Generate pages for trackers with insufficient data (instead of 404)

[philipp-classen]

To derive meaningful data, WhoTracks.me only generates pages for the trackers with the most traffic. For other remaining trackers, we should still generate some nicer page then having a 404 (Page not found). For instance, having a fallback page that explains that it is a potential tracker, but not enough data is available to confirm.

[ghostwords]

Does analytics.tiktok.com fall into this bucket?

[philipp-classen]

As a first step, I would expect it to generate pages for everything listed in the the input file trackerdb.sql. Currently, I don't see references to analytics.tiktok.com there, so it would not be covered. I wonder if analytics.tiktok.com is exclusively used on the tiktok domains (that would explain why it is not automatically picked up, as the system distinguish between first-party and third-party tracking).

For reference, this it where the file can be found, though its content might change over time:

aws s3 cp --no-sign-request s3://data.whotracks.me/trackerdb.sql .
grep analytics.tiktok.com trackerdb.sql # no results

[ghostwords]

I wonder if analytics.tiktok.com is exclusively used on the tiktok domains

Good question, it's not though. See https://ads.tiktok.com/help/article?aid=9663 and https://publicwww.com/websites/%22analytics.tiktok.com%22/ (for instance).

[ghostwords]

You should have started picking it up in the summer of 2020.

[philipp-classen]

Thank you, I also found it in our raw data listed as third-party requests. Would need to investigate why it is not classified as a tracker. I created a separate issue for it: #262

[ghostwords]

Excellent, thank you!