Community Call May 20, 2021

[SydneyhSmith]

This meeting will be hosted as a Teams Meeting instead of as a Teams Live Event. We're experimenting with the new format, but this should enable us to allow folks to speak, use video, and share screens for demos.

As always, you can join the Community Call at https://aka.ms/JoinPSCall (9:30a PDT) with a Teams client or the browser web app.

Agenda:

• OpenSSH update
• Working Groups/Community update
• Crescendo preview 2
• SecretManagement 1.1 design change affecting extensions
• PowerShellGet v3 update
• Community Demos
• Q&A

In addition to questions and topics--which you should feel free to file below--we're looking for one or two demos from the community. These should be no more than 5 mins, and should be PowerShell-focused. If you'd like to give a demo just comment what you would like to show below.

[awakecoding]

Will the recent wave of PowerShell community members giving up on their own pull requests be discussed? Without pointing fingers, I do feel like a tipping point has been reached, and it deserves an appropriate response. The elephant in the room has to be discussed, ignoring it would be a grave mistake.

[JustinGrote]

Will the recent wave of PowerShell community members giving up on their own pull requests be discussed? Without pointing fingers, I do feel like a tipping point has been reached, and it deserves an appropriate response. The elephant in the room has to be discussed, ignoring it would be a grave mistake.

I imagine this will be addressed in the "Working Groups" agenda item and its probably no coincidence it is top of the list :)

[awakecoding]

@JustinGrote I really hope it's to give the news that they are finally happening with some actual decision power given to non-Microsoft contributors, because a lot of the heavy criticism has been about the failure of those working groups to take form despite prior announcements.

I really want to form a "vendor extensibility" working group and make some actual, meaningful contributions in that underserved area of PowerShell, and take part of important decisions like custom transports, because after all, who is going to implement those if not third-parties?

[iSazonov]

Please keep in mind that strong direct pressing of MSFT PowerShell team makes little sense and embarrasses them. Obviously they are first class PowerShell fans but they work within the company's policy and its investment plans, which they do not create.

This project is like a time bomb. Not only does it cause reputational damage to the company now, but it could eventually torpedo the entire .Net community in the future.

[awakecoding]

Please keep in mind that strong direct pressing of MSFT PowerShell team makes little sense and embarrasses them. Obviously they are first class PowerShell fans but they work within the company's policy and its investment plans, which they do not create.

This project is like a time bomb. Not only does it cause reputational damage to the company now, but it could eventually torpedo the entire .Net community in the future.

Your assessment is unfortunately correct. I apologize for the strong wording, and I do not hold the PowerShell team responsible. I personally met with a lot of the team back at Microsoft Ignite, and they are good people. As for those I haven't met but only interacted on twitter and GitHub issues, I can attest they are true PowerShell fans and definitely not the enemy here.

It is very hard to tell as an outsider how much power is given to the PowerShell team, and if the recent events have even been internally brought to the attention of decision makers higher up the chain. At a bare minimum, I hope that the severity of the current state of things has been acknowledged and that some flags have been raised. For obvious reasons, I don't think the PowerShell team is at liberty to discuss in public such internal discussions, but just a sign that internal action has been taken, even if unsuccessful, might help accept the situation.

[potatoqualitee]

I imagine this will be addressed in the "Working Groups" agenda item and its probably no coincidence it is top of the list :)

I think clear, direct language would be beneficial so that we don't have to imagine or assume. If you're right, @JustinGrote, Community concerns as item 1 followed by Working groups as item 2 would help address our uncertainty.

[michaeltlombardi]

Please keep in mind that strong direct pressing of MSFT PowerShell team makes little sense and embarrasses them. Obviously they are first class PowerShell fans but they work within the company's policy and its investment plans, which they do not create.

This project is like a time bomb. Not only does it cause reputational damage to the company now, but it could eventually torpedo the entire .Net community in the future.

I have... complex feelings about this comment. Basically:

1. Not pressing the team didn't lead to any changes so arguably speaking up was a necessary step
2. Beating people up as individuals doesn't make sense - assigning blame and being vitriolic towards individuals for systemic problems is a waste of time and energy (and something I wouldn't condone even if effective, which: it's not); luckily, I don't think I've seen this as a major component of the ongoing discussion
3. The reputational damage seems to correlate directly to the systemic problems plaguing the language and in particular the mismatch of expectations/promises and reality since the open sourcing; that is to say, sometimes reputations take a hit bc systemic issues and shortcomings were not addressed soon enough
4. C# and the broader dotnet community likely could not be less interested in this in the same way they're not very interested in things that happen in other languages they don't use

[JustinGrote]

EDIT: See submitted lightning demos below!

On the subject of lightning demos, obviously put me at the bottom of the list as I've already done more than my fair share, but if none otherwise I can demo my PowerConfig module that brings ASP.NET Configuration to powershell scripts and modules: https://github.com/justingrote/powerconfig

[JasonFossen]

If you can, please talk about why Microsoft under-funds the PowerShell Team and what that means long term. The hard work and dedication of the Team is beyond doubt. It's the lack of funding that prevents the hiring of additional Team members and then important work has to be pushed to the back burner (there are only 24 hours in a day).

If you expect the under-funding to continue, is it time to make strategic sacrifices to maintain a healthy rate of progress overall? For example, Desired State Configuration (DSC) is slowly dying and will never catch up to Ansible, Puppet, Chef, etc. It's sad to say it, but DSC seems like a "sunk cost" now, it's just pulling Microsoft and Community energy away from projects that are much more important for driving PowerShell adoption.

[JustinGrote]

@JasonFossen "If you can, please talk about why Microsoft under-funds the PowerShell Team and what that means long term."

That's a pretty loaded TMZ-style question, how about you air out all the dirty laundry of what you don't agree with about your employer and CC your boss as an icebreaker first? :)

I think a fair question though is based on the current roadmap of 7.2 priorities, why was DSC chosen and what does the Powershell team envision for its future that justifies the investment of resources.

[iSazonov]

Perhaps there are people available who supervise open projects at MSFT, who have decided to make this project open and who can be invited to this meeting.

[xmichelleyang]

I would like to give a lightning demo on some updates I've added to the GuestConfiguration commandlets to support set scenario.

[gaelcolas]

And I'd like to follow after Michelle to quickly show some Linux DSC resources and module I'm working on. What @xmichelleyang and I will share is early pre-release stuff, but we'd like to share and get it on people's radar, so that we can get some feedback and keep iterating on it.

[JustinGrote]

@SydneyhSmith it may be worth notifying and locking down which lightning demos the team wants to schedule for the call by the end of today so that people know they should be prepared, maybe specify 1 or 2 alternates if there are no-shows.

[SydneyhSmith]

Thanks @JustinGrote when we get to the community demos we will do @xmichelleyang , then @gaelcolas and then you if time permits/you are still interested in demoing

[potatoqualitee]

Community question: Can we get some in-depth reassurance about PowerShell's importance to Microsoft? It seems that Python is very important to Microsoft, and I know a few of us are questioning our own career investments in PowerShell. Will more resources be given to the PowerShell team and if not, is there anything we can do as a community to encourage Microsoft to reinvest in PowerShell? Perhaps help bring more visibility of PowerShell to the Linux community?

[michaeltlombardi]

(question xposted from teams chat)

Circling back to the community again a bit:

The community appears to feel like there has been a gap between what was promised and what was delivered re: open sourcing and contributions from the community. Can you speak to the circumstances that led to that expectation gap and what we can do to monitor the changes being made with working groups to ensure it doesn't continue to happen?

How are you measuring success for this endeavor? What roadblocks do you anticipate? What can we do to help?

For folks who love this language and community, it's probably the most important piece of this conversation.

[techthoughts2]

One of the other things that came up on the recent community discussions on Twitter etc was along the theme of PowerShell getting a more dedicated developer advocate role. Any thoughts / comments on that at this time?

[michaeltlombardi]

(question xposted from teams chat)

A follow up question for accountability/process based on conversations with folks who have contributed several patches or features:

It often seems like it's easier to get your first PR reviewed and merged than your fifth; that the barriers (real or perceived) to contribution goes up as community members get more capable and comfortable submitting code to the project.

Is there a reason for this? Do you have any ideas as to how this perception formed and what we can do to reverse it?

[iSazonov]

@joeyaiello You were talking about quality, which is pretty great in the last couple of years. You are in a dangerous area.

1. Code coverage check doesn't work and the index is stuck at 60% In reality the situation is worse since a lot of code is not covered by any tests at all, moreover there are quite a few scenarios not covered by tests.
2. CIs occasionally fail (false pass tests) from time to time.
3. Moreover release process catches failing tests even though CIs pass them.

Can we talk about quality in this situation? I'm not talking about the low quality of our work, I'm talking about the quality assurance process. I don't think it is possible to move forward until this process works so well that we can trust it.

[JustinGrote]

@iSazonov probably best to bump this to the latest issue post.