demo.tramline.app - Notes

[kolharsam]

Context

Currently to help potential customers try out tramline, a throwaway "viewer" account via the rake task create_external_viewer Link. Some data from the staging / production environment(s) is then anonymized and seeded for this new "viewer" account and given to the customer to try out.

Potential Issues with the current approach

There are multiple reasons to why this is not an ideal approach to help new and potential customers to try the platform. I'm listing some down below:

• Since this is done on the staging environment, this doesn't yield a very professional appearance to Tramline as we approach new prospects.
• The whole process of
  • creating a new viewer account
  • anonymizing data
  • seeding the newly anonymized data
    is pretty annoying and time consuming. Since we also risk the data becoming stale and unusable after a certain point
• Seeding data from either staging or production environments is invasive and expensive.

---

Hence having a dedicated demo will help alleviate these issues and also

• reduce the potential feedback loop while selling tramline to potential customers
• eliminate the need to anonymize data from either staging | production environments
• cutdown the potential sales time for Tramline
• keep the data on the demo fresh and easy to get most information and context out of trying the app

Steps involved to resolve this ticket

• Understand the data model of current version of tramline
• Figure out a way to easily anonymize data (without the need of existing data on production | staging)
• Create a new render environment for demo
• Disable certain actions (e.g: make a release) on the new demo site

---

Other notes

WIP

---

[kolharsam]

Important Links:

• https://github.com/sunitparekh/data-anonymization
• https://github.com/evilmartians/evil-seed
• It should be configurable similar to: https://github.com/simpledotorg/simple-server/blob/master/lib/seed/config.rb
• https://driverjs.com/

[kolharsam]

Other notes and links:

• https://dbdiagram.io/d/Tramline-DB-Schema-67bebc2b263d6cf9a07d5ccc
• tables that are affected by the anonymization task: https://dbdiagram.io/d/Most-important-tables-for-anonymization-of-data-Tramline-67bebfb0263d6cf9a07de025 (this is only around 30 tables out of the 90 tables or so that are present in the full schema)

[kitallis]

Old seed script that is much simpler: https://github.com/simpledotorg/simple-server/blob/a9caebff67d748ae1535097ca145744f862725de/app/jobs/seed_user_data_job.rb

seeder
-- model_seeder/
    - model_seeder.rb
    - model_seeder.toml
-- <model>_seeder.rb (builder config)
-- config.rb
-- top_level_config.toml

Strategies

• dev_starter
• mock_data

Sizing

• small
• medium
• large