Connell Blackett

Data Warehouse Architect @ Typeform | Doing Analytics Engineering with dbt | Snowflake, Looker and Tableau Certified

[Woojin Kim @woojinrad]

Stop Chasing “Efficiency AI.” The Real Value Is in “Opportunity AI.” - @TDataScience

Companies should focus on "Opportunity AI," which uses AI to solve new problems and create innovative business models, rather than just improving existing processes through "Efficiency AI." AI-native companies can leapfrog traditional businesses by designing systems that address inefficiencies and adapt quickly. By embracing AI's potential for transformative change, organizations can secure a competitive advantage and avoid becoming marginalized.

https://t.co/Bfag5SKoyZ

[Connell Blackett @connellblackett]

@berenguel Do you own a didgeridoo? That should have been the subject of a stand-up demo

[Connell Blackett @connellblackett]

@datavizdad This is my favourite inside joke of all time

[Connell Blackett @connellblackett]

@matthewmmiller It might have worked if it paid for the costs for some kind of identity verification. As it is I really don’t think it’s worth it.

[Connell Blackett @connellblackett]

@FranziCros I much prefer the experience developing locally using Core. I’m fairly sceptical of the value Cloud provides. Airflow scheduling and GitHub CI ticks all the boxes for me.

[Connell Blackett @connellblackett]

@berenguel I had pairs of what they called running tights/leggings in the UK. Not brand specific. Flexible but keeps the wind off

[Connell Blackett @connellblackett]

@datavizdad I can’t believe it took them this long 🙄 I’m sure there’s a hacky solution with my name on it that only works for the first 6 words in the string 😂

[Connell Blackett @connellblackett]

@berenguel @hillelogram I can’t remember the last time anything was smooth 😂

[Connell Blackett @connellblackett]

@ilmariaalto @ergestx I find distinct can be a nice shortcut when you’re using window functions where there’s not an aggregate equivalent, like last_value, and you need to dedupe

[Connell Blackett @connellblackett]

@GergelyOrosz I remember the golden days when Google didn’t allow you to bid on a competitors brand terms, so this was never a thing.

[Connell Blackett @connellblackett]

@berenguel @criccomini @ilmariaalto I’d actually go for the popen at this stage, dbt doesn’t have good external methods to call from python from what I’m aware of.

[Connell Blackett @connellblackett]

@VizWizBI @tableaupublic Does the axis have manual number formatting on it , or is it auto-formatted? I’ve seen cases where the auto-formatting is evaluated differently and I’ve resorted to defining the number format manually.

[Connell Blackett @connellblackett]

@ianwhitestone @sarahmk125 The concepts aren’t as scary as they sound, there’s a parallel for all of them in Tableau, as an example.

LookML = Workbook XML
Looks = Worksheets
Explores = Published Data Sources
Models = Projects (ish)

The learning curve is okay and it’s the most productive tool I’ve used.

[Connell Blackett @connellblackett]

@getdbt After the first post I had to Google KonMari and spent the next hour reorganising my t-shirt drawer instead 😂

[Connell Blackett @connellblackett]

@berenguel We could argue that the majority of human art is derivative too. I just worry that some of this is *so* good that an original artist is robbed of attribution somewhere… but I still love it, reminds me of concept art from old SF novel covers

[Connell Blackett @connellblackett]

@FireboltHQ Outer unions. Matching columns by name at least, and handling missing columns in either part of the union. Admittedly there’s a dbt macro for this.

[Connell Blackett @connellblackett]

@berenguel System requirements: Windows OS

[Connell Blackett @connellblackett]

@berenguel Accidentally dropped production table