1) Where we are on our road map.
2) Open Action Items
3) JIRA Issues Review - https://jira.edmcouncil.org/projects/RDFKIT/issues
4) Todays content discussion.
1). spec sign in page
2). From Atkin Spring report to members
3) Finish Dallas discussion
5) For next week.
20170530 FIBO FPT RDF ToolKit
Wiki -we are using RDFKIt Wiki space for FPT meetings since March
Items: the spec sign-in page. JG has taken this on. Looks nice. This is where we have the for that was asked for, that JG did over the weekend.
Filling the form sends an email to Dennis Wisnosky. There is one minor problem. If you let it sit for a while it times out, which is fine, but when it does time out it then does something inappropriate.
MA: downloaded Protege last week to look at the OWL files. However it doesn't open up FIBO when he clicks on Protege. Should be able to follow the instructions, so it seems these are not working. Should be able to do off line or on line. There is a problem with the font.
ACTION: DA - the instructions need to support on line and off line working. Imagine what it is not to know Protege, so that non Protege users can download and use. New user / audience requirement: people who do not know Protege but have been encouraged to download and use that. The Protege site gives "Download and install" instructions so we need to include a link to that and an explanation that you should follow what is written there.. So add Step 0: Download and install Protege.
JG used an out of the box thing from Google that also has Captcha. We need to deal with the un-catered for state where the time-out has happened, and write an action for what happens then.
Spring report to members. This is being released in Toronto. Also a note went out today at am Eastern Time from MA.
Review of slide on FIBO Access. DW: want to make sure everyone knows that this is for real, and that we have kept our promise wrt this web page.
Now need to be sire we can deliver onthe new deliverables on the 0th.
Q on the Spring Report Slide above. JL: diagrams not fully compliant with the SMIF standard yet. Please clarify. DW we don't have all the diagrams there, it will only be an example of how the diagrams are going to look. The "yet" is in the wrong place - we don't yet have all diagrams, in a SMIF compliant form.
MA can change this. JL or CC to write up the needed word. "Not all FIBO content currently has SMIF diagrams". So the slide now shows something we all understand and agree about. Only this one slide is now a page in the wiki, as it spells out the deliverables. PR Are we saying that FIBO OWL is not going to change between now and end of June? DW not so - this is the completion of the "Dallas discussion".
Next agenda item = where we are (the Dallas meeting). DA: We are working on the round trip of what is currently in GitHub back into CCM so we can pick up the diagrams. What CCM are we bringing it into? Both Pink and FIBO-V Full (the rest. After we have done that we can identify whether or how to reconfigure the model / project file. This depends on the outcome of tomorrow's meeting on getting TWC working again. So everyone is dependent on Jovaldas who is gathering the needed information. By end of tomorrow we will have a confident round trip cycle going.
DW question for JL - any more on the progress on the TWC blocking matter. JL has exchanged emails with Jovaldas who is asking what is expected from the meeting tomorrow. We need to have everything needed from NoMagic before tomorrow's meeting. IT seems that by having individuals with different configs, we have confused the TWC server. They will have to tell us how to resolve. Also everyone who ever touches the file. We should not need to be able to open Pink as read-write in the first instance, we will need to do that for a later iteration. The OWL should be in piles. One that is everything we need in Pink and one that is everything else. MB this is how we have been maintaining these models
JL is it the case that any ontology in Pink does not need to import anything from the rest of it? MB we need a tightly managed process for the CCM models, at the same level of discipline that we have been doing to date for OWL things. DA we also need the FCTs staff otologists to know the details of those processes when complete.
To Dennis diagram slide about FCT roles and environments. Comments on this: OK for CCM but does not include the separate matter of people being able to edit in OWL tools. JL: Last Thursday not everyone was on the same page. Needed more understanding on what the big chunks. This upper right includes stuff that was in MB process slides from OMG Reston in March. JL there is also subtlety in the upper left. For example the local branches, Git branches, pull requests.
DA also need to coordinate between the upper left and upper right. This was covered in MB slides which need expanding with more detail. Right now we do not have a place to check in the MD.Zip files. At present this is done manually. Eventually this needs to be automated, so that every time someone checks in a project file into TWC, we have something that checks things into GitHub. DW can take a shot at the upper circles. Can also take care of the bottom part.
ACTION: MB to re-send the slides from Reston. MB to update that current simple slide. Covers Git, CCM and the rest, and then other Git experts can improve upon that part of this. On the Left Hand Side of the diagram, JL may need to write some more.
JL comment: re the comment above isn't everything Pink now? MB: FIBO Master has large chunks: Pink + Dirty Pink + Extensions. Pink is Pink and has not changed. The Pink inside Master is the same as the Pink in the Pink branch n GitHub. DA has to sometimes do something to keep those in Pink. Pink corresponds to TWC:PinkFIBO. In TWC you can have one project import another, which is not something we can do in GitHub. So there is a sub-set of one branch being identical to a different branch. This has been an issue for policy management on the GitHub side. JG and TC aware. So there is a lot of process and education after tomorrow meeting is successful.
What are the next steps for the Cory presentation that corresponds to the lower left of this slide? This is the deck with the proposal for the header etc. This has stagnated for reasons: 1) we needed the content (30 June). 2) The revised EDM Council site needed to be there, as schedule for end of Q3.
The FDTF Content group has been meeting and will have feedback on what kind of thing they will be recommending. Timeline on that? No specific timeline, but should be able to give you some feedback at next week's OMG in Brussels.
CC if we are generating content and are automating it we should be able to do something without needing to wait on the EDM Council work. So there is both technical and user experience. We did reach consensus that we want to be able to land on one page from which different people can select different viewpoints (diagrams, written definition etc.). JL there is a technical matter: if you want to show a SMIF diagram we have to go from a URI to a global UID to a URL for the page n collaborator . So someone needs to write a script that would map the IRI to the eventual URL Ideally a java script. If we use naming conventions about this, the script should know about the naming conventions. This is one of those things where there are many viable ways of doing it, we just need to pick one and write it down. Cory's slides include such a proposal.
JL stresses that a java script will need some sort of script to tell it the mapping between the IRI and the GUID since the URL in CCM is based on the GUID.
Need a comment "This URI links to tie URL in Collaborator". They will look nothing alike.
CC suggests a small group to work on this and write the java script. This will be on Thursday agenda.
- Dean Allemang the instructions need to support on line and off line working. Imagine what it is not to know Protege, so that non Protege users can download and use. New user / audience requirement: people who do not know Protege but have been encouraged to download and use that. The Protege site gives "Download and install" instructions so we need to include a link to that and an explanation that you should follow what is written there.. So add Step 0: Download and install Protege.
- Mike Bennett to re-send the slides from Reston. MB to update that current simple slide. Covers Git, CCM and the rest, and then other Git experts can improve upon that part of this. On the Left Hand Side of the diagram, JL may need to write some more.
- Dennis Wisnosky Put CC ideas for URLs and URIs on the June 1 FPT agenda
- Dennis Wisnosky Write some more words on the JL OWL slide
Here are the slides I presented in Reston on the initial requirements for part of the process around maintaining CCM model repositories in line with the OWL.
Collaborator and FIBO.pptx
The relevant stuff starts on Slide 5.
The main thrust of these is what to do when something that was DirtyPink or Ext needs to becoem Pink. There are other usage scenarios that also need to be mapped out.
This is the basic process requirement, and does not show detail about who does what and when. That detail is implied in the vertical swimlanes showing different environments and activities, e.g. GitHub, Pull Requests.