I think a great way to write machinelearningapi and cre in 6.x, is to leverage cck's nodereference + userreference, actions + rules, and flag. With rules, you can flag content on behalf of a user. You can also auto-fill nodereference and userreference fields.

So, users can have content that matches patterns automatically flagged into their viewing spaces, and they can choose to unflag it. If they browse for data, then a user could flag it into their personal db tables. Separate standards for content can trigger different sets of actions on nodes on behalf of individuals.

The benefit also is in cck's content_copy module: you can import node_types and fields just by installing a module. The same is true to a similar degree for the rest of the combo as well.

Comments

kyle_mathews’s picture

kyle_mathews commented
Status: Active » Postponed

Some cool ideas there. I definitely see potential in integrating with actions/rules/flag. Those are powerful modules. Right now I'm just trying to get the core functionality of memetracker working correctly. After that, I want to add some hooks so 3rd party modules can add integration like you're suggesting. I think there's a good chance that actions will be build into memetracker in the future.

mitchell’s picture

mitchell commented
Title: Wild idea » Discussion of Actions, Rules, + Flag Integration
Status: Postponed » Active

Kyle, if you don't mind, I'd much rather keep this issue active; hopefully, more contributors and commentators will look at your modules and add to this issue.

Could you please explain the existing mechanism by which individual users currently accumulate and store their memes?

kyle_mathews’s picture

kyle_mathews commented

Fair enough.

I'm planning on writing documentation (yeah!) over the thanksgiving break. I'll try to answer your question there.

mitchell’s picture

mitchell commented

:-)