% This file was converted to LaTeX by Writer2LaTeX ver. 1.0.2

% see http://writer2latex.sourceforge.net for more info

\documentclass{article}

\usepackage[utf8]{inputenc}

\usepackage[T1]{fontenc}

\usepackage[english]{babel}

\usepackage{amsmath}

\usepackage{amssymb,amsfonts,textcomp}

\usepackage{array}

\usepackage{hhline}

\usepackage[pdftex]{graphicx}

\author{till }

\date{2013-04-30}

\begin{document}

OOR / Ontohub API

\bigskip

\includegraphics[width=15.822cm,height=12.277cm]{OOROntohubAPI-img1.png}

Federation and general

General remark: Any ontology and any link can be optionally accompanied

by a version id.

Ontology ids are instance specific

\bigskip

Ontology Services

\bigskip

Not formalized in OORService

\ \ \ index OntologyVersion\ \ \ \ (ontology[id,vid])\ \ $\Rightarrow$

(ontology)

&find Ontology SymbolsAndSentences

&(ontology[id])&$\Rightarrow$ (list(symbol), list(sentence))

&find OntologyVersion SymbolsAndSentences

&(ontology[id,vid])&$\Rightarrow$ (list(symbol), list(sentence))

&get Ontology Prefix&(ontology[id])&$\Rightarrow$

(prefix)

&list Categories&()&$\Rightarrow$ (list

(category[id,name])

&find Category Ontologies&(category[id])&$\Rightarrow$

(list (ontology[id,name])

&list Groups&()&$\Rightarrow$ (list

(group[id,name])

&find Language Ontologies&(language[id])&$\Rightarrow$

(list (ontology[id,name])

\bigskip

&upload OntologyVersion&(ontology[id],file)&$\Rightarrow$

(vid)

download OntologyVersion&(ontology[id,vid])&$\Rightarrow$

(file)

\bigskip

\bigskip

\bigskip

Formalized in OORService

\# not to be implemented in our system

\bigskip

Ontology

&find Ontology&(name{}-fragment) &$\Rightarrow$

(list (ontology[id,name]))

&create Ontology&(ontology) &$\Rightarrow$

(ontology[id])

&index Ontology&(ontology[id]) &$\Rightarrow$

(ontology)

&update Ontology&(ontology) &$\Rightarrow$ ()

&delete Ontology&(ontology[id]) &$\Rightarrow$ ()

\bigskip

&get OntologyVersion

Metrics&(ontology[id,vid])&$\Rightarrow$ (metrics)

&update OntologyVersion

Metrics&(ontology[id,vid],metrics)&$\Rightarrow$ ()

\ extract OntologyVersion

Metrics&(ontology[id,vid])&$\Rightarrow$ (metrics)

\bigskip

&get OntologyVersion

File&(ontology[id,vid,sid])&$\Rightarrow$ (file)

&find Latest OntologyVersions

&find Latest ActiveOntologyVersions

\bigskip

&get AllNotes ForOnto&(ontology[id])&$\Rightarrow$

(list (note))

&get AllNotes ForOnto ByAuthor&(o[id],

author[id])&$\Rightarrow$ (list (note))

&get AllNotes ForConcept&(o[id],

concept[id])&$\Rightarrow$ (list (note))

&get AllNotes ForIndividual&(o[id],

indiv.[id])&$\Rightarrow$ (list (note))

&get AllNotes ForNote&(o[id],

note[id])&$\Rightarrow$ (list (note))

&create Note

&update Note

&archive Note

&delete Note

unarchive Note

&get Note Bean

&get RootNote

&archive Thread

unarchive Thread

\bigskip

Project

&create Project

\ retrieve Project

&update Project

&delete Project

\bigskip

Review

&create Review

\ retrieve Review

&update Review

&delete Review

&get Reviews ForOnto

\bigskip

Rating

&create Rating

&update Rating

&delete Rating

&get AllRatingTypes

\ retrieve RatingType

\bigskip

Finding Commands

&find OntologyOrView&$\Rightarrow$ find Ontology

&find LatestActiveOntologyVersions&$\Rightarrow$ find

LatestActiveOntologyVersions

&find LatestOntologyVersions&$\Rightarrow$ find

LatestOntologyVersions

\bigskip

cleanupOntologyCategory

getOntologyFile

Search

There is only one method (search), having the following parameters:

\begin{itemize}

\item search string (with Boolean operators and wildcards, e.g.

{\textquotedbl}foo bar -baz{\textquotedbl} will expand to

{\textquotedbl}foo* AND bar* AND NOT baz*{\textquotedbl})

\item

ontologyids={\textless}ontologyid{\textgreater},{\textless}ontologyid{\textgreater}…

- limits the search to specific ontologies (default: all ontologies)

\item searchontologynames=[1/0] – search in the ontology names (default:

1)

\item searchsymbolnames=[1/0] – search in the symbol names (default: 1)

\item isexactmatch=[1/0] – match the entire ontology resp. symbol name

(default: 0)

\item pagesize={\textless}pagesize{\textgreater} - the number of results

to display in a single request (default: all)

\item pagenum={\textless}pagenum{\textgreater} - the page number to

display (pages are calculated using {\textless}total

results{\textgreater}/{\textless}pagesize{\textgreater}) (default: 1)

\item maxnumhits={\textless}maxnumhits{\textgreater} - the maximum

number of top matching results to return (default: 1000)

\item symbolkinds={\textless}kind,kind,..{\textgreater} - limits the

results returned to these kinds, multitple kinds can be included in the

parameter.

\item includedefinitions=\{true\} - if a search result is a hit for a

symbol, adding this parameter will include the definition in the search

result xml.

\end{itemize}

\bigskip

Persistence

\begin{itemize}

\item Synchronize two repositories (also non-git ones, like triple

stores)

\end{itemize}

Difference

createDiff

createDiffForLatestActiveOntologyVersionPair

createDiffForAllActiveVersionsOfOntology

getAllDiffsForOntology

getDiffFileForOntologyVersions

Distributed Inference

Open question: should we use Hets development graph sessions, or only

send around updates to ontologies and links?

Here is the session based API:

\begin{itemize}

\item POST /libraries/{\textless}coded-iri{\textgreater}/sessions -

create a new proof session for development graph

\item GET

/sessions/{\textless}id{\textgreater}?format={\textless}f{\textgreater}

- get proof state of session

\item GET /menus - Get development graph menu structure

\item GET

/nodes/{\textless}coded-iri{\textgreater}?library={\textless}coded-iri{\textgreater}\&session=id

- Get info for node

\item GET

/nodes/{\textless}coded-iri{\textgreater}/theory?library={\textless}coded-iri{\textgreater}\&session=id

- Get theory of node

\item GET

/edges/{\textless}coded-iri{\textgreater}?library={\textless}coded-iri{\textgreater}\&session=id

- Get info for edge

\item PUT

/libraries/{\textless}coded-iri{\textgreater}/proofs/{\textless}id{\textgreater}/{\textless}command{\textgreater}

- execute command for session

\item PUT

/sessions/{\textless}id{\textgreater}/{\textless}command{\textgreater}?node={\textless}iri{\textgreater}\&edge={\textless}iri{\textgreater}-

execute command for node in session

\item GET

/sessions/{\textless}id{\textgreater}/provers?node={\textless}iri{\textgreater}\&translation={\textless}iri{\textgreater}

- Get provers for node

\item GET

/sessions/{\textless}id{\textgreater}/translations?node={\textless}iri{\textgreater}

- Get logic translations for node

\item PUT

/sessions/{\textless}id{\textgreater}/prove?node={\textless}iri{\textgreater}?prover={\textless}name{\textgreater}\&translation={\textless}iri{\textgreater}

\&timeout={\textless}secs{\textgreater}\&include=true - Call prover

\end{itemize}

List of available Hets commands (which ones do we need here?)

dg-all auto

&Apply

automatic tactic - needed

dg-all glob-decomp &Apply rule

global-decomposition - to start with, auto should suffice

dg-all global-subsume &Apply rule global-subsumption

- to start with, auto should suffice

dg-all loc-decomp &Apply rule

local-decomposition - to start with, auto should suffice

dg-all local-infer &Apply rule

local-inference - to start with, auto should suffice

dg-all comp &prove

composed edges - to start with, auto should suffice

dg-all comp-new &create composed

proven edges - to start with, auto should suffice

dg-all cons

&Apply rule

conservativity - to start with, auto should suffice

dg-all hide-thm &Apply rule

hide-theorem-shift - to start with, auto should suffice

dg-all thm-hide &Apply rule

theorem-hide-shift - to start with, auto should suffice

compute-colimit &compute colimit -

not needed, since this is called by static analysis of “combine”

compute-normal-form &Compute normal forms for nodes

with incoming hiding links \ {}- needed for proving in presence of

hiding

triangle-cons

&triangle-cons -

needed

freeness

&freeness

- not needed in DOL

flattening importing &Flatten all theories

and delete all importing links \ {}- needed for interfacing to standard

theorem provers

flattening disjoint-union &Create intersection nodes and

ensure only disjoint unions \ {}- needed for interfacing to some (but

not many) theorem provers

flattening renaming &Flatten out renaming

\ {}- needed for interfacing to some (but not many) theorem provers

flattening hiding &Delete all

hiding links \ {}- needed for interfacing to some (but not many)

theorem provers

flattening heterogeneity &Flatten out heterogeneity -

needed for interfacing to some (but not many) theorem provers

qualify-all-names &Qualify and

disambiguate all signature names

undo

&Undo

last change - not needed

redo

&Redo

last change - not needed

use

&{\textless}File{\textgreater}

&Read HetCASL file - not needed

dg basic

&{\textless}Nodes{\textgreater}

&Select node - needed

translate

&{\textless}Comorphism{\textgreater}

\ Choose translation - needed

prover

&{\textless}Prover{\textgreater}

&Choose prover - needed

set goals

&{\textless}Goal{\textgreater}

&Set goal - needed

prove

&Applies

selected prover to selected goals - needed

check-consistency &check

consistency - needed

drop-translations &Drops

any selected comorphism - needed

cons-checker

&{\textless}ConsChecker{\textgreater} Choose

consistency checker - needed

conservativity-check &{\textless}Edges{\textgreater}

&Choose conservativity checker - needed

set time-limit &{\textless}Number{\textgreater}

&Set the time-limit for the next proof - needed

set axioms

&{\textless}Axiom{\textgreater}

&Set the axioms used for the next proof - needed

set include-theorems true &Include proven

theorems - needed

set include-theorems false &Do not include

proven theorems - needed

nodes

&Show

Nodes - not needed

edges

&Show

Edges - not needed

show-undo-history &Show

Undo-History - not needed

show-redo-history &Show

Redo-History - not needed

show-proven-goals-current &Show Proven

Goals of selected node - needed

show-unproven-goals-current &Show Unproven Goals of

selected node - needed

show-all-axioms-current &Show All

Axioms of selected node - needed

show-all-goals-current &Show All

Goals of selected node - needed

show-computed-theory-current &Show Computed Theory of

selected node - needed

show-taxonomy-current &Show

Taxonomy of selected node - not needed

show-concept-current &Show

Concept of selected node - not needed

show-node-info-current &Show

Node-Info of selected node - needed

show-node-info &{\textless}Nodes{\textgreater}

&Show Node-Info - needed

show-computed-theory &{\textless}Nodes{\textgreater}

&Show Computed Theory - needed

show-all-goals &{\textless}Nodes{\textgreater}

&Show All Goals - needed

show-proven-goals &{\textless}Nodes{\textgreater}

&Show Proven Goals - needed

show-unproven-goals &{\textless}Nodes{\textgreater}

&Show Unproven Goals - needed

show-all-axioms &{\textless}Nodes{\textgreater}

&Show All Axioms - needed

show-taxonomy &{\textless}Nodes{\textgreater}

&Show Taxonomy - not needed

show-concept &{\textless}Nodes{\textgreater}

&Show Concept - not needed

show-edge-info &{\textless}Edges{\textgreater}

&Show Edge-Info - needed

expand

&Extend

current node - ???

addview

&Add a

view - ???

help

&Show

all available commands - see DG menus?

quit

&Quit

- not needed

\bigskip

\bigskip

Here is an API for sending around updates:

\begin{itemize}

\item prove link. Input: IRI of link. Output: list of new links and/or

proof goals for simple ontologies that will prove the link

\end{itemize}

\bigskip

\bigskip

\bigskip

Evaluation and other services

OOPS! and similar services

we propose the following abstraction from the OOPS! API:

input: ontology\footnote{\ OOPS! has more inputs, but we let the list of

pitfalls empty, and the output format be XML.}

output: list of response elements of the following form:

type (for OOPS: pitfall, warning, suggestion)

code (an integer)

name

description

list of involved symbols\footnote{\ OOPS! outputs structured XML

elements that may contain multiple n{}-ary relations between symbols

(e.g. oops:MightBeEquivalentProperty and

oops:MightBeEquivalentAttribute). We prefer to have only one such

relation per response element.}

\bigskip

Annotator Service

\bigskip

This service it specific to bio ontologies. How to generalise it to

other domains? It seems that some (more static) list of service types

and (more dynamically growing) list of actual services (conforming to

these service types) would be useful. This of course also includes

services like OOPS!

\bigskip

Ontology Recommender

\bigskip

Interesting challenge to generalise this to ontologies written in

arbitrary languages...

\bigskip

Resource Index Service

\bigskip

could be adapted for Ontohub, if {\textquotedbl}concept{\textquotedbl}

is replaced by {\textquotedbl}symbol{\textquotedbl}

\bigskip

Notes Service (Term Proposals and Comments)

\bigskip

Logic-specific services

OWL specific services involving the class hierarchy

These services could also be used for other languages if there is a

suitable projection to OWL.

\bigskip

Remaining stuff from OOR

&find AllOntologyOrViewVersionsByVirtualId\#

&find LatestAutoPulledOntologyVersions\#

&find LatestActiveOntologyOrViewVersion\#

&find LatestActiveOntologyViewVersions\#

&find LatestOntologyOrViewVersion\#

& find LatestOntologyViewVersions\#

\bigskip

\end{document}