For anyone following along at home, here's the issue: I get a different result depending on whether my slicer is in "Between" mode or some other mode. @myersmiguel

Here's the relevant tables. Note that Year[Year] has a different max value (2004) than Time[Year], and this seems to cause the DAX to return the wrong result when a "Between" slicer used

^ @marcorus I wonder if you can confirm whether this is a DAX issue, or as i suspect a bug in Slicers in "Between" mode (and not the first potential bug I've raised regarding "Between" Slicers with Microsoft that has gone nowhere)

Further to this, why does my DAX return an incorrect result when in "Between" mode - unless I change the max value of the table used to populate the Slicer to the max value of the table it is connected to. Note the correct result when I change the Slicer to a Tile layout.

^ Better just to make my measures that are called by the FP conditional. Stand down.

I wish you could conditionally change the measure called by a Field Parameter, based on a Slicer. The below syntax isn't illegal, but doesn't work. ("Category", IF([True Or False],NAMEOF([MeasureA]),NAMEOF([MeasureB])), 1),

If I had the ability to edit that and an audience of geologist data analysts I would have said “I get that there is a gold rush going on, but mostly it looks like a Pyrite of 💩”

In mine it’s often the price. I don’t think Gartner factors in much in my government town. But god forbid someone should edge past MS on the quadrant based on higher investment in Clippy 2.0 I get that there is a gold rush going on. I’m just yet to see the gold.

Damn you Jeff you are right.

TBH if I was in the business of selling #PowerBI licenses I’d probably do the same. But because you mention it in your post @cwebb_bi I feel compelled to ask: Has anyone here met these C-level execs who supposedly deliver organisational value using Copilot for #PowerBI?

I think I’d rephrase it like this: When Gartner have very strong Magic Quadrant criteria and your competitors are rushing to meet them in order to bedazzle C-Level Execs, you fulfil these criteria regardless of whether it truly improves the product.

But this has been my entire complaint about #PowerBI’s visual capability for ever, and it irks me no end that resources are getting applied to copilot for hapless C level execs who will QUICKLY realise copilot don’t deliver, and meanwhile where is my Tableau-level viz capability?

Never mind the fact that there is a circular relationship between modelling and visualisation in meaningful report development m. Heck, if I were assimilated into copilot I would require write permission to *add* to the data model to produce maximum (sometimes minimum) insight.

Copilot’s gonna get in their heads enough to add insight? Occasionally perhaps. I’ll give you that copilot will produce meh reports cheaper than humans currently producing meh reports. But the better reports will be hidden in a sea of meh as C-Level execs race to the bottom.

Im not sure you non-Bruhler types know just how friggin hard it is to build truly useful dashboards with UX users enjoy and adopt, even leaving aside tooling deficiencies. It is damn hard even for experts because you need to get inside users’ jobs, needs, problems, heads.

Did they announce #powerbi audit bookmarks mode yet? Because we’re *really* in the dark there…

I just turned out a pretty good dashboard in 2 weeks with the tool. I’m gonna have to spend the next 2 documenting the hacks that made it good. I can’t imagine how hard it must be to build a principled tool, mind. But I wish the tool’s makers could see how hard I have to hack.

I don’t like commenting here, but have to say that I must have screwed up because I’ve made criticising my mf tool *part* of my identity. TBH I’m quite good at the tool partly bc it’s quite bad. I get relatively good results bc I hack the mf tool like the stubborn mf I am.

You lucky, lucky bastard.

You want a good data analyst? Hire the high-functioning lazy kid.