I am presenting Pypowerwall at PyCon on May 17th -- Suggestions for what should be in the presentation?

[Nexarian]

Here's what I'm thinking so far:

• Show Powerwall Dashboard.
• Show my integration with 2x solar inverters and the custom modifications I had to do so.
• Talk about how Python is best when it integrates in an ecosystem, not a standalone monolith like Java apps tend to be.
• Discuss the community and its vibrancy.
• Gently address some of the problems with Tesla api access.
• Detail how Powerwall Dashboard has helped me diagnose issues with my system.

...Others?

@jasonacox ?

[JohnJ9ml]

I was able to take the data from pypowerwall, combine it with my data from my smart meter, and demonstrate to Tesla that my PW3 cluster was going through "violent mood swings" and flapping back and forth between consuming 300-500 watts from the grid and then sending 300-500 watts to the grid. Usage-wise, it was a wash, but billing-wise it was costing me more than $2 a day, even though my site was totally self-sufficient.

I have no idea if my data and the submission is what made the difference, or if it was just something in the plan anyway, but a couple weeks later 25.2.4 or 25.2.6 came out (I now forget which), and the flapping went back to what others experience, down in the 30-50 watt range in general. Now my daily bill is closer to $0.26 / day. I was only able to do this convincingly with the data from pypowerwall.

[Nexarian]

This is becoming true of my system, too. Multiple strings on one inverter randomly drop out for no reason, and it took using pypowerwall data to finally convince my installer of it. I just got a text from them today that they are now "true believers."

[JohnJ9ml]

BTW, not everybody writes monolithic Java apps. All my stuff here at the house is in Java, but it's a series of purose-built daemons that gather from various places and push them into the greater ecosystem of data at the house. So... it could happen. 😉

I do know what you mean, though, and it is likely generally the case that more and more functionality (even only nominally related) gets glommed onto the same process.

[Nexarian]

Yeah, not saying that Java or others can't be a microservice languages. Only that Python excels at being part of an ecosystem as it seems to have struck a good balance between technical rigor and ease-of-use/accessibility. Many other languages have come and gone that got that balance wrong (Looking at you Perl and Ruby)

[jasonacox]

Great list, @Nexarian !

Discuss the community and its vibrancy.

I really love this. In the context of PyCon, I think it would be fun to highlight is that Python has made the project approachable to enthusiasts of all levels of experience. We've seen new Python devs submitting PRs all the way up to guru level software engineers helping us shape the project. In fact, this project has encouraged some to learn Python and even try their hand at contributing to OSS for the first time.

[Nexarian]

@jasonacox I plan on putting a blurb at the end which is my tips for getting involved in OSS. This is a top request people I mentor ask me: I want to do open source -- How?

My thoughts:

• First, read. Read a lot. Read issues, read code. Try to understand it. If you set out to make big feature contributions at the beginning, you'll overwhelm yourself and likely fail.
• Second, post issues. Post questions. Try to get to know the people who have already done it.
• Start with writing documentation or small bug fixes. Have a mission as to WHY you're doing it. For many projects, the documentation for installation and setup and building it is often lacking, improve it, and you'll get familiarity.
• Don't bias yourself to only one project. You won't succeed unless you have motivation. If the ADD kicks in and you flitter to another project, that's OK!
• At the end of the day, this is world-changing. It's probably the easiest lever any of us can pull that TRULY affects hundreds if not 1000s of people, the ROI is insane.

Any other thoughts here?

[JohnJ9ml]

Watch videos.

I normally hate watching videos on how-to things, but my first check-in here is my only time ever doing something with github that was contributing to a project. At some point, watching something that went eggs to olives made it much easier to follow the terminology and the ordering, and allowed me to better focus on what it was I was trying to do.

The same would likely be true if someone had done the same for a project I was coming to. The "why" can be a rather dry topic when trying to read someone else's code who's not in the next cubicle. If you're going to contribute to a project, my thinking would be that your work should embody the spirit of what you're contributing to, not work against it. "Why" it came to be put together as it is would be good to know and maybe remove some doubts as to whether you're doing the right thing in the right way at the right time.

[jasonacox]

Maintaining OSS has been an incredible laboratory of learning for me. I love the community, diverse points of view, experts and novices that come along and help nudge things in the right direction. I'm biased to accept as much as I can, but learned the hard way that you need to be thoughtful about what you add to a project. While generally everyone has good intentions, I have come to understand that most people, especially first time contributors, focus on their own particular need, use case, hardware, problem, or interest area (a good thing) but often neglect to think though the wider usage or user base of the project. Additionally, there is a built in desire to impress, a very human and healthy thing, that can translate to being more clever than is needed, fix what is not broken, tending to complicate things over making them easier to maintain. Documentation is often omitted and even the PR comments can be hard to decode. I have seen (and learned) that we all have thin skin at times and critical feedback or misunderstandings cause hurt feelings. Lean in to critical feedback to make your contribution better and assume best intentions even if you disagree. Maintainers need to do the same, but also likely have 10 other things that they are thinking about for the project when a PR hits, so be patient, kind and helpful. :)

So adding to what you had:

• Understand the Project: Before contributing, thoroughly understand the project’s goals, architecture, and existing codebase.
• Focus on Wider User Needs: Consider how your contribution will benefit all users, not just those with specific use cases or interests similar to yours.
• Keep It Simple: Avoid unnecessary complexity, or being too clever; simpler solutions are often better maintained.
• Document: Ensure that your code and contributions are well-documented, making them easier for others to understand and maintain. Use clear and concise language when describing issues, proposing changes, or documenting issues.
• Make it Clear: Think about the reviewer and maintainer. Make it easy to understand your change and merge. Follow the "style" of the project as best as you can. Stage whatever you can to make it easier for the maintainer to add and release. Best way to do this is to think about what you would want as a maintainer.
• Seek Feedback: Be open to feedback from the community and be willing to iterate on your contributions based on constructive criticism.
• Test Thoroughly: Ensure that your contributions are tested and validated before submitting them. This helps maintain the quality and stability of the project.

You are right about impact! This project alone has been enjoyed by thousands of us... and all of those are wrapped up in a fantastic human story that we somehow impacted and (hopefully) encouraged, made better, easier or happier. :)

[Nexarian]

Here is what I presented yesterday. It got a couple laughs. I think there were about 20 people attending. It was at the David L. Lawrence convention center in Pittsburgh, PA for PyCon 2025: https://us.pycon.org/2025/events/hometown-heroes/

What I said is largely captured in the presenter notes.

I tried to record it but unfortunately the camera fell asleep in the middle of it so it only captured the first 15 minutes of the 27 minute presentation.

One question that was asked that I didn't know: Outside of Tesla, do other manufacturers provide such a rich data set?

[jasonacox]

Congratulations @Nexarian ! Nice presentation!

[Nexarian]

I will be giving a re-hash of the same presentation at a Pittsburgh PyData meetup sometime in October or November this year. I'll keep this discussion open until then.

[jasonacox]

Congrats and good luck, @Nexarian !