Future directions

[apiology]

Hi! I wanted to have a place to dump ideas related to potential future directions for Records Mover that aren't really baked enough to be an 'issue' quite yet - maybe we can use this thread?

[apiology]

Generally one of the things I think would have a ton of value is having a cloud-hosted web interface.

Some benefits this would unlock:

• It'd get rid of the pain of installation and upgrading given the plethora of dependencies Records Mover uses to get the job done.
• In cases where we need to do some kind of CPU/IO intensive transformation of a CSV file, we could get it done much faster in parallel in the cloud.
• We could use (probably more mature) tools like https://github.com/yobulkdev/yobulkdev / https://flatfile.com / etc to do parts of the job of Records Mover like detection of the CSV variant and hints but don't do the job of translating formats and talking to different databases.

Different hosting needs pop up. It could be one big public thing, and/or include documentation for organizations to set this up more privately?

Public hosting in particular open some important issues to think through:

• Billing for CPU time could be an issue depending on usage level.
• Management of users' credentials, and the consequent need for authentication/authorization.a
• Careful handling of user data.
• Compliance needs of various sorts.