I let chatgpt have a go at creating our project strategy

The Fedora Council put in a lot of work last week, with the primary focus around creating a top-level draft for our five-year focus. I have a post about that which should go live Monday or Tuesday. But I figured, with this conversation going on, might as well let the fad of the year take a swing at it…

Me: Do you know what a logic model is?

ChatGPT: Yes, I am familiar with the concept of a logic model. A logic model is a graphical representation of the relationships between the resources, activities, outputs, and outcomes of a program or project. It is used as a planning and evaluation tool to help program managers and stakeholders understand the underlying logic and assumptions of the program and its design.

A logic model typically includes a clear statement of the program’s objectives, the inputs or resources required to achieve the objectives, the activities that will be carried out, the expected outputs or intermediate results, and the ultimate outcomes or impact of the program. The logic model provides a roadmap for program implementation and helps to ensure that all aspects of the program are aligned and working towards a common goal.

Me: Do you know what a North Star strategy is?

ChatGPT: Yes, I am familiar with the concept of a North Star strategy. A North Star strategy is a term used to describe a clear and compelling long-term vision that guides decision-making and aligns the efforts of an organization or team towards a common goal. The term “North Star” refers to the North Star, which has been used as a navigation aid for centuries and serves as a reliable point of reference in the night sky.

In a business or organizational context, a North Star strategy is a strategic framework that provides a clear direction for the organization’s growth and development, and helps to ensure that all efforts are aligned and working towards the same objective. The North Star strategy is often used as a way to maintain focus and ensure that the organization is on track, even in the face of change or uncertainty.

A North Star strategy typically includes a clear statement of the organization’s long-term vision, its core values, and its mission. It also defines the key metrics and goals that will be used to measure success and progress towards the vision. By having a clear North Star strategy, organizations can make informed and aligned decisions, prioritize their resources effectively, and ultimately achieve their long-term goals and objectives.

Me: Good enough! Explain how the North Star concept might fit into the Logic Model framework.

ChatGPT: The North Star concept and the Logic Model framework can complement each other and work together to provide a comprehensive strategy for organizations. In the context of a Logic Model, the North Star concept can be seen as the overarching long-term vision that guides the development of the program or project.

[Deleted a long and somewhat strange example that ChatGPT generated, with snake-eating-its-talk logic model outline showing how to use a North Star metric in a logic model.]

Me: Write a potential Logic Model outlining a five-year strategic plan for the Fedora Project. The North Star metric for this plan should be: “Double the number of contributors active every week.”

ChatGPT: Here is a sample Logic Model outlining a five-year strategic plan for the Fedora Project with a North Star metric of “Double the number of contributors active every week.”

Objective: To double the number of contributors active every week.

Inputs:

  • Dedicated team of Fedora developers and community members
  • Adequate funding for development and outreach efforts
  • Community feedback and suggestions for improvements
  • Partnerships with other open-source projects and organizations

Activities:

  • Conduct user research to understand the needs and motivations of potential contributors
  • Develop and implement programs to encourage and support new contributions, such as mentorship programs, coding sprints, and online communities
  • Expand outreach and education efforts to raise awareness of Fedora and encourage contributions from a wider range of individuals and organizations
  • Foster partnerships with other open-source projects and organizations to share resources and collaborate on development efforts
  • Continuously monitor and track contributor activity to measure progress towards the North Star metric

Outputs:

  • Increased contributor engagement, as evidenced by the number of individuals actively contributing to Fedora
  • Improved contributor experience, as evidenced by increased satisfaction and retention among contributors
  • Enhanced collaboration and partnerships with other open-source projects and organizations

Outcomes:

  • A more vibrant and sustainable open-source community, as evidenced by a significant increase in the number of active contributors to Fedora
  • Increased contributions and collaboration from community members, resulting in a more robust and sustainable open-source project.

Note: This is a sample Logic Model and the actual five-year strategic plan for the Fedora Project could differ based on various factors and considerations. The North Star metric serves as a guiding principle and measure of success, and should be regularly reviewed and updated as needed to ensure that the strategy remains aligned with the goals and objectives of the Fedora Project.


What do you think? Still need humans?

Well heck, why did we even spend a week discussing this? We could have been playing lazer tag instead!

In seriousness: like a lot of ChatGPT stuff, this sounds blandly good — but, for example, “Develop and implement programs to encourage and support new contributions, such as mentorship programs, coding sprints, and online communities” sure is lacking in specifics.

Our actual plans will need to, alas, go into detail.

One of my issues with AI is not in where the technology is now, but where it could be in the future. This plan may be lacking detail, but having an AI do this at all would have been mind blowing in 2015. Heck, it would have been mind blowing in 2022 if AI image generation didn’t steal the thunder. If you give this tech 5 more years and point it specifically at all of Fedora Discussion and the Fedora subreddit, for example, I think it could get a lot closer to giving actually helpful guidance even in complicated, high level strategy decisions. I could even imagine another project running this exercise first and using it as a rough draft for further discussion, as others online have suggested for writing. If AI can write something good enough to start you off, I think it can do better than that over time.

In the anime Neon Genesis Evangelion (haven’t finished it, no spoilers please) there’s AI. Story takes place on Earth after a big almost apocalyptic disaster happens. From what I’ve seen it seems like they built back with AI by developing this AI council called the Magi. When something important is happening on the level of threatening a city, they run a simulation in the Magi and try to make a decision from there. Otherwise it seems like the Magi is running the automated day to day operations of the city. Is ChatGPT on par with the fictional Magi? Not at all, of course. Could you draw a line from all the experimentation we’re doing now to more powerful AI being entrusted with important decisions? I think so. Or at least I haven’t heard an argument convincing enough that better AI isn’t in our future.

The reason I’m even spouting on is because the time to say something about the potential consequences of AI is not when or if the AI gets good enough to start taking away knowledge work jobs. At that point it’s almost moot. The time to have those discussions and decide on potential mitigations is now, before it gets better.

Soapbox over, lol. I just have had these thoughts knocking around in my head without a blog to talk into the void with.

Just don’t ask ChatGPT to solve all the world’s problems and play lazer tag at the same time! Based on the SciFi shows I’ve seen, that could go badly! :stuck_out_tongue:

2 Likes

Develop and implement programs to encourage and support new contributions, such as mentorship programs, coding sprints, and online communities

You had me at lazer tag :smiley:

chatgpt just repeated what it read somewhere. None of the listed inputs or activities are new to the Fedora Project. The activities are very broad and general, nothing specific at all. I think this is quite useless.

1 Like

To be clear, so do I.