OpenOntologyRepository: presentation on "Ontohub" - Tue 2012_03_27    (37VH)

Topic: Ontohub: an OOR-compliant Repository for Distributed Ontologies    (37VI)

Session Chair: Dr. habil. TillMossakowski (Bremen University)    (37WQ)

Archives:    (383D)

Abstract:    (37WL)

Topic: Ontohub: an OOR-compliant Repository for Distributed Ontologies    (37WM)

by: JulianKornberger, ChristophLange, TillMossakowski, HenningMueller    (37WN)

We demonstrate Ontohub, an ontology repository engine with a web frontend. Ontohub specializes on managing distributed ontologies in the sense of ISO Working Draft 17347 (Ontology Integration and Interoperability – Distributed Ontology Language). Distributed means: logically heterogeneous, modular, interlinked, and distributed over the Web.    (37WO)

In its current state, Ontohub is the fully functional result of two weeks of agile web development – supporting a small but growing set of features. Users can upload, browse and annotate ontologies in OWL and Common Logic. We have developed a backend service that analyzes the structure of ontologies in various logics and returns their structural outline (axioms, entities, soon imports as well) as a lightweight, logic-independent XML description, which the web frontend makes browsable. The next aspects of distributedness that we will realize are cross-logic search for entities, cross-logic links between ontologies, as well as linked data compliance for reusing ontologies over the Web.    (3837)

Ontohub does not aim at reinventing mature ontology repository engines such as BioPortal. It does aim at supporting a level of distributedness that existing systems do not support, due to their strong focus on single ontology languages and logics. Aiming at practical support with distributed ontologies and at quickly gaining experience with them in realistic application settings, we had to build a self-contained repository engine. For realizing its general, "non-distributed" functionality, such as the user interface and user management, we followed a subset of the requirements that had previously been established for the OOR.    (3838)

See Ontohub (demo) site at: http://ontohub.informatik.uni-bremen.de/    (3839)

References:    (383M)

Conference Call Details:    (37VJ)

Attendees    (37WA)

Agenda & Proceedings    (37WP)

Full proceedings will be made available in our archives after the session.    (383P)

1. Meeting called to order ... opening (chair), participants' self-introductions. (~10 min.)    (383Q)

2. Introduction of the "Ontohub" project ... (ref. slides)    (383R)

3. Demonstration of the "Ontohub" interim deliverable - the fully functional result of two weeks of agile web development ... (ref. video clip)    (383S)

4. Q & A and Open discussion (~20 min.) ... (ref. process above)    (383T)

6. Wrap-up: summary and Follow-up actions (where applicable) - chair (~5 min.)    (383U)

Transcript of the online chat during the session:    (383V)

 see raw transcript here.    (383X)
 (for better clarity, the version below is a re-organized and lightly edited chat-transcript.)
 Participants are welcome to make light edits to their own contributions as they see fit.    (383Y)
 -- begin in-session chat-transcript --    (383Z)
	[08:21] PeterYim: Welcome to the    (38D1)
	 = OpenOntologyRepository: presentation on "Ontohub" - Tue 2012_03_27  =    (38D2)
	Topic: Ontohub: an OOR-compliant Repository for Distributed Ontologies    (37VI)    (38D3)
	Session Chair: Dr. habil. TillMossakowski (Bremen University)    (38D4)
	Session page: http://ontolog.cim3.net/cgi-bin/wiki.pl?OOR/ConferenceCall_2012_03_27    (38D5)
	Mute control: *7 to un-mute ... *6 to mute    (38D6)
	Can't find Skype Dial pad? ... it's under the "Call" dropdown menu as "Show Dial pad"    (38D7)
	 == Proceedings: ==    (38D8)
	[08:28] anonymous1 morphed into HenningMueller    (38D9)
	[08:28] PeterYim: Hello Henning    (38DA)
	[08:28] ChristophLange: Hi all, I will join the skype call in 2 minutes    (38DB)
	[08:29] HenningMueller: Hey, I'm sorry but I could not organize a working microphone setup for skype.... :/    (38DC)
	[08:29] HenningMueller: So I am just listening today..    (38DD)
	[08:29] PeterYim: @Henning - no problem, please just use the chat to communicate when necessary    (38DE)
	[08:31] HenningMueller: Ok    (38DF)
	[08:33] anonymous morphed into JohnBateman    (38DG)
	[08:38] PeterYim: == session started 8:37am PDT    (38DH)
	[08:39] PeterYim: TillMossakowski opens the session    (38DI)
	[08:39] PeterYim: == ChristophLange presents (see slides)    (38DJ)
	[08:40] HenningMueller: I sent Julian the link to the session page    (38DK)
	[08:41] TillMossakowski: Demo site: http://ontohub.informatik.uni-bremen.de ... Sources: http://github.com/ontohub/ontohub    (38DL)
	[08:45] anonymous morphed into JulianKornberger    (38DM)
	[08:47] TillMossakowski: Please find the video at http://youtu.be/QixyGs8YpHc    (38DN)
	[08:50] MichaelGruninger: there is sound on the video (just a little on the quiet side)    (38DO)
	[08:50] ChristophLange: video file also available as: http://kwarc.info/clange/exchange/ontohub.ogv (full resolution) 
                & http://kwarc.info/clange/exchange/ontohub.avi (lower resolution)    (38DP)
	[08:53] PeterYim: == demo video clip shown    (38DQ)
	[08:54] ToddSchneider: So you're not 'creating' an ontology (from scratch), but rather registering an ontology?    (38DR)
	[08:58] AliHashemi: shouldn't the button be called "import" vs "create" ? :D    (38DS)
	[09:00] JulianKornberger: "create" is the rails default. i agree that "import" fits better in this context    (38DT)
	[09:01] ChristophLange: @all: Is there a risk that "import" may be confused with "one ontology importing another 
                one"? If so, do you have suggestions on the terminology?    (38DU)
	[09:02] ToddSchneider: Register    (38DV)
	[09:03] AliHashemi: or "Add"?    (38DW)
	[09:05] ChristophLange: @ToddSchneider: "register" might be confused with registering a user account    (38DX)
	[09:06] AliHashemi: @Christoph, presumably it's "Register Ontology" ?    (38DY)
	[09:03] ToddSchneider: I'm done with the video demo    (38DZ)
	[09:03] AliHashemi: finished here as well.    (38E0)
	[09:05] PeterYim: == resuming slide presentation ... slide #10    (38E1)
	[09:06] PeterYim: continue from slide#15    (38E2)
	[09:07] TillMossakowski: Please find a demo of Hets under: http://ontolog.cim3.net/cgi-bin/wiki.pl?OpenOntologyRepository_Architecture/From_HeTS    (38E3)
	[09:17] ToddSchneider: Apologies, I have to go to another meeting. Thank you.    (38E4)
	[09:27] HenningMueller: (ref. the use of postgreSQL as the rdbms backend for ontohub) At this point, it would be 
                absolutely no problem to change to another DBMS.    (38E5)
	[09:29] JulianKornberger: actually we do not use any special functions of postgresql yet    (38E6)
	[09:32] PeterYim: @Julian & all - if there is no particular preference, then would you please consider using mysql 
                (since it is already part of the BioPortal technology stack, at least for now) ... allows for an easier integration    (38E7)
	[09:34] Till / Michael / Peter: (ref. BioPortal moving to a triple store backend) discussion of the possibility of 
                using a triple store as the backend for ontohub as well    (38E8)
	[09:36] ChristophLange: Ken et al.: The question: When I have an ontology with a URI and create a new version of 
                it (possibly importing from a different web URL), what happens then?    (38E9)
	[09:35] JulianKornberger: the metadata is actually only stored for the current version    (38EA)
	[09:35] JulianKornberger: (the way it is implemented now) the new version will be parsed and stored in the database    (38EB)
	[09:36] JulianKornberger: everything that does not exist in the current version will be deleted    (38EC)
	[09:39] PeterYim: (capturing for the record): Christoph and Michael will work together on making this "ontohub" 
                demo available at the OntologySummit2012_Symposium demo session    (38ED)
	[09:44] JulianKornberger: i have to leave in 5 minutes    (38EE)
	[09:45] AliHashemi: (ref. versioning) this may be relevant: http://www.w3.org/2001/sw/wiki/White_paper_Vocabulary_Management    (38EF)
	[10:03] AliHashemi: the above W3C doc pointed to SKOS and DC (Dublin Core) recommendations re versioning URIS    (38EG)
	[10:05] ChristophLange: They just used the SKOS spec as an example for "any W3C spec", because these specs are 
                versioned. SKOS itself doesn't answer our question.    (38EH)
	[09:46] JulianKornberger: i would append the version number to the URI to keep it RESTful    (38EI)
	[09:49] MikeDean: (ref. versioning) OWL 2: http://www.w3.org/TR/owl2-syntax/#Ontology_IRI_and_Version_IRI    (38EJ)
	[09:49] JulianKornberger: I have to leave now, but I leave the chat window open to read the log later. bye    (38EK)
	[09:50] TillMossakowski: many thanks, and bye Julian    (38EL)
	[09:50] PeterYim: @Julian - thank you ... we will post the chat-transcript and the audio recording to the session 
                page later too    (38EM)
	[09:55] AliHashemi: this also seems relevant to the uri-versioning that was discussed earlier: 
                http://stackoverflow.com/questions/389169/best-practices-for-api-versioning    (38EN)
	[09:57] PeterYim: please refer to the BioPortal "tab" implementation details 
                - see: http://palexander.posterous.com/extending-bioportals-rails-ui    (38EO)
	[10:01] ChristophLange: Note: whatever versioning approach we come up with, we still need to consider that other 
                repositories and other ontologies might not use it. There will still be ontologies keeping the same URI 
                while their content is changing. We also need to be prepared for "interoperating" with such ontologies, 
                e.g. when a perfectly versioned ontology inside our repository references (e.g. imports) a "messy" external one.    (38EP)
	[10:04] AliHashemi: thanks, bye    (38EQ)
	[10:05] HenningMueller: Thanks for the feedback! Cya    (38ER)
	[10:06] PeterYim: impressive work ... and a very productive session! Thank you ALL from the Bremen team, and 
                everyone who were here today!    (38ES)
	[10:02] PeterYim: -- session ended: 10:02am PDT --    (38ET)
 -- end of in-session chat-transcript --    (3840)

7. Schedule Next Meeting & Adjourn:    (3843)

 --
 notes taken by: PeterYim / 2012.03.27-11:17am PST
 All participants, please review and edit to enhance accuracy and granularity of the documented proceedings.    (3849)

Resources    (384A)

For the record ...    (388G)

How To Join (while the session is in progress)    (386T)