

Hi Plotly Team,

Currently I am working in science. I am a big fan of Plotly and use it both in my research and when I teach methods (e.g., data visualization).

However, I am running into problems. I am a proponent and also teach methods of open science and reproducible research (e.g. here). Reproducibility means that an analysis can be reproduced by another researcher from start to finish relying on open-source software (e.g. a researcher in Africa might not have the money to pay for a software license). That’s why people in academia move from commercial software such as STATA to R or Python.

The issue with Plotly is that it doesn’t seem to be possible to produce publication-quality (high resolution) static graphs for free (e.g., I just tried orca() but it requires an API Key). So when I teach Plotly to students they would need a key and an account as I understand.

I understand that Plotly needs to make money somewhere. But I wonder what Plotly’s long-term strategy is in that matter.

Thanks!

Paul

See https://github.com/plotly/orca

1 Like

Ok! Basically, I work in R. I got the following message when trying to use orca().

Error: No mapbox access token found. Obtain a token here
https://www.mapbox.com/help/create-api-access-token/
Once you have a token, assign it to an environment variable
named ‘MAPBOX_TOKEN’, for example,
Sys.setenv(‘MAPBOX_TOKEN’ = ‘secret token’)

After your answer I simply tried entering a random number as token and it worked. That solves it!