Page History

Provide an email update to NCBO, NLM, and Mayo with current activities and what has been built.

open

Scott to work with Gilberto to determine what is needed for OWL 2 requirements

on hold

Larry: 

Contacted Ivor.  He is interested in looking at the prototype.  They have the beginnings of an RDF model for MESH.  Both are suggested as topics for the fall session.  No current time frame sent. 

Scott: MySQL 5.5 - Tomcat verification testing, Loader testing .

against 5.5

, Responding to requests around REST API. 

Final Tomcat verification tasks are in progress.

Cory:  Out

Craig: Project Plan Updates,  Delays in the 6116 environment and Shady Grove will impact this.  Might be an extra month. 

Document that captured a performance document that was submitted to justify better hardware.   Also include a basis for target performance.  After review, there was followup required.   Will require ongoing discussion and refinement of the document going forward.  

JJ asked if there was a timetable to address performance concerns.   The goal is that in the 6.2 time frame, plan to have adequate hardware and consistent hardware from Dev to Prod.   This timetable would lead to Sept 2014.    This performance will need to be included in the project plan - include creating performance profiles of 6.2 on DEV hardware on Shady Grove and then evaluate performance at that point and determine what else would need to be done.   Additional work would need to be prioritized against the currently planned tasks at that time.    Ongoing task and concerns however plans for this are tracked in the current version of the project plan.

Currently looking at all aspects of process - including updates of terminologies.   Agile approaches still being decided.  EVS will be a good product to evaluate approaches. 

Would like to have common education to provide a definition of how we will run the project with Agile.   This will ensure that all teams/groups will understand what aspects of Agile will be in this process.

Provide an initial analysis of how Agile could work in LexEVS environment that could feed requirements to the formal process.  

Risks on Mayo's side - small development team that is required to develop and also address immediate issues/concerns.  Typical agile sprints are focused and limited to defined sprint.  Moving forward, do we need to support agile and the existing expectations of support?  Need to consider such expectations.    TODO:  Scott to capture issues/concerns that we may have. 

Project plans to start reflecting agile in the September time frame.

Discuss issues and plan for any training/adoption framework.  The plan is still to start the discussion.  What are good ways to approach this.  Questions from Scott will be presented to an Agile expert who will comment on our current and future processes. TODO Jason.

URI format was agreed.  No additional news.  One concern was if the old URIs are being used by anyone.   Need to look to see if the URI resolver has logging.   TODO:  Scott to determine if there is logging. 

Report on URI Resolver logging (Scott) Scott has done some preliminary work.  Requirements for the logging service:  

• Who is using the service. 
• What URI's are hitting the resolver. 
• Can old URI's be dropped. 
• Might see people using URI's we do not yet support and use those as a standard for upgrades. 
• Figure out if we have to change marketing. 
• Can logging tell us where requests are coming from and who is making them. 

TODO Scott to look into this further.

Update.  See above in project plans updates.

Not much to add from the email from Gilberto.  NCI is not allowed to go to external sites for builds.  There is a local mirror. There is a mechanism allows them to create this.  In the case of LexEVS we will  have our own channels. We will notify systems about this.  The URL's for these will be entered in the PTE and the external github will be noted to the systems team.  Builds to be done from the local repository.  When we are ready to do a push to PROD at that point the code needs to be update the code in the NCIP channel.  A pull request will need to be made to do this.  NCIP Channel will be locked down – no code changes.  Pull request only for changes.  We should make the request that we do not fork from NCIP but from lexevs.  TODO look into how or whether you can fork from a fork – Scott

Recently received a forum posting from the old VKC forum.  Realized the readme was linking users to the VKC forum.  We can fix the readme, but now that we need to modify the files and push to NCIP channel, we need to figure out how we would make the change and push to both channels. 

Will need to discuss with Gilberto next week.

We will hold off and hope we don't have a lot more visitors to our current forum.   TODO Ask Ann Wiley to look into creating a forum sticky to direct users to our current github.

List of value set client - NCBO, TLAMP issues.  Scott and Craig to provide list of any existing issues.