00). Progress on spec......
0. Deans Publish question
1) Where we are on our road map.
DW the ask is can now all load in Protege and TopBraid. Next is VOM and CCM. Meanwhile we have JG and DN agreeing on what an acceptable first cut in HTML is. We have http//schema.org now. If we can do something with SMIF to show what it would look like, and if can put a pointer on that, e.g. do BE and have it point back to the be folder in the ontology structure to close the loop - can we do that to be published June 30 as our 2nd quarter deliverable." Want to announce at 8 June EDMC members meeting in NYC. DA will be there.
2) Open Action Items
3) JIRA Issues Review - https://jira.edmcouncil.org/projects/RDFKIT/issues
4) Todays content discussion.
5) For next week.
20170518 FIBO FPT RDF ToolKit
Dean walks us through revised spec.... and asks for the team to test what has done? Must have Protege 5.2 loaded. Bobbi: can we test in CCM? DA, no can u write this up? Bobbin yes.
ACTION: Bobbin to write up testing CCM ingesting FIBO from spec……. Dean needs more information on some of the things BT will help
Some of the questions DA has only JG can do.
ACTION: JG answer Deans email questions.
OK: verifies that Deans updated page is published.
DW: getting this generated is the main thing. MA reacted positively to this. There are a lot of acronyms. Can we make those into hyperlinks so we can add definition? BT: Suggest you leave the SEC etc. and put the URLs in parentheses). However we need a strategy for consistent urls.
CC: What's missing to have a pattern. JG the versionIRI is still not there. Also what tech we use to do these? E.g. generate html files for every directory simple and searchable by crawlers. This is on the content in pervious question was on common acronyms like OMG. Can generate pages at whatever level e.g. ontology, product, branch. Single page app would read the json files and generate the html page.
Back to the easier question i.e. SMIF OMG etc. Links to external sites can be done by adding something to infra. MA wants all to get access to all. Change of policy. Makes things easier. JG: the infra repo does not need to be public. it contains this file in git format. when we change, and push this, Jenkins picks up the change and publishes it. Can we do that now? Yes.
CC talking about how every class has a specific url for each product type e.g. organization, owl v organization, glossary. need a pattern that is symmetrical across these. for the urls for the web pages for each concept. JG outlines a solution for what order the parts of the urls go. Each such page needs to integrate with others. should be possible to change out each part of the url and get to the relevant thing e.g. vocab v owl relevant to header because from a common page header pattern for navigation it needs to be able to have links to the related views for the thing. Would not work for vocab since that has the URI at a different level.
OK: I would like to try a parallel approach; caveat David Newmans approval. https://github.com/LinkedDataFragments/Server. JG: Would need to generate a folder structure for vocab to do this. OK: http://linkeddatafragments.org/concept/ Need not be : as long as it can be done algorithmically. Next to the ttl file would be an html file with all the terms for the glossary. We know which class defines each concept. This does not cover schema or SKOS want to keep these things decoupled so we do not need to rely on the same structure. To do this we would need a diagram for each concept. This is ok, we just need a treatment for concepts that do not yet have a defining diagram.
UML is one thing. Diagrams are derived from that. The pub for the uml/smif part should be command line driven. This needs to be headless. Can we get the content out initially in some way until then? JG: no. all content should be generated first time. JG should not be possible to put anything on spec that is not auto. DW Yes.
OK has an idea Linked Data fragments, from university of Ghent. Can provide this. Various docs, has instructions to run it on nGnix. Lets you take our artifacts and query independently. Can this be updated dynamically? looks like it. OK would have to do some work with permission from DN. can pub triples, ttl docs out of the box. Maybe use for an end point we can point something to. HTML choice is dynamic HTML page generation on server versus generating the content. Ghent University
JG we can install this in parallel. The aim of http://spec.edmcouncil.org is to reflect all and only what is in git. The Ghent thing could be another namespace e.g. http://linkeddata.edmcouncil.org although LD is usually ABox.
DW: Are CC and JG on the se page? JG I don't see SMIF on there. SMIF is reported now at Collaborator. This is not on Git. DW we need to back out of what we were doing in the earlier debt session. Need to back into Git machine readable version of all the concepts needs to be there these are not in GitHub, but they can be. So if we put these into git then every push we do in git generates a new version of the whole site. same structure for SMIF as everything else. So instead of having the SMIF output on the Collaborator we can put these files on git. These are binary files. So if we insist that all fibo is all on git then we have non binary on git.
Action is under way with CC JL MB.
The diagram from DW to MB was sent for the getting started guide action to use about file. There is an app to write pdf docs from a range of pages. use for the policy pages.
Next item: The thing Kevin was to show will be next week. All to digest his diagram. JG has talked Kevin re quad file. Is easy.
HTML action. LODE v OWLDoc review. Also TQ HTML. DA talking to TQ. JG knows LODE, not v impressive. We can support them all, as long as can be done from Jenkins. Could do them all on slightly different urls we need one for now.
- Elie Abi-Lahoud Update the FIBO Product Ontology
- Bobbin Teegarden to write up testing CCM ingesting FIBO from spec……. Dean needs more information on some of the things BT will help
- Jacobus Geluk answer Deans email questions.