All Hands on Data #77
The All Hands on Data crew will be busy watching football and eating turkey next week. These articles should leave you with some leftovers to make it to after Thanksgiving.
The Fall of Data Moats
The author revisits an article he had written a few years prior pointing to the decreasing viability of large amounts of data as a beneficial resources for businesses. Advances in AI/ML model construction have led to less and less data being needed to derive value. Pretty accurate given the meteoric rise of AI as a startup buzzword. - John Forstmeier
Thoughts on data in the Cloud (S3).
This article is a good reminder of how crucial cloud storage is in the modern data team because it offers teams scalable, durable, and cheap storage. That being said, it is important to not just treat your cloud storage as the "dump" for all your data, that will get unruly very fast - Wes Poulsen
4 Ways Effective Data Management Can Boost Business Growth
With the tightening of the belt at a lot of companies, it's important to understand how your data architecture can play into the success (or downfall) of any business. Forbes does a great job of outlining how to be effective in your data management approach. - Angel Catalan
The Danger of Untethered Use Cases
Danger! Danger! There are plenty of pitfalls with a use case-centric approach to data and analytics without first considering an organization's key business initiatives. Organizations often fail not due to a lack of use cases, but because they have too many without strategic alignment. It's important to tether use cases to key business initiatives in order to ensure focus on meaningful outcomes and collaboration between business and data science teams. The Schmarzo Digital Asset Economic Valuation Theorem is a framework explaining how organizations can increase the value of their data and analytics assets through reuse, continuous improvement, and network effects. - Reed Cowan
Predictions 2024: Data And Analytics Set The Stage For Generative AI
"Unstructured data represent less than a third of managed data today. With AI poised to unlock a wealth of text insights, companies investing now in unstructured pipelines will gain a competitive edge." This all sounds well and good, but I believe only the very largest of enterprises will be able to make the necessary investments anytime soon. But will they? With the focus on "efficient growth" and the layoffs hitting data people in 2023, I'm very much in a I'll have to see it to believe it mode. - Shawn Fergus
A Coder Considers the Waning Days of the Craft
As someone who has experimented with ChatGPT to write code and do data engineering work for the past few months, I share the feeling that writing code from scratch may be a dying art. You can solve almost any problem incredibly quickly while cutting out the grind of sifting through documentation and spending hours troubleshooting. Why shouldn't we welcome that future? - Blake Burch