Determine 90th percentile and 99th percentile average on both mCaptcha/mCaptcha instances and mCaptcha/survey. #6

Closed
opened 2023-02-13 15:14:14 +05:30 by realaravinth · 3 comments
Owner

This milestone lays the foundation for "Milestone:
Use local instance statistics to guide webmasters when configuring PoW”. The work from this PR will be used to improve the “easy mode” for installing mCaptcha0

We currently make guesstimate difficulty factors in easy mode, but with this work, we could ask questions like: "How long do you want your visitors to wait before entering site and what percentile of devices do you want to cover?"

If successful, computing percentile scores will enable us to migrate from difficulty factor based configurations, which is very unwieldy, to wait time in seconds.

This milestone lays the foundation for "Milestone: Use local instance statistics to guide webmasters when configuring PoW”. The work from this PR will be used to improve the “easy mode” for installing mCaptcha[0] [0]: https://mcaptcha.org/docs/webmasters/configuring-difficulty-factor#easy-option We currently make guesstimate difficulty factors in easy mode, but with this work, we could ask questions like: "How long do you want your visitors to wait before entering site and what percentile of devices do you want to cover?" If successful, computing percentile scores will enable us to migrate from difficulty factor based configurations, which is very unwieldy, to wait time in seconds.
realaravinth added this to the 2023 project 2023-02-13 15:14:14 +05:30
Author
Owner

mCaptcha/survey

Benchmarks with duration less than or equal to the requested time is retrieved and sorted in ascending order. Percentile score is computed over it. This functionality is available via an API endpoint.

PR: mCaptcha/survey#20

## mCaptcha/survey Benchmarks with `duration` less than or equal to the requested time is retrieved and sorted in ascending order. Percentile score is computed over it. This functionality is available via an API endpoint. PR: https://git.batsense.net/mCaptcha/survey/pulls/20
Author
Owner

Tests in the previous PR were failing due to unrelated reasons:

Creating a separate database for each test is provided out-of-the box in “batteries included” web frameworks. I was using hacks to get test isolation, and it finally broke for this test. This PR creates separate databases for all tests. So tests are more stable now.

Fixed in follow-up PR: mCaptcha/survey#21

Tests in the previous PR were failing due to unrelated reasons: Creating a separate database for each test is provided out-of-the box in “batteries included” web frameworks. I was using hacks to get test isolation, and it finally broke for this test. This PR creates separate databases for all tests. So tests are more stable now. Fixed in follow-up PR: https://git.batsense.net/mCaptcha/survey/pulls/21
Author
Owner

mCaptcha/mCaptcha

PoW analytics records from all mCaptcha configurations on the instance with duration less than or equal to the requested time is retrieved and sorted in ascending order. Percentile score is computed over it. This functionality is available via an API endpoint.

PR: https://github.com/mCaptcha/mCaptcha/pull/121

## mCaptcha/mCaptcha PoW analytics records from all mCaptcha configurations on the instance with `duration` less than or equal to the requested time is retrieved and sorted in ascending order. Percentile score is computed over it. This functionality is available via an API endpoint. PR: https://github.com/mCaptcha/mCaptcha/pull/121
Sign in to join this conversation.
No Milestone
No project
No Assignees
1 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: mCaptcha/2023-NLnet#6
No description provided.