EPICS Version 4 Telecon, 19-Jan-2016
 APS Thurs f-2-f meeting minutes, See topic “Combining V3 with V4 — discussion (Andrew)” https://docs.google.com/document/d/1lO7X91LqaWQZulnZIu82lQkDoweXF8QwkpOZwtkvCXQ/pub
 Andrew’s presention on merging V3 and V4, see slides 8,9,10: http://epics-pvdata.sourceforge.net/files/talks/2015/ANJ-F2F-201511.pdf
 Dave’s email, https://sourceforge.net/p/epics-pvdata/mailman/message/34776144/
Proposed training rubric,
Present: AJ, DH, GW, MK, MS, MD, KK, SH
GW: [Action Item 2] progressing on obtaining funds for dbGroup @ SLAC; LCLS-II experimental side will use V4.
AJ:  done. [5, 6] no target date, yet.
MK: See pvDataCPP license file.
AJ:  ongoing.  ?.
GW:  mostly about finding the face-2-face minutes, having only user guides on lit. page.
DH:  working on wrapping up release.  contacted TK; based on what was used at FRIB EPICS meeting.
DH: Looking at pvaSrv bugs: 1st update of monitor may be missed; queue size potentially wrong.
GW: This is about RPC exampleCPP regressions. Outstanding pull requests should be discussed.
MK: Examples that rely on running EPICS base (V3) database and PV (V4) records are moved. Let’s merge so DH can continue.
DH: OK to merge, but what about git history when files are moved? Should MK merge to preserve file history?
AJ: Githib states “too many changes to show”
KK: Github website may not be able to show you all changes, but git command line would show all
MK: I’d Like DH to do the merge. Please merge and then do pull request
DH: I’d like also to resolve issue of folder called “test”. What is it? Genuinely test code, or more example?
MK: pvaClientTestCPP can go away since moving to exampleCPP
AJ: Prefer to have module that can be executed as integration test, not buried as example and then not automatically executed. CPP/Java integration test may be hard, but at least CPP tests.
MK: It’s a good example for writing a regression test. It’s an example, not integration test.
MK: pvDatabaseCPP pull request that uses weak pointers instead of shared pointers. Would like to merge before more test/example reorg.
MD: Unclear about extent of changes, worried about exception handling.
MK: May need review, but that’s separate issue.
AJ: There are lock/unlock calls in the change. MD more suggestions?
MD: Need to add guards for locks or try/catch to assert unlock in case of exceptions.
AJ: Need more time.
MK: Go ahead and merge?
KK: Merge, since it’s master and not a release, then add bug report?
MD: Not easy to demonstrate, but pretty obvious from code review that there is a problem
GW: Want to avoid adding potential bugs to master
→ MK: OK, will revert weak pointer changes, then await MD’s suggestions. Will go ahead with example/regression test update.
AJ: Not ready to merge EPICS base & V4, but could provide a tar file that combines everything.
GW: Talked to Mark @ DLS. Their priority is a merge of V3/V4. First step is combined release. Second step - and much longer term, is a true integration of V3 and V4 (based on Andrew’s list given in the f-2-f).
MK: Having everything (asyn, synapps, …) all as one will not be manageable.
AJ: Correct. Won’t be able to do separate releases of base, pvData etc.
MK: So need to consider carefully what parts of V4 to merge into base
AJ: Yes: pvData, pvAccess, normative types.
GW: For ESS EPICS meeting, wanted to have combined download. Longer term plan would be to discuss/decide on what full integration means.
AJ: There are still incomplete prerequisites to merging: Resolve/remove Boost, ..
KK: Is the short term plan, of a combined download, still on track for the ESS meeting?
AJ: Appears likely, using base 3.15 and 4.5.0
GW: Need to start release planning earlier
AJ: 4.6 time table?
GW: Nothing at this time
MK: Start in March. 4.6 release should be before EPICS @ ESS
AJ: That might be 4.5.1
→ Have combined download at time of Lund meeting
MK: Good list of topics. Emailed some comments. Concerned about time available for hands on. Requires people already have something pre-installed on their laptops.
DH: Certainly too much for ½ day. Would we have 1 full day?
KK: Suggest one day?
GW: Need to ask TK as organizer of meeting who also asked for this to train his coworkers.
AJ: Maybe that’s a designated set of training days, not part of the EPICS meeting.
AJ: Split Java and C++ sections? Or emphasize similarities and do both at once?
GW: Emphasize C++ and mention that Java is similar
AJ:  is done, by DH, not TK
GW: New action item on TK: How long? How many rooms, participants, ..?
MD: Pending pull requests from last year. share libs
AJ: Module owners should review their pull requests, discuss, merge
See who can fund dbGroup
Create a release links file (docbuild/<version>/index.html) for previous versions
Write the one license file (using the MIT license in most cases)
Update your module to include copy of that one LICENSE. Update your headers to contain the header text.
ALL MODULE OWNERS
wait on 5
On your module pages, remove the current/previous links; replace “Working Draft” with version number of module, which has its own numbering scheme, like 5.0.2 for pvDataCPP. Eg “Version 5.0.2 18/Nov/2015”
ALL MODULE OWNERS
Change the C++ bundling to add creation of a tarfile that includes Base 3.15.x
Simplify Literature page and home page to help users find novice oriented documentation.
Usher the patch fix for buffer overflow through to release.
Prepare a list of course contents for ESS EPICS Meeting training, and practicalities