The Fabric Magic Wand – Ep. 458
What Fabric feature would you conjure if you had a magic wand? Inspired by Stephanie Bruno’s LinkedIn post, Mike and Tommy share their wish lists for features that would make the biggest difference in their daily work. Plus, OneLake File Explorer improvements and a new Fabric Data Warehouse migration guide.
News & Announcements
-
OneLake File Explorer: Smarter, More Reliable, and Seamlessly Integrated — Improvements to OneLake File Explorer, making it more stable and better integrated with the OS.
-
Migrating to Fabric Data Warehouse Guide Now Available — A comprehensive migration guide for teams moving from existing warehouse solutions to Fabric Data Warehouse.
Main Discussion: Magic Wand Features
The Concept
Stephanie Bruno’s LinkedIn post asked the community: if you could wave a magic wand and add one feature to Fabric, what would it be? Mike and Tommy run with the concept and share their own picks.
Wish List Highlights
The conversation covers features that would reduce friction, improve developer experience, and fill current gaps in the platform. Key themes:
- Better debugging and observability — Understanding what’s happening inside the engine when things go wrong
- Improved collaboration — Real-time co-editing, better conflict resolution, smoother team workflows
- Performance transparency — Clearer insight into what’s consuming capacity and why
- Simplification — Reducing the number of steps and clicks for common workflows
- Cross-workload integration — Smoother handoffs between Fabric workloads without manual wiring
Community Wisdom
The value of this exercise isn’t just wish-listing—it reveals what practitioners find most frustrating about the current experience. Mike and Tommy encourage listeners to share their own magic wand features with the Fabric team through feedback channels.
Looking Forward
Microsoft has been responsive to community feedback (UDFs, TMDL, calendar time intelligence all came from strong community demand). The magic wand features of today may well become the announcements of tomorrow.
Episode Transcript
Full verbatim transcript — click any timestamp to jump to that moment:
0:00 Good morning and welcome Back to the explicit measures podcast with Tommy and
0:32 Mike. Hi Mike. How you doing? I’m doing well. Sorry. I I usually do a little bit more talking in the intro. Tommy, I realized I just stopped talking. You’re like I was like, “All right, I guess I’m now I guess I’m starting. All right. To me.” All right. Good. Sorry. We’re starting a little bit late. We Tommy and I were talking beforehand and having a fun time talking. I totally totally lost track of time. This is what happens when you have two ADHD people running a podcast. like we’re just it’s not the best, but we’re definitely going to have fun with it and have a
1:03 Lot of fun and enjoy all the conversations. So, that being said, our main topic today is an article. , you may have guessed who the article was by just based on the last couple episodes we’ve been doing, but Stephanie Bruno is has a really fun post on LinkedIn about if you had a magic wand and would ask for something from PowerBI team, what would you build? What would you create? And so, , we’re going to go through her article and actually I want to just review all the different features or just talk
1:35 About the things that people are asking for inside the chat that comes through on that message. Just want to chitchat about those things. See how that happens. So, that’ll be our main topic today. The fabric magic wand. , let’s jump over here to some news items before we get going. Quickly before we get going here, I do want to pay homage to two events coming up. There is the Chicago event, the PowerBI user group coming up on October 2nd. Tommy, do you have more information about that? October 2nd. You’ve probably heard us talk about it enough by now, but sign up
2:07 In person downtown Chicago on October sec 2nd is the fabric user group and we’re going doing Chicago’s crash course on Microsoft Fabric. We’re going to take advantage of being in person. You’ve heard it already. Mike and I will be there. There will be Jimmy John’s. Make sure you go to the meetup link, sign up. We’re really excited to be there. So, should be super fun. looking forward to getting the user group back up and running again, particularly in Chicago. , I think it’s about time to get out of our homes, start meeting people, and doing some inerson events. I
2:39 Think it’ll be very fun. So, make sure you use the link. , I’m going to put the link here in the chat description right now so that way you have it. , this is the meetup link. You need to make sure you sign up. It is at the Microsoft AON building and that means you need to sign up because they need a list of people and when you show up you must have an ID with you so they can verify you are the right person and they’ll let you into the building. So make sure you sign up on the meetup link and then make sure you also sign in. There’s a little bit of a coverage cost on top of the event. So that covers just to make sure that you show up, you
3:11 Committed. and then also I want to plug another one here. Pittsburgh SQL Saturday is happening. it will be on October 18th and there is a also a precon day October 18th this year and it’s an in-person event and there is a precon day of a fabric in a day. So I definitely want to plug that one as well. Another great event coming up that we really want to support. The link is also in the chat as well and I think that’s all the details for events. Tommy, give us some news items here. let’s talk about what’s what do you
3:44 Got going on first article? We’ll we’ll we’ll get our feet wet. And the first one is update to the one lake file explorer. And Mike, do you use the one lake file explorer at all? This is a Windows application. Yeah, I was going to say this is this is basically one drive for one lake. Correct. it’s like it’s like a little Exactly. So I do use that and I actually find that there is no way to download a file from the one link experience inside the browser which I find actually a
4:18 Little bit frustrating cuz sometimes I’d like to just download the file right from the browser and then work on it and then just push it back up. So I wish I wish that was there. I will also admit you can edit files in one lake which I found which is very interesting. So if you look if you have like a JSON file or some definition of something, those things exist in one link and you can edit those files. So that’s fun. but yeah, I do use this tool. I’m happy that it’s getting an update, but it just feels like simple. It’s just like one drive and just Okay, fine. What do you think, Tommy? What’s your
4:49 Opinion? the the idea between the one lake file explorer is really the ability for people if they’re adding images they’re adding excel files quickly to the one lake and adding that and this is for teams working together I think for you and I sure maybe for an individual but this is this solves the hey put everything in SharePoint and PowerBI will connect directly to SharePoint this completely solves that. So there’s a lot of opportunity. You can
5:21 Do I believe your Python packages as well, wheels. So for teams working together, this is pretty essential. The only thing I just have a little bit of hesitation around this one is cuz I’ve already got one drive. Now I have one lake. Now I have two like applications sitting on my computer that are just synchronizing files and things back with my machine. I feel like as a little extra bloat that I don’t really want all the time running on my machine because there there are times when if I’m uploading something big or doing something large, it’s taking a bit of time for me to get that file up to
5:54 Those storage accounts and sometimes it feels like it bogs down my machine. Again, I’m talking more for like of the one drive perspective. because I have some pretty large files that we move around for the podcast and those get synchronized through one drive. But in general, I like it. It works pretty well. I’ve had good good experiences with it. Yeah. One of the this has been on already but I think for a lot of people consultants it’s one drive is a pain to try to sign in different accounts you can do this in the one lake file explorer there’s some smarter sync for your temporary files and they just work
6:26 On some stability improvements so nothing feature-wise but just again I don’t need this to have a really cool UI just make it sync if I add something make it show that’s what this is here awesome. I like it. the other article we have here and oh I just lost my notes. the notes here is the second article is migrating to fabric data warehouse guide. So it’s it’s migrating to a fabric data warehouse. There’s a guide that is now officially available. So Tommy, what do you think about this?
6:58 This is this is just generally, , get yourself into fabric. Here are some migration steps, an assistant, a strategy, and methods to get out of synapse dedicated SQL pools and migrate Synapse stuff into fabric. What do you think about this? So, Mike, this is right off our conversation with Brad a few weeks ago, and it’s simply just we had the migration assistant. They’re coming in all shapes and sizes now. , especially when you’re dealing with all the different cloud sources. But this is great because you actually have three
7:31 Great migration resources for the fabric data warehouse for planning for Synapse dedicated SQL pools and again for Azure Synapse dedicated SQL pools into fabric. So this is great again I think this is just the starting point to get this a a one-click solution but obviously we know that’s not always going to be the case. So the more resources the better. And this goes off our conversation to me from last week about capacity planning. So awesome. Yeah. So I think this is really good. I I’m I’m enjoying like all these
8:03 Different tools. I will say this and at the end of the day whenever I’m bringing more items, let’s call them items or systems or processes. When I bring them into fabric, I’ll say I really substantially enjoy the integration of all the things that can talk to each other. It’s actually pretty seamless to go from, oh no, I need to get a little bit more data here. the other day I was just trying to rebuild a SQL server. I went to the old SQL server and then I didn’t know how to do like a DAC pack or extract the old the SQL information. So I literally just went to the database,
8:34 Ran queries from the tables. It was only a handful of tables. I ran the queries in the tables. I scripted them. Create as went over to fabric. Just ran the create tables, did a little quick data factory pipeline, and then uploaded a little copy data. Hey, here’s the exported data from the old thing. Import to the new thing. Boom. Data just works. and my my whole report just like lit up and the server was moved. It was it was very easy, very seamless. , and the fact that everything works together. I could write the data in using a pipeline. I could access the data using a notebook. I could go to the lakehouse and look at things from there.
9:08 Like it it was helpful. I will say I really I like having all the tools in the same place. This feels like what Azure should have been doing, which is make everything talk to each other. And I think this is a platform where they can really rethink that and make sure everything does in fact talk to each other in a sim se seamless way. I really like it. Anyways, really good there. Check out the article. The article is in the link description here. I’ll put that out here as well that if you’re looking to get into fabric or looking to migrate other things, Synapse or other solutions directly into fabric, you can see that article here as well. I highly recommend
9:40 A read through. Thank you, Priyanka, for making that article. All right, that being said, Tommy, , let what do you say we jump into our main topic, unless you have any other news items? No, I’m excited for this one. All right, we have a caller. Welcome, caller. , caller number one today. Caller number one, we have you here today for our call. Welcome to the Explosive Metric podcast again. We really appreciate having you Welcome back. Thank Thank you. Thank you so much. Hello.
10:11 And thank you for taking my call yet again. I feel like we’ll always answer the phone call, bombarding you with phone calls these days. So, thank you. It’s really fun to be here with you again. Excellent. Well, our main topic today is going to be centered around one of Stephanie’s posts. I think Stephanie, you did a great job of just teeing up a very I think I feel like on the community, you just lobbed a softball over to us here on LinkedIn. And with like always the community is very opinionated about what it what it wants to talk about and what features they like and what features they don’t.
10:43 So give us a an overview. What was your what was your statement? What was your initial post here about and what was your idea? Let’s start there and then we’ll go from from there to what other people said on the on the post as well. Okay. So like all of us whenever you’re busy building in PowerBI you’re just thinking ah why do why is it like this? there’s just this one thing that I want and it would make my life a whole lot easier. So, I assumed everybody had this feeling, right? So, I just wanted to know if you could wave your magic wand and conjure one new feature in PowerBI, what would it be? , and the funny thing is is that the feature that
11:16 I wanted actually is already there. It’s just that I can never find it. It’s like my brain will never it just will not stick in my brain. And what it was for me is just, , sometimes I’m I I’m working on a thin report connected to a model and maybe I have a different version of the model or whatever, maybe dev test. And so I just want to swap it out really fast. Yep. Every single time I click on one lake the one lake button there. Yep. Sure. In the get data ribbon and then it says would you Okay, now you have to add a
11:48 Local model. Ah, that’s not what I want to do. I just want to swap it out. So, of course, the right way to do it is in the the data source settings, but I never remember that. So, for me, I guess what I’m asking is for my own brain to actually have that stick. But fortunately, everybody else had had some great ideas. So, I think that’s what we’re going to talk about. Well, , again, let’s let’s be really clear here. Like, it’s it’s not good design. , if so, my my issue here is like if you’re looking to do this and you continually forget where the button is to do that, it probably
12:20 Should be a bit more discoverable in general, like I’m in this report. I’d like to exchange my connection. There should probably be a bit more menu or another item that says exchange connection or an item a button that’s on the ribbon that’s, , hey, you’re one of the things that’s also very difficult to understand. I I’ll add to your feature here is how do when you’re in a live connected thin report, right? It’s actually you have to go to the very bottom right hand corner and there’s this very tiny line that says live connected. So for users who are newer to the program, how would they know like how would how to
12:53 Switch things? You would just assume oh shoot I just got to rebuild the whole thing or dig around. But to your point that should be highlighted more that should be I think that’s a common use case and that should be a bit more visible on the main ribbons. So I’d unear it. All right. Thank you. Thank you for making me feel a little less embarrassed about feature that already exists. Man, if we only had this pipeline thing that could make data move across these I I have a question. How often are you wishing for a magic wand when it comes
13:25 To when you’re working in PowerBI? Is this coming from a sense of frustration? Is this like after seven times I’m going to, , pull out, cross my u cross my fingers and hope it turns out well or when does this come about? Because I think we all have sometimes that frustration whether it’s a UI change or it’s like man if this could be a certain way. So yeah, love to hear. Yeah, and I really like hearing from people who are newer at it because I think, , with us, we’ve been
13:58 Using it for so long, you just get used to how things work and, , then you start to to know the quirks and it’s harder to think about, oh, this it would be great if this one thing was a single click instead of these five, , that I know by heart. , so yeah, I really like hearing from people who haven’t been using it quite as long. , they they’ve got a great perspective. I’m going to I’m going to say that when whenever I’m trying to teach someone about PowerBI, it feels like these feelings of I wish it was smoother were heightened for me, right? So, when I’m
14:32 Preparing class material or I’m preparing like a demo or I’m working on something, I’m always thinking through like trying to explain it to somebody else. And for there’s something maybe maybe I’m weird. I don’t know. Yes, I am weird. , but maybe there’s something therapeutic about being able to explain something to somebody else, but you really have to boil down what you’re doing in a way that someone else could understand like the sequence or the steps, right? So for me, I feel like a lot of these feelings of man, I wish this wasn’t so much frictionfilled or man, I wish this
15:05 That was my feeling around when I built the power designer andor theme generator because I was like, man, working with theme files could be really powerful, but it just feels like it’s so much effort to do anything. And so I was like, I this has got to be better. , and we’ve got to improve this experience. So, a lot of a lot of my frustration is born out of like trying to let other people know about a thing that I’m trying to do or produce. And in doing that, I find all these rough edges. And I think us as MVPs, we’re just more vocal maybe than everyone else. Everyone else might be thinking it and being like, I have no voice. I’m not going to do anything
15:37 About it. But like, I think as MVPs, we’ve already made the acknowledgement that we’re loudmouthed. Like, we’re going to say things like whether we like it or we don’t like it, we’re going to we’re going to let people know with feedback like here. And one of the things we try and do in the podcast is if we come up with a a friction point, we always want to try and come up with a solution like always be like on the like yes there’s maybe friction there’s no matter what tool you buy or build or whatever there’s always going to be a parts of like ah that was awful right they could improve that but then I think always coming back with at least a
16:09 Resolution to say if I were building it or how we would look at it how would we solve the thing better and I like that that positive spin on it yes it’s got friction but what’s the opportunity that exists here to fix or make it better. Awesome. Well, with that then, Tommy, let’s go through some of the items here. Is there any features? There’s there’s a lot of comments on this one. So, we’re going to just roll through here and pick out some of the the , the comments that had new features attached to them. I’ll give it to you, Tommy. Let’s go through your stuff. What do you think first is one of the other feature requests that came in the comments here from this post that you
16:42 Want to play on or pick out? It’s interesting when you when you put this out to the social sphere what would people ask for? Would it be bugs? Would it be design? And it looked like by far and away the most requests were around developing and authoring specifically and visuals. I think visuals a lot. Visuals and charts came in at a solid number two. but developing and authoring. Yeah. So how can we actually have templates? And that was a huge theme. theme that was a huge shrine
17:14 Throughout story. Yes. Yeah. The story throughout this was how can we have better templates and components around what we do in PowerBI reusable things and because that Mike outside of the theme generator that you have doesn’t really exist does it? there I’ll be very frankly honest like there’s some multiple theme generation type tools out there but they really just do like the theme thing only and they really and they’re also very expensive honestly if you look at the just price comparison like some people
17:45 Give this stuff away for free. I try and give away a lot of the core part of the program for free and we have some premium features that we want you to to use and help us pay for cuz I’d like to fund more features. So that’s my way of like, hey, if you like if this is saving you time, please help us support the the the furthering of create continuing to creating new features in this tool. We we’d love for you to like, , help our community build a better tool and that’s why we’re trying to keep the price really low. But most tools I see out there are like $8, $10, $12 per month just to do theming. And I think to myself, dude, I can buy PowerBI, all of PowerBI desktop and everything in
18:18 PowerBI.com for $14. Why would I want to double my price just to get a theme generator to stylize some things? So that’s that’s why we price things at around $2 or $3 per user for using our theme generator tool. But you’re right, Tommy, there’s really nothing that exists. And I’m now starting to see tools that exist that are like wireframing. We’ve had this for years already. we templating, we’ve got templating already as galleries that’s already existing. So I think you’re right, Tommy. There’s a lot of One of the main feedbacks I’ve always heard with PowerBI is I’m good at we us I I’m
18:52 Respect for all of us. We’re all really good at shaping data, getting tables together, dropping in a little DAX. We can do that. It’s when we get to what does the aesthetics look like? And I feel like that’s a slightly different eye or or study or field where I just enjoy the numbers. I don’t really love I enjoy the the the graphics side of it too but that’s something that we don’t study I think as much as getting the numbers right because at the end of the day if you don’t have the numbers right you might as well not even report anything right you just stop
19:26 Right there because that doesn’t help anyone but on the flip side of that I also think you don’t drive a lot of engagement with your reports if they don’t at least look pretty and function well and so now with translitical being added to like the you can make a full data application where you’re reading data and then you’re pushing data out and you’re getting data back and like you really do need to think about the whole user experience the UIUX experience and I think a lot of things here Kurt pulls this out as well Kurt talks about using visual templates or a better way of having a component system like let’s
20:00 Imagine putting three or four visuals together to make a group of things that you just drop on the page as a group and it just always stays together like I’m always going to put this background I’m always going to put image and then I’m going to put this visual on top of it. Make it a a collection, an item as a dedicated element that you can reuse. , and we’ve been doing a lot of exploration on that front. So, we actually have projects with clients right now that are doing really heavy report automation. So, stay tuned. That’s probably going to come to part of the Power Designer product in the near future where we’re going to be more heavily automating your reports. Think of it like have you guys both used
20:33 DAB? A little bit. Yeah. Okay. In DAB, you define like it’s a JSON. I don’t really love the fact it’s JSON, but you define a structure in JSON that describes what you want. Hey, I’m going to make a mark. My mark will be a bar. This bar has a height. This bar has a width. This bar has a color. There’s like properties. You basically bind data to properties. It’s called a grammar a grammar language to build visuals. You just describe what you want basically. And then the code interprets what you need into an actual visual. Why can’t we
21:08 Have the same thing, a grammar visual for reports? I want a page. We all know what a page is. A page has a width. A page has a height. Why can’t I just speak to the tool and tell it what kinds of properties I need it to adjust and it just works? So, stay tuned. We’re we’re working on things like this. And so why not have heavy report automation using these report grammar language designs? So we’re actually actively building on that. So that’s I think a really good point here and I think that’s a really good feature. I’m not sure Microsoft will ever really build
21:39 That though, right? We talked about this before about how I Microsoft does depend on the community for for some of the tools, right? So they’ve said we’re not going to build that because we know that our our community is doing it and they’re doing a good job. So I think some of those things we might see from Microsoft, but some it’s it’s on us. I’d rather have them focus on again going back to like what is Microsoft good at? Microsoft’s really good at being like slow and methodical and building like the scaffolding, right? I don’t want Microsoft really to come in
22:11 To like put the furniture in the room, right? I don’t want them to really come in and decorate it cuz my my preferences may be different than theirs. I may go minimalist, right? I may have lots of, , warm clothes or warm, , lots of fuzzy stuff on the floors and carpets and like someone else may just do minimalism. Like I don’t really want Microsoft to come in and tell me this is the design you get. I think we want the framework. I want them to give me like the scaffolding of the of the building and then I’ll come in and decorate how I want. And so I think this is part of where the community can come in and build the things that they feel like the community needs. And then Microsoft should really give space to
22:44 Allow the community to really step in and fill these gaps. And then at some point I think this is a better it’s a better product for the end user anyways, right? Microsoft could let’s imagine I’m just going to spitball here for a bit. This is this is going to be an episode filled with lots of tangents and randomness. All right, here we go. Would you would you my my question would be is would you would would you like to have Microsoft add $10 to the product and give you these features? Right. So $14 a
23:17 Month. Microsoft says no problem, Tommy. We know you want all these really fancy visual things you want. You want components. Not not an issue. How about we add $10 to your product and we’ll give it to you. And I think most people would be like, I’ll just deal with it. Like I’m okay. I won’t I won’t act. Nah. what? On second thought, how about not? Like I think I think some people would actually pull back some of their requests a little bit if Microsoft just said, “I’m just going to add money to the product to to add developers to get it done and have that done.” But here’s where I think the community and
23:49 Particularly people who use PowerBI, this is where the free market, I think, makes a lot of sense. Right? Again, my space of theme generation, there’s lots of tools out there that do it. Some of them are 100% free. Some of them are supported and pay money for them. I think at the end of the day, the tools that are the right price with the right value, they’re going to bubble to the top and be like the leaders in that space. And so that’s what I’d like to compete in. I think I’d like to think of, , put your money where your mouth is community. If you want this, go buy the products and the tools that are helping you build that
24:20 Direction. That way, they can build better stuff for you. So, I I think it’s fair. And then I may not want to pay $10, but it may be worth it for me to pay $3 or $4. that might be the right balance for what I need for extra features. So, I I get it, but we can’t , it’s unsustainable to have Microsoft say for $14 a month, you get everything you would ever want and we’ll always build all new features and it’ll it’ll never go up in price ever. That seems like you’re going to be building more code. It’s going to be more costly to build and run and do the things. At some point, they’re going to have to push the price up a bit if they keep
24:53 Doing that. Anyways, that’s just my thought. Tommy, what what’s your reaction to that? I think you were going to say something. So, I I to a point I’d agree with you. Of course, it’s never going to be the perfect product and it’s going to have major updates every week. And yeah, maybe we’re also spoiled at first with PowerBI with the monthly updates of 10 years ago, but honestly, yes, you can build your own. It’s open source. And I know you you love saying that because Microsoft is is more much more open source. say they but with the fabric environment Mike heck even with PowerBI and the external tools that was
25:26 Great for a single user but the workflow still wasn’t seamless it wasn’t really integrated with your daily workflow I use external tools all the time but with templates Mike and trying to have these integrated assets that are unique to me Yep. It’s it works or it worked in PowerBI desktop but it wasn’t something that was part of the tool the application made it
25:58 Much harder and this is something we’ve been asking for for a long time and I think this will be more of this with fabric with notebook templates be able to customize that I don’t think that’s a major ask and you can’t really do that yourself now in fabric well I want to maybe point on like yes and no right so workloads give you the ability to drop stuff in automatically and that’s one of the the advantages that we get with our power designer which is the themed gener re revised right this is why I’m so big on this
26:30 Workloads are amazing they’re incredible if you’re any company who’s trying to share data with another company who has fabric you better be looking at workloads call me if you need help right so I I think this is an amazing I so this is one of the things that Microsoft has opened the door to where organizations can step in and say I’m going build a proper data product that will allow me to share easy data shortcuts. , it’ll allow me to share reports and semantic models. You can autopublish things directly into customers tenants for them. So, and you can shut off the data stream from
27:04 From a permission standpoint as well. So, there’s a whole bunch of really rich features that Microsoft is now opening the world up to with workloads. And I think this is this is huge. , I don’t see I don’t see data bricks letting customers build things like this. I don’t see Tableau letting people build like custom things like this or even snowh snow snowflake for that matter. Maybe they have something I’m not aware of it. They they could, but the the fact that I could embed my app directly inside of the fabric product and Microsoft’s giving me a handshake of like we’re going to trust you. You’re going to be able to do things on behalf
27:36 Of the user inside the workspace. That’s huge. And so, , it’s stuff like that, Tommy, where I think like, yes, we were in a world that this was like hindered, but now we can have all the things, right? SAP can show up. , SA, SAS, SSAS, whatever the the statistics program is, like they can they can show up. They can build their own data product that’s around their stuff. , , Oracle could show up and do things with what they’re doing. Like, who knows? I I don’t know what people are going to build, but this is the idea of like what is going to be most valuable to the end user. It’s up to now these companies to decide what’s
28:08 The level of investment they want to produce. And now to me, this makes PowerBI just that much more robust. You you have all the artifacts that are coming from Microsoft notebooks, pipelines, those things. What if someone comes up with something new with vibe coding? I don’t know, maybe someone comes up with a vibe coded solution where you just chat with it and it vibe codes out some data engineering pipelines and runs its own thing. I don’t know. Something more efficient may come along. This is going to allow Microsoft to adapt to that. You can use it whatever you want. So, I think it’s I think it’s really good, but I think it’s just changing is what I would say. Let me pause there. What do you think?
28:41 I still believe there’s still only six workloads in PowerBI, if I’m not mistaken, or eight. So, more now. Yeah. And I’m going to be adding hopefully two more in the next couple of weeks. That’s the goal. So, we are full steam ahead on making sure that PowerBI is very equipped with really cool products and neat tools and visualization things. So, stay tuned. There’s more more to come. So yes, and I I think that’s a great point like we can split these feature requests out into things that we as the community can build but there are
29:13 Some that we can’t right like some things are still just a part of the product that we need. So like some a lot of the core visual ones is what I’m thinking about and especially from the perspective of somebody who’s maybe again newer to PowerBI they want certain basics in the visuals they don’t want to have to go look elsewhere for it. oh man I love what you’re saying here. This is great. This is this is you’re hitting on all the things that I friction point with the team right now. Keep going. Keep going. I love this. Let’s go. Keep going. Well I was thinking like there was
29:45 Something I wanted to do and I can’t remember exactly what it was. I think I was trying to do conditional formatting. I wanted bars in a matrix, but then I also wanted to be able to change the colors of the bar based on a measure and I couldn’t do it, . And then I thought, oh, I’m going to get tricky and I’m going to go into the PBIR format, , and I’m gonna I’m going to copy conditional formatting for color and add it to the bar. And I and maybe that’ll work. It didn’t work. It just broke the visual. So, , that’s something I just could not do. And so even with, , community
30:17 Tools, I don’t think that’s something I could do. So I think to your to your point, well, I’m going to challenge you there, right? So I think your way of thinking about things is good. , but I would I’m going to you can do this, but you have to go towards SVG as a measure and then you just hit a big cliff because now I’m like, okay, it’s possible, but to your point, right, it’s not like part of the tool. There’s not an easy UI to get you to a place where like oh I in your mind this is and this is where large language models should help us honestly right this this is exactly
30:48 Where this should sit right Mr. language model I’m going to build a measure the format needs to look like this and I’m going to bind this data to this bar and the color of that bar must be this build me an SVG bar chart that has these things and this is the input that I’m going to give you. So to your point, right, it should be easier. Not not disagreeing with you, but here’s where I think I want to push a bit more on the community tools, right, when you open up things like workloads, why do you need to build a PowerBI
31:21 Report with a matrix in it? Like maybe maybe the answer is using something more like DANB, which does give you that level of flexibility. Now again, it’s very coddriven, but what if there was a UI that helped you get to that starting point or helped you get to that next level of things that made it a bit easier for you to use these different tools? And so I think my to your point, my biggest gripe with the core visuals is if Microsoft doesn’t build it, it can’t be done. And so there’s all these workarounds you have to know about with core visuals to get them better. I really feel like Microsoft should have
31:53 Embraced the the Neb and Vega and Vega light type mentality of like look there’s a spec out there we’re going to add a visual. the fact that Microsoft didn’t go by DAB and just drop it in as a core visual is like beyond me. They they should have just done this. Like that could have been like the easy button win of like any new visual that you I think literally the DAB visual by itself could kill the entire app marketplace for all visuals. I’m I’m like 90% of them could disappear if you had to n because it can build everything. It does all the stuff. And can you add a slider button to a visual
32:28 In anything core visuals? Like literally like put a number, , hey, I have a bar that you slide up and down that adjusts something on the visual. Do we have that other than the x or y axis? No, this doesn’t exist. So the idea is like the Neb and the the Vega visual itself allows you to add like specific data controls on that visual with like that’s the stuff we want. Like that’s can we make that easy? How can we get this easier to put in people’s hands? So and the capabilities are really high. It’s just that we’re limited by the team and what they’re
33:01 Allowing us to do inside core visuals. But I don’t want to discredit your idea, Stephanie, cuz I I I I think it’s a brilliant idea, but I think we’re on the the cusp of something really changing here. and now that Microsoft has opened the world to other visuals or other workloads, I think we’re actually to a place where right now someone could really come in and disrupt some of the existing visual market because of that that’s that’s that pattern. I I am going to refrain on the db. So just say I just want to say for the note I disagree because then it’s going to be
33:35 Why is DB we’re going to have to change the title. No, go ahead. No, Tommy, go ahead. I want to hear I want to hear it. Let’s This is where the goodness comes in from the podcast. We argue about things. Okay. The craziest thing you’ve said in 460 episodes is you think that DB can take over 400 of their custom visuals. No, hold on. You went wrong in episode one. All right. You got you got one thing right think maybe two. Maybe two. There are a lot of things you say that I do agree with and
34:07 Sound logical. This one though with Danb. Yeah. What? What happened to Charticulator? Go watch your 10 episode Chartic playlist because that was going to be a thing too. Danb is great. Yeah. Shots fired. No, not Shots fired. You think Chartic is dead. I’m bringing it back to life. Don’t worry, Tommy. It’s it’s coming back with a vengeance. This is going to be like , Terminator 2, , with with Tarticulator coming back with like more power, more more
34:40 Everything. You There is my bad. There’s things on my idea roadmap, Tommy. You will eat your words. Keep going though. Let’s keep going. So, DB can at the final output do to your point a lot what the custom visuals can do. However, now you’re expecting everyone to be a de true developer working in JSON and the DAB is awesome. It’s an amazing tool, but that is not for the faint of heart. especially in the environment that
35:12 It’s in the from the UI point of view, the documentation the documentation is incredible, but my gosh, you need to understand some concepts that if you’re coming if you’re not coming from web design, you’re not going to understand it at all. And to try to make those small changes for you cannot expect maybe 90% of the visuals can be solved, but 90% of the people cannot. And this I’m going to challenge you here. Okay, I understand. I I I fully want to
35:43 Acknowledge your point. I agree with you, Tommy. Writing JSON and reading the the JSON visualization spec for Vega and Vega Light difficult. I would agree with you 100%. But are we smart enough to figure out how to uncomplexify that spec and make it clickable and easy to use and other I think the answer is yes. I think the answer is yes. And I do I think there’s there’s a tipping point here where large language models are shocking me. It it’s changing how I want to work with things. So even in so I have no fear
36:18 Going into any Python notebook now and be like I can data engineer anything I want now. Like I know a good amount of Python but I have been learning a lot more Python by just saying oh shoot I can’t get these two table things. I have two data frames. I got to get them together. Hey chat, co-pilot, whatever. How do I take two data frames that look like this and put them together? And it says, “Oh yeah, you need to use the the function called MEL. Didn’t know MELD existed. Okay, let’s use that.” Boom. It works. Happy. Like so there’s there’s a there’s this layer of we’re on I don’t think we understand
36:52 How much AI and large language models is going to take anything code related and make it just that much easier to build on top of. So it’s going to get better. So all this to say is Tommy I hear your point and the way it looks the world right now I would agree with you probably not the widest adoption but how can we change that? How do we influence the size and scale of like danb visuals, custom visuals? There’s a whole new world here that we just need to dream up and it becomes easier. And I think with
37:25 Large language models, this really changes the game. , , I was we were talking earlier about report grammarss, a grammar for the report building. We’ve already developed one. We’re using it now in some projects and that will probably be coming to power designer. How great would it be to talk to your AI and say, “Build me a report with three pages that has this, this, and this on it, and this visual here, and this visual there, make it look pretty, and then it just does it.” So, I I really do think if you simplify the technical aspects of some things and put a little bit of an abstraction layer in front of it, there’s actually some
37:57 Really good large language models that can build you really good visuals. I think to your point, Stephanie, we should be have enough examples of things to say, I need a bar chart inside a table looking matrix thing and I want these bars to have these colors. That should be easy enough for a large language model to go figure out how to do it and come back with an answer for you. Actually, I would actually would encourage you go into a large language model, go test it out, see if you can have it build a SQL SVG measure, sorry, an SVG measure. Maybe that can do it. I don’t know. Like so maybe you if maybe
38:31 If you give it a simple bar chart or bar SQL SVG measure and say here’s one that I found on the internet. Can you adapt this so the bars are changing color by this and see what it does and maybe it actually gets you the answer that you want based on the data that you’re inputting. I don’t know. So anyways I’ll just pause right there. What do you think Tommy? We are getting there but I I think we should probably get back to some of the other feature requests in the next 20 minutes. So totally fair. Totally fair. as as much as want to dive in. But regardless, Mike, I think you was still bringing out
39:02 Oh, by the way, there are currently 12 workloads available or 13. Oh, yeah. Really? So fast. So fast. So fast. You never know. So fast. , let’s kick it back over to Stephanie. Let’s Stephanie, what do you what do you want to any thoughts on that point or you want to move on to another another topic on the another idea on the post? Both. Lots of thoughts. I think I agree with you both because I do think I’m totally with you, Mike, on the notebooks. I used to be way more intimidated by writing Python, but now like you, I feel like I can make it do
39:35 Whatever I want. This is great. Yeah. So, I think same same with with visual design. but also, like Tommy said, I just don’t think we’re there yet. Like I knew I could have done that in SVG, but I was working with a customer and they didn’t want to deal with, , the complexity of a longer piece of DAX to create that SVG. , so we were trying to find something that was just going to be simple in the product. So I think where we are today, like you said, , it would be great to have more of these features available. I hope that Microsoft doesn’t, , give up on
40:08 That and just wait for us to be able to do them all with LLMs because I I do think there’s value in in putting these in the in the core visuals at least at this point. So that’s my two cents to agree with you both. Thank you’re very you’re very diplomatic about how you went about that one. That was good. Like Tommy, you’re totally wrong. Mike, you’re an idiot. Stop doing that. Awesome. let’s go through some other items. Any other items? One that stuck out to me which I think I find some friction around is someone
40:41 Talked about decent lineage. Hey, I want to take a column that’s used in a measure or like go both ways, right? I want to go from this visual and I want to click on that visual and say on this visual. How can I see the lineage all the way back to source? Right? We know all this exists. There’s columns being called out from like bronze through silver to gold. There’s transformations that are happening. There’s lineage that’s happening there. And then there’s the semantic model. We know where that links to. And then we now have the report. Dog on it. Everything has an ID
41:13 On top of it. Like the measure has an ID. The measure has things in it that have other items that have IDs to them. I’m like, why can’t all this just be mapped out and easy? Now, someone does have a solution. Chris, I believe, has a solution to this one, which by the way, I’ve been talking to Chris. His solution is pretty dang sweet. And I would like to demo it at some point. So, stay tuned. We’ll talk to Chris at some point in the future. But, , it’s not built into the product like we’re saying here. This is like not a product based thing. And I would agree if I’m going to modify a semantic model or I’m going to modify a report, sometimes I just need to know where did this stuff come from, how did it get here, and that would be
41:45 Very helpful. What am I going to what am I going to break downstream? Yeah, totally. That’s the that’s the big one. I think some Semantic Link Labs has something also similar to this where you can say here’s the report, here’s the model, and it’ll like map some things together for you. So, I think some of that exists. , well, and I put that into the fielder, but it’s it’s a little hacky. , it’s my own tool, so , it doesn’t always work. , I would love to have something right in desktop or in the service that says, “Oh, this visual or this measure, , has all of
42:20 These downstream impacts.” But this is this is the point of like what if you got paid a little bit of money to help support that and just you could spend like an hour or two a week just like imagine the community coming back and supporting you in a way that lets you continue developing the tool that on the edge the periphery of what Microsoft’s developing. This is the point where this is my my initial point earlier which was like the community should rally behind things that they like to use and they should help out contribute back to what you’re doing there contribute wise so you can keep doing a little bit more of it making it better. Well yeah and I think Gregor did that right with measure killer. So I think
42:54 Measure killer does a really good job of that and that’s I think he’s got a free version and then he’s got the paid version. Exactly right. So I think that’s a great example of the community rallying and supporting that. I would totally agree there. What do you think, Tommy? What What about decent lineage thoughts? , listen, I I’m pretty happy with Lineage right now. Everything else is I think is a license for purview. there of course there’s a lot more we can do with Lineage. I think honestly we’re that’s just the game we’re playing in where that’s always going to be a
43:25 Complicated endeavor. Anytime we have our dependencies between models, tables, visuals, it’s just something that we have. I don’t think there’s going to be a magic UI to fix that. I did actually want to talk about the a few others on the list here just to make sure that covered them unless there’s anything else there you wanted to do. I think down the visuals route I do want to call out the one the person who said we need a Gant chart. Yeah, we do. Sure. We need a good G I would say we need a good Gant chart. Like I the number of
43:57 Requests I’ve gotten from customers about hey what’s a good Gant chart I should use and all the ones I’ve ever found were like meh at best which is then I usually just fall back to DAB because then I can customize a crazy good Gant chart because it’s got a really so someone has made a really amazing spec that is an amazing Gant chart like it’s incredible like yeah I want to use that thing and then I could change the point is once you get to something that works like what do you want to do like I want to change the colors I don’t want it to be sorted by day I want it to be sorted by month like there’s these little like nuancy things that like I don’t think you could even build a one visual that you could be
44:29 That flexible to do all these things. So anyways, just some other ideas there. There’s actually a few here on data governance that I’m surprised actually was one of the higher requests for a magic wand and we we talked about the lineage and deployment is another one Mike that we’ve talked about for a long time. We have the deployment pipelines right now which is great and we have GitHub But for a lot of the things when we push things live, I think that yeah, for me this is a magic wand that I would
45:04 Definitely want to have in terms of different ways that we can deploy the content because the GitHub’s great right now and like I really do enjoy but Mike if I had to ask you if you had one and you can’t say something that’s going to come out there’s one. Yes. Yes. actually work great in GitHub too. Yes, because it’s all code. Just everything to Neb. Sorry, Tommy. Go ahead. Everything to Neb. Everything to Neb. The integration between Git and PowerBI workspaces. I’m not talking about
45:38 Features that can be implemented. I’m talking about just those two products or platforms. Is there anything you felt is where there’s the bridge is can’t cross over in terms of the workflow that you do in Git and PowerBI workspaces? so you’re just basically asking what limitations I’m seeing through GitHub that I just don’t love. Is that what you’re asking? Is that the the question you’re asking for? Exactly. Exactly. I think the security aspect of GitHub personal access tokens back to Microsoft not the best experience. I don’t I
46:10 Honestly I don’t even think personal access tokens are even like a sanctioned Microsoft approved way of connecting to things anymore. I think there’s better ways of securely connecting. I was talking with Matias about how to connect GitHub to or even DevOps, Azure DevOps directly to talk to repos and to PowerBI and using APIs and things. There’s better ways of doing this where you are just providing dedicated access. I don’t I don’t want to refresh tokens. I just want it to be securely this website is securely able to talk to this website. That’s what I need to have it to do, right? That’s the end goal here is to make sure that it’s super secure.
46:42 Set it up once, I authorize it, and it’s good to go. And it can’t use my personal credentials in any way because that just adds me as a weak link to any part of the process. Most security breaches, I think, Tommy, if you look at like the documentation, it’s the guy who has the password on his screen writing it down because something is in plain text somewhere that someone can see and steal. Like people are the problem with security, but that how they store things, where they put the information. It’s the people that are the problem. Most of technology is pretty secure if
47:15 You use it correctly. So I that that would be my one thing. There’s too much friction in my opinion getting from GitHub to PowerBI, especially in the enterprise space. If you buy GitHub enterprise, all kinds of weird things that people do there that make it difficult to work with. So that’s one area of friction I would like to see improved. , what were you thinking Tommy? I’m guessing your question came with like you had a thought that was like friction that you saw to me right now when I’m publishing in GitHub and then I have to go back to PowerBI. I feel like the workflow is a little weird
47:46 Right right now where once I have everything published in I agree and that’s what I’m trying to say. This is not necessarily a technical bug, but once I publish my workspace and it’s or my repository that branch is main that means in a sense live go there’s some not there’s sometimes a few steps I’m like okay this is good we’ve committed this I made those changes like oh yeah now I have to go to PowerBI as well and get sync those changes in to me it’s like the work the repository the repository for me should
48:19 Be the lie what what’s actually live, but that may be my own magic wand internally. I’m I’m gonna I’m gonna answer your question with a the community is already solving your problem, Tommy. And again, I’m going to I’m going to point at Matias, who’s the gentleman who actually created Tindle, which yay, thanks, Matias, for creating Tindle. Like, you’re amazing. he’s a genius. by the way, if you didn’t know he’s so smart, very good at thinking about code, been doing software development for 20 plus years. incredible developer. He’s building something new, Tommy, that will
48:53 Be exactly what I think you need. Right. Honestly, again, this is back to the community and figuring out things. , really to your point, Tommy, we should be sitting in one tool and one tool alone, right? If you’re going to choose GitHub, I should be able to say, “Here’s Maine. Maine is this workspace inside powerb.com, and whenever I create a branch, automatically the workspace and the data and the testing is automatically created for you.” That’s really what needs to be happening. So, I I really do firmly believe that there are solutions out there that are coming that are going to make some of these experiences, to your point, Tommy, friction is going to be much better. And I think Matias is on it.
49:27 Stephanie, your reactions. What do you think? Oh, I was just going to echo what you said about Matias and his genius. And thank you, Matias. So, sorry about that. But, , no, no, don’t don’t apologize. It’s totally he’s doing such good work. I really appreciate him as a person. He’s a very smart kind individual and and again he’s giving back to the community in a way that like I don’t think anyone else would have seen from something that has so fundamentally changed how I build things. Tim is one of those things like it changes. I’m using it every day now. Like every day I’m using something that he dreamed up
49:59 And created inside the PowerBI ecosystem. Like dude amazing very very happy about like that that type of thing. Anyways, other thoughts? Any other notes here on the LinkedIn post? Anyone else want to point out here? What was the one that you were most surprised with, Stephanie? Then when you were looking through all those comments, we were like, I was not expecting people to say this at all or you might have even just completely disagreed with. I don’t think I disagreed with any of
50:32 Them. A lot of them were things that I just like I said earlier I I don’t think I would have thought of because I’m just so used to how things are. So it’s really nice to see these. now I think there were multiple comments on on bookmarks and I think that did not surprise me at all. I think that’s great because whenever I have to whenever I open up a report and there are a ton of bookmarks I’m always like okay here we Walk away slowly. Don’t touch it. Yeah. buckle up because I don’t know what how these things are going to work.
51:04 I don’t know what they’re supposed to do. And yes, I think that that one is really fair. It’s just bookmarks are are great. , love them, but they’re also a beast to work with. So maybe I don’t know. Any thoughts on that one? Yeah, I definitely want to give you the echo on bookmarks. We just did a project where someone was rebuilding a report and we needed to rename a bunch of columns. Back to your point, bookmarks include pieces of data inside the bookmarks themselves. So if you rename a column, you could potentially also break the bookmark because it’s filtering or adding properties of that bookmark data
51:36 In the bookmark itself like if you select something. So while we were doing the migration of reports, we found again back to Matias and and how and Ruie and Ru is another one of the individuals that I think has changed my career as far as PowerBI goes. His invention of the PBIR format and the second version of this, we couldn’t have done it without it. The bookmark is clearly laid out. the the columns are there. You can search for old columns and replace them with new columns. Really really helpful. but I would agree with you. I shouldn’t need to dig through the JSON or the files itself to be able to
52:10 Replace something like in a bookmark. it would be even nice just to have a visual visualization of like what things interact with the bookmark. What things on the page are highlighted by the bookmark? Is there any data that’s there? Like it’d be nice just to see what the bookmark’s doing to some degree in simple form and you could then evaluate whether or not that’s correct or not. But I would agree with you wholeheartedly there. Oh, and what? One more I want to I want to shout out is default values for slicers and filters. I would really like to see a little bit more more choice there. That’s a good point, too.
52:43 Yeah, I called that one out and I I Yeah, that was just like that’s one of those basic nice to haves, ? Yeah. Yep. Like, wait, what? I can’t default this. Yeah. Yeah, that’s true. You can’t you don’t get to like you don’t get to default stuff. It just just shows up the way it is. or or you have to save it with selected the item, save the report that way, and then it becomes saved as part of the report. But then if that if someone doesn’t have access to that data, then it filters to nothing and then you have a page that’s blank. That doesn’t seem right either. So like or it filters to that item, but then like you see that item even if you don’t
53:15 Have access to that. Yeah. Correct. Yeah. Not good. It’s weird stuff. I would agree with you. It’s just to your point though, right? The the default selection of things I think does make more sense, right? There should be some more options there. I agree with that one. All right, great. How about you, Tommy? Yeah, Tommy, what do you think? Yeah. , people are still asking for their machine learning integration and which I’m like, cool, that’s we’re going to have LLMs for that. So, we’ve been trying to do that for years. Like, so like already done. That’s like I love
53:47 When every time I do training I for the PL300 there’s always the AI section and I always have to preface it with I’m like and now we’re going to talk about the AI features in PowerBI pre- chat GBT. So because it’s crazy all the AI visuals came out way before chat GBT was a thing or all those elements were a thing. So, we’ve tried it to try to integrate it, but again, there are some things that are just never going to be an easy button. Agreed.
54:19 Well, and we when we did have that those one easy button, those three easy buttons in Power Query, the sentiment analysis. Yeah, those are and now they’re gone. So, they got deprecated. Yeah. Must not have had a lot of usage, I presume. , I I didn’t really feel like there was Right. But then when I I did post about that one too and there was a lot of response on people being disappointed about the loss of that. Yeah. Interesting. Maybe it’s going to come back in a different form. Maybe maybe we’ll stay tuned for FabCon Vienna which is actually coming up next week is
54:53 What’ll be. So I think it’s going to be coming out in fabric. it’s in it’s in fabric. So I think that’s the idea. Use fabric. Oh, I I see you’re saying maybe people who aren’t using fabric then they’re like hey you just took away stop stop stop stop stop stop stop stop stop stop stop stop stop moving my cheese away from powerbi and only in fabric now yes I could see that yeah I sent sentiment taken I I get it and this is where I’m like just jump on the fabric bandwagon I get it like but it’s my life is much better with fabric these days I they’re doing a good job awesome all right that being said I think we’re about time I
55:25 Want to say thank you everyone for listening sorry we started a little bit late today some technical difficulties slash We were just talking and having fun as MVPs together before the meeting, so we got we lost track of time a little bit. But that being said, thank you all so much. For those of you who are watching this after the fact, you don’t really care because you just listen to it on 2x speed, whatever you want. and good luck on your run. Hopefully, you have a good run or a bicycling exercise doing things like that, exercising, which I should do more of. That being said, thank you all so much for attending. If you like this episode, please make sure you share it with somebody else. We also have a
55:57 Memberships area. So, we just we’re introducing something new. we have a memberships area on YouTube. So, if you want to watch this episode without any advertisements, come on over to YouTube. You can subscribe to us there. We have a very low price tier where you can get all the episodes for free. and there are no ads in any of the videos. So, if you like that, come on over to YouTube and hang out with us over there. Thank you all so much, Stephanie. Thank you so much for your time. I really appreciate you just being such a good support about this. Your topics are amazing. , I learn a lot from you every time we talk and you’re just so
56:29 Generous and kind with your time. So, thank you very much for joining us today on the episode. It was a pleasure having you and great conversation like always. Thank you so much. I do want to call out Stephanie is definitely running or helping out with the Pittsburgh SQL Saturday. I know call at the beginning. I do want to plug it one more time here. SQL Saturday October 18th is the event. The day before the 17th, there’s a full fabric in a day free training. You can go get it for free and Stephanie will be there running it for you and you can see her in person in there as well. So Pittsburgh, SQL
57:02 Saturday. Definitely want to get a plug there for her event. Tommy, where else can you find the podcast? You can find us on Apple, Spotify, wherever we get your podcast. Make sure to subscribe and leave a rating. It helps us out a ton. Do you have a question idea or a topic that you want us to talk about on a future episode? Head over to powerbi.tip/mpodcast. Leave your name and a great question. And finally, join us live every Tuesday and Thursday, 7:30 a.m. Central on all PowerBI tips social media channels. Thank you all so much and we’ll see you on the next episodes. Cheers.
Thank You
Want to catch us live? Join every Tuesday and Thursday at 7:30 AM Central on YouTube and LinkedIn.
Got a question? Head to powerbi.tips/empodcast and submit your topic ideas.
Listen on Spotify, Apple Podcasts, or wherever you get your podcasts.
