Tweet by TJ
Bunch of thoughts on data I want to write down so I can start next year fresh.
BI (charts with pickers) feels further from solved than ever. Looker is great at 1-2 things but otherwise awful. All the startups are dead. Tableau will never go away. Everything else sucks, nothing is 10x better.
Every data tool is fighting Google Analytics for mindshare and market share. Know thy enemy.
If you’ve been fighting SQL as the language for analytics, now is a good time to reconsider. There’s momentum now that can’t be stopped.
Y’all still don’t know how to count.
If you want to learn to count, understand why we count at all. Look at control systems. Econometrics. DAGs. Causal inf.
Notebooks are a workshop. Production systems are the factory.
Not everything needs to be put into production. Not everything should be a notebook.
You need both. Lean in to the strength of each.
When do you hand-mill vs injection mold?
Excel forever and ever, a hundred years excel.
Cloud costs are high? Learn cost diligence and quit complaining. So bored of seeing complaints with zero effort behind it.
dbt is fine. It’s commodifying data modeling. That is a good thing.
Y’all still don’t know how to model data.
I never want to write a funnel in SQL ever again.
The data industry feels like frontend did 15 years ago. We just got jQuery. We’re about to turn off Flash and ActiveX. We’re almost ready for React, but not quite yet.
I wonder how much ML is motivated by the sunk cost effect of getting clean, reliable data in the first place.
The success of Snowflake lies in its ability to create data pipelines on your behalf with 1/100th of the pain of using spark
Just go work on the org problems already. You’re not different. Your company isn’t different. That tool isn’t going to change anything.
If you ever want self serve to work, you need to make a grocery store. You can’t give folks directions to the farm to pick their own produce.
You think because you pick the produce or run a farm that you know how to design a grocery store. You don’t.