• View
• Edit
• History
• Print

back to SeedeaCore

Seedea API (all those function are available through remote calls)

User

	CreateProfile
	DetectSimiliarProfile
	UpdateProfile
	ListIdeas (list ideas where Ive made edit)
	ListMicroservices (list microservices with a notification mode)
	SubscribeToMicroservice (pick default notification, optionnaly choose others plus otions)
	UpdateMicroserviceOptions (notification/timeslice/integration)
	UnsubscribeToMicroservice (no notification)

Microservice

	Register (add a new microservice to the list of subscribable services)
	Enqueue (call a service from a page for an user)

Idea

	AddToSandbox (append well structured idea in Username.SandIdeaBox )
	TransformToOimp
		separate Username.SandIdeaBox
		extract the targetted idea
		write it in Oimp.Uniqideaname
		replace the original in Username.SandIdeaBox text with a link to Oimp.Uniqideaname
	TransformToPimp
		same but within Pimp.uniqideaname
	TransformToLab
		extract from pagename the synthesis and append it to Seedea.Thelab
	SplitTo
	MergeWith
	FindAuthor(s?)
	Display
	Edit

authorinfluencers(user U, page P)

	read configuration Profile/U
		mode of notification (if none means unsubscribed)
			profile page (profile page, email, RSS, Twitter, etc)
		time slice 
			CET 2am-6am (default on geoIP night then adjusted from stats)
		mode of integration
			direct append / direct prepend / replace tag / overlay (with buttons to append/prepend)
	if not (done today!?  && out of time slice)
		update Oimp-Microservices.X "cloud currently working with stats", stats
		run OpenCalais API call with Oimp.X
		find authors then call the other API (or directly use Freebase scripts or directly DBpedia)
		if no error
			integrate in Oimp.X
			remove from Microservices.Queue
		notify in profile page
		update Microservices.Stats-Cloud user, date, time required
		update Oimp-Microservices.X remove cloud
	else
		if not present in Microservices.Queue
			queue itself

Separation

Used in

• Seedea::TransformToOIMP
• Seedea::TransformToPIMP

Principle

Detect repetition of (non-text items)* and order them by (length of separated content)*

	(X)* are local heuristics

Example

the study of this text would (should?) find

1. \n----\n
2. \n\n
3. \n
4. ,
5. " "

Simulation

try with explode() and let the user decide

What is pseudo-code? What is code? Is it just discussion?

Avoid recursively head spinning abstraction while designing, get simple things going THEN and ONLY THEN abstract/template/etc them!