Japanese survey - #273
Conversation
|
Thanks so much for doing this @AndrejKiri |
|
@AndrejKiri @E-STAT We didn't ask for the user consent to share the answer publicly. Given that, the raw data should not be publicly shared. |
|
@ymotongpoo Fair point. Thanks for raising it. In the past, we shared data for all surveys that SIG End-user conducted in our repo by default. I don't think we asked for consent for any of them. But that doesn't mean it wasn't a mistake. We should start doing that for sure. Here is an issue to make sure we do it in the future. For this particular case, I am wondering if there is anything we can do to still publish the data. 🤔 I think the data is pretty anonymous already. We can make potential identification of individual respondents even harder by breaking the data set into multiple .csv files—e.g., having demographic questions separate, open-ended questions separate, and the rest of the questions separate. That way there is IMO no way to identify who responded what. Would it be sufficient, or do you think we shouldn't publish the data anyway? @avillela @danielgblanco @reese-lee What do you think about it? Did we get similar feedback in the past? |
| ## Resources | ||
| <!-- Add links to the resources you created --> | ||
| - [CSV survey data](/end-user-surveys/japanese-community-survey/japanese-community-survey.csv) | ||
| - Survey report in `.pdf` |
There was a problem hiding this comment.
@danielgblanco Yes, the analysis is not finished yet. I would add it later on. My intention here was to share out the data so if folks in the community are curious, they can do their own analysis.
However, before merging, we should for sure address Yoshi's comment above. I shared my perspective there. I would love to hear what you think about it.
|
I'm not a lawyer, but does any of the questions (or combination of them) fall under the regulation for PII? Just reading https://www.security.org/identity-theft/what-is-pii/ it doesn't seem like there is. One thing we could do is ask for CNCF's opinion, as they have a service to run specific CNCF surveys (see https://contribute.cncf.io/resources/services/marketing#surveys) and could shed some light? |
|
Maybe in the future what we could do is a have a disclaimer around the information collected, how it's not PII, and how we use the data. |
|
Just an update here: We discussed this multiple times at the SIG meetings. I will split the data into several smaller csv and will open a new PR where I will follow the convention to submit PRs from a fork rather than submitting directly to the source repository. 💪 Will close this PR once the new PR is openned. |
4 participants
Adding a new survey to the repo. Please review. So far it is just a README file and csv with data. We will add analysis later, once it is ready.