little-megabyte-1074
gorgeous-dinner-4055
01/25/2022, 1:17 AMoften the people who care about one thing are really disconnected from where it matters and vice versa. So you get kind of these pockets that don't listen to each other very well and are lacking a part of their worldview, a part of the universe of what matters because they haven't seen the other side. I'll give a really silly example, which is the oil market. If you talk to oil traders, they can tell you a lot about the buyers and sellers. So they know all about suppliers and all that OPEC psychology. They really, really understand it. But they actually know shockingly little about this huge thing going on in the world, which is that every government in the world have signed on to say we're going to phase out oil and never use it over the next 30 years. And basically have said, "Your industry is going to be obsolete, but we don't know how."
And then if you talk to those guys, they often sound like climate activists. And they're talking to you about all this beautiful like we're going to make the world a better place. But they often really lack the basic understanding of what are the incentives that people involved in the oil industry. It's not like they're just going to stop drilling oil. Why would they? How is this actually, possibly going to work?
And so I find that both sets of knowledge are there, but they're often not very well connected. And so people in oil often under weigh how much they have a very serious risk to governments taking this idea more seriously and saying, "Wait, I signed onto this. I should really do something to actually get people to phase out of oil."
I loved this quote I recently heard that struck a note with me on what some of the challenges with working with data ownership/governance is. IMO there's a producer and consumer dynamic that similarly mimics the patterns of oil traders vs governments trying to remove fossil fuel usage.
On one side, the data platform teams say, THIS time all data is going to be governed with comments, descriptions, owners, SLAs and validations. On the other side producers of data have a hard time understanding all the moving parts in the data ecosystem to really take ownership of their data as it moves through the platform and tech debt continues to pile up.
Things that come to mind when I think about this problem are:
• How do we make ownership of data easier(can it be point and click vs tons of configuration)?
• How easy is it to communicate with stakeholders?
• What is the impact of the data I am producing?
• When my data is bad, what are the challenges that consumers of data face?
What kind of features can we implement through our datahub project that will alleviate some of these issues? When wee do implement them, how do we validate them?
For the last part, I love your suggestion of:
Start with a small team or domain to test out workflows; iterate until you find a Data Ownership framework that works for your specific organization's structure & tech stack.
little-megabyte-1074