iiinova — where your digital self keeps evolving

On the one hand AI has saturated my benchmarks for simple, unambiguous DAX questions. On the other hand, don't turn off your brain.

361

I think the biggest piece of empathy I have for Jensen is in in his world, as I understand it, you spend way way more time verifying chips than you do designing them. And if you don't you spend millions on recalls. So I can understand his frustration some. 3/3

000

Revealing the details of how OpenAI agents hacked Hugging Face

Like, even if you think this is all a regulatory capture ploy by the AI companies and even acknowledging that they don't seem to have done their due diligence with basic sandboxing, airgapping, security etc, Some of the way these agents get around restrictions is REALLY impressive swarmtraces.org 2

swarmtraces.org

When a swarm of 700 OpenAI agents hacked Hugging Face in July, they left behind a public trail of evidence.

100

Perhaps I've drunken the Doomer kool-aid but I was in Ezra's camp listening to the podcast. From my perspective, it felt like he was trying to get Jensen to admit that "alignment", however you define it, is a really really hard problem. And that there are game-theory coordination issues. 1/

100

Posted this on X too: Hey Bryan, I just wanted to say that while I think you and I disagree on some things (Jensen made me furious in his interview), you always come across to me as thoughtful and good-faith. Keep up the good work! I love the podcast as well 😁

000

One or two years ago I install a save as PNG browser extension because I was so annoyed with them. I think support has gotten better.

000

Seen elsewhere: "industry built on modeling humans as rational agents panics as humans adopt rational agents"

010

This is both funny and poignant.

110

I think a lot of folks will move to value pricing. I think a lot will be expected to do more work in the same time and will drown.

210

Eugene Meidinger @sqlgene.com · 26/09/2026

Thoughts on this one? bsky.app/profile/carn...

100

Current status: building a phone app so I can easily get grocery counts for my mom.

010

Reposted by Eugene Meidinger

John Kerski @jkerski.bsky.social · 26/09/2026

Relationships are critical to Power BI semantic models, but are you testing them? Learn how the latest PQL.Assert release helps automate relationship testing and catch issues early. youtu.be/4aIJD9q1Grw

021

Eugene Meidinger @sqlgene.com · 26/09/2026

In their defense, it was one of these

030

Home - Apache Ossie (incubating)

Yeah I work at Tabular Editor. I think the futur eof Power BI is very dicey but the future of DAX and semantic models is good....as long as MSFT doesn't screw it up with this ontology stuff....and as long as they don't get clobbered by Databricks, Snowflake, et al with OSI ossie.apache.org

ossie.apache.org

Apache Ossie (incubating) is a collaborative, open-source effort dedicated to standardizing semantic model exchange across analytics, AI, and BI platforms.

100

Having looked only 10 years ago, it felt like it was good at the filtering and performance. It was a little weird in that it seemed to want everything to be star schema-ish but with bi-directional filtering (their "associative" engine). But that was a decade ago and some basic poking around

210

And there's a lot of things that customers ask for that would be a huge pain in the ass with SQL and is a few lines of code of DAX. Also having the equivalent of clustered columnstore indexes as your DB is very nice for performance in most cases. 2/2

120

I took it as a vanilla versus chocolate thing and didn't feel dismissed. I guess what I can say is based on my experience having been a SQL DBA and then PBI consultant is that when you work with analytic data every day, you do lots of wacky comparisons and business logic. 1/2

130

Justin Bird 🏴󠁧󠁢󠁷󠁬󠁳󠁿 @justinjbird.com · 25/09/2026

The DAX formula language is really really good and manipulating filters and aggregating numbers. It does it in a way that makes it easy to encapsulate complex business logic and make comparisons, say year-over-year or month-to-date. Imagine WHERE, window functions, and some bizzaire UNDO-WHERE, etc

120

Reposted by Eugene Meidinger

RetroData is tomorrow! TEN HOURS of free practical learning. 16:00 UTC+1 Jean-Pierre Riehl on what agentic analytics can learn from decades of BI semantics. 17:00 UTC+1 Edward Pollack closes the day with parameter sniffing, and why the quick fixes hurt later. swfta.uk/retro-beta

swfta.uk

Retro Data Beta, Sat, Sep 26, 2026, 8:00 AM | Meetup

Our agenda is live! 📅 Saturday, September 26, 2026 📅 🕜 8:00 am ⭐️ SUPERCHARGE YOUR DASHBOARDS WITH POWER BI We’ll explore advanced techniques for optimizing data models

033

When you let the AI do the UI design

060

That makes complete sense! He's an authority in the field for a reason

020

I found those big data warehouse books a bit dry and hard to read through from what I recall, but it's been yeeeears

110

Yeah, dumping people into Kimball seems a bit heavy lol

000