publish 1.4.0

[pixelzoom]

The PhET-iO client would like a final production version (1.4.0) of ph-scale-basics by "late August". RC testing will include both ph-scale and ph-scale-basics, for brands=phet,phet-io.

@arouinfar let's review open and blocking issues, and put together a timeline.

[pixelzoom]

At 7/2/2020 dev meeting, we noted that there are several sims (ph-scale, EFAC, SOM, ... others?) that all have the same "late August" deadline. So this should be part of a coordinated effort. See phetsims/tasks#1040.

[arouinfar]

Looking at the open issues in this repo, it doesn't seem like there is anything sim-specific to be done before 1.4.0. I think we're also down to just one blocking issue in common code, and it's actively being worked on.

@pixelzoom if I'm missing something, or if you would like to chat, please let me know.

[pixelzoom]

In 7/9/2020 phet-io meeting and phetsims/tasks#1040, it was decided to send pH Scale through the RC process first, to shake out any PhET-iO issues. Then we'll presumably use those shas for other sims that are part of the "late August" milestone.

Blocking issues, to be addressed before creating the RC branch:

• HomeScreen text is not dynamically centered for PhET-iO joist#600
• PhET-iO design for NumberSpinner sun#575
• https://github.com/phetsims/phet-io/issues/1678

[pixelzoom]

I ran into RC build problems on the build server. I deleted ph-scale 1.4 branch, and will start over.

These issues are now blocking:

• Puppeteer fails on CT phet-io builds chipper#940
• Phet-io Build fails on build-server due to need to npm prune and update in studio chipper#961

[pixelzoom]

RCs published 7/9/2020:

pH Scale 1.4.0-rc.1
QA issue: phetsims/qa#514
phet: https://phet-dev.colorado.edu/html/ph-scale/1.4.0-rc.1/phet/ph-scale_all_phet.html
phet-io: https://phet-dev.colorado.edu/html/ph-scale/1.4.0-rc.1/phet-io/

pH Scale: Basics 1.4.0-rc.1
QA issue: phetsims/qa#515
phet: https://phet-dev.colorado.edu/html/ph-scale-basics/1.4.0-rc.1/phet/ph-scale-basics_all_phet.html
phet-io: https://phet-dev.colorado.edu/html/ph-scale-basics/1.4.0-rc.1/phet-io

Testing completed 8/3/2020. Issues discovered:

2 sim-specific
6 common-code
12 PhET-iO

[pixelzoom]

RCs published 8/17/2020:

pH Scale 1.4.0-rc.2
QA issue: phetsims/qa#533
phet: https://phet-dev.colorado.edu/html/ph-scale/1.4.0-rc.2/phet/ph-scale_all_phet.html
phet-io: https://phet-dev.colorado.edu/html/ph-scale/1.4.0-rc.2/phet-io/

pH Scale: Basics 1.4.0-rc.2
QA issue: phetsims/qa#535
phet: https://phet-dev.colorado.edu/html/ph-scale-basics/1.4.0-rc.2/phet/ph-scale-basics_all_phet.html
phet-io: https://phet-dev.colorado.edu/html/ph-scale-basics/1.4.0-rc.2/phet-io

These RCs were cancelled on 8/18/20 due to new problems found in Studio, including phetsims/studio#183 .

[pixelzoom]

Summary of a long discussion in the Slack phet-io channel:

I am confused about what PhET-iO fixes (and especially Studio fixes) need to be included in the next RC, deferred for a future maintenance release, or can be safely ignored. There are so many changes that I have lost the ability to track and manage them. So before I publish another set of RCs, I will request sign-off that the list of "ready-for-QA" issues in the ph-scale repo is complete. According to @ariel-phet, he will request RCs, and @arouinfar will provide sign-off.

[pixelzoom]

In Slack phet-io channel 8/26/2020, there was a discussion about how iO changes have become difficult to track and patch. There have been many iO changes, the 1.4 release branches were created long ago (7/9/2020), and the next RCs for ph-scale and ph-scale-basics now pushed out. So we decided to abandon the 1.4 branches, and start over with new 1.5 release branches (from master) when RC testing is resumed. Further work is on hold until the client's schedule is known.

Closing this issue.

Slack discussion

Chris Malley 5:03 PM
For ph-scale, I suggest that we cut our losses, plan to abandon ph-scale 1.4 and ph-scale-basics 1.4, and start over with new release branches.

Ariel Paul 5:03 PM
But that seems to be the likely correct approach for pH-Scale

Chris Malley 5:04 PM
I'll put it on the phet-io meeting agenda to confirm. If you really want me to totally ignore ph-scale until the next RC (which I'll be happy to do), it's the only way forward.

Ariel Paul 5:06 PM
I think at this point it is clearly the correct call, it will incur a QA cost, but it will probably be smart to test off master at that point anyhow (edited)

@kathy-phet agrees @pixelzoom so no need to be on the agenda. You have the confirmation