one and welcome to the fourth and last session of today's Microsoft fabric blast off we could even call it so we spend the day going deep diving into some of the technical aspects of Microsoft fabric for those of you who are just tuning in here for the first time we will do a short intro of what is Microsoft fabric just to catch you up and and summarize so don't be worried if you don't know what it is yet today here in the at the showroom I have Henry Kim who is our CTO here at philomine so welcome here henlike thank you do you want to introduce yourself yeah just briefly uh Henry Kim Christensen CTO here in Vermont Denmark I'm not an expert on this in this field and I must admit I've been listening in on all the sessions this morning and afternoon and yeah I'm just happy to be here and maybe go through some of the questions and then also reaching out to you listening in here please feel free to ask questions and we'll try to pick them up along the way yeah and I've prepared a session today about the I will call the data driven organization which is a session about well end-to-end analytics in Thailand sorry in fabric but also a session about how what we should think about as our organization our collaboration and architecture around Microsoft Microsoft fabric so I'm looking very much forward to to this session but I've invited you here today to ask me all those questions that a business leader out there may be sitting with and don't have the chance to ask and if you are sitting with questions out there and you are not afraid to ask them go to the LinkedIn live Channel and just write them in the the chat and the Henry will be monitoring that and just abruptly asking me all the questions during the session making sure that that we talk about what you guys care about exactly yes yeah and from listening in since morning I I must say it's been a long day for you uh also maybe leading up to this they a lot of stuff to look together and comprehend I guess um you said in the morning session I think it's kind of magic but I must say it's kind of mind-blowing uh yeah it's been mind-blowing it's been magic the last session I told Lawrence that some of what he told me gave me literal Goosebumps and it did I was yeah this is some of these things are really mind blowing yeah yeah now yeah and MVP in the field so it's also as you say amazing to to feel your amazement and excitement about it it really comes comes through yeah one of the things I've been thinking about listening in the sessions and also one but I've been able to read up on is that Microsoft quickly in kind of the first sentence jumps into talking about AI in the context of the fabric that's not something you really talked about today what is that well there's two reasons um first of all AI is going to be a huge productivity enhancer for us there is um what seems to be a lot of of focus on giving us for example co-pilot tools to accelerate the way we work that's really interesting but for me the technology is more a productivity enhancer than it is about the impact we can have with our data so it will it will make our work easier and faster but it's not so it won't make it much different technically but the other thing is that there's also a component in fabric about not just using something like open AI or copilot but developing your own machine learning and your own models and it's just such both an important and huge topic it even could use its own day in itself just talking about what we should do with so maybe we should do that and then some other but but there are definitely a bunch of features and we did talk a bit about them in the first session with Brian about um how we could make machine learning models in the data science experience but for the people out there who are new here and and welcome welcome here welcome to you who haven't heard about fabric yet I think everyone did the past few days in the industry but here's a quick sum up of what it is so I have here on some slides here we have fabric here as to really it takes what we had in power bi we had data flow we had data sets we have reports and it just gives us a bunch of new tools and the pipelines to orchestrate all of these workflows and furthermore it collects these workflows into one unified Platform One Unified Service which are neatly tied together with the common storage engine and a common storage data file format in the data LED parquet format which means that whoever we are in organization we have one place to to to put our data so a bunch of new functionalities if you're coming from synapse you some of them Azure synapse you know some of them may all will be familiar to you but now you'll just be able to work much much tighter and more collaborative with your colleagues in the other data professional areas okay I think and that that sums it up so we got the functionality for a data Factory we got data engineering we got data warehousing capabilities we got data science capabilities speaking of AI we got real-time analytics which is also in itself a really interesting thing area and then of course we have the power bi that we know and and love and all of that in one neat package first to just use I I just had a brief conversation with one of my colleagues so now he's coming more from the developer space and he was like a Microsoft fabric what is that isn't there a fabric already this I think there's an Azure service prep work something okay but but the big thing is that of course the service fabric resource is kind of a little thing in the edge of some of the areas where this is this will be the one platform to rule them all in some sense of course there's organizations and they can safely do so that will want to stay in Azure synapse analytics because there are benefits of managing that cloud layer yourself and and and doing that but I think for for most organizations um they don't need and want to care about the nitty grittiness of how to fine-tune those services in the cloud if they can get a service software as a service um that just works out of the box yeah I think it was um it was telling to talk with with last last Anderson here right here before if you didn't watch that third session it was really interesting Deep dive into the direct Lake functionality in power bi uh so mind-blowing but but I think one of the key messages was that some of the things that are in Microsoft fabric is things that we could do before if we tied together different services and made sure that there was probably governed and architected and everything and then we could achieve maybe 80 of the functionality which we will get just out of the box with with Microsoft fabric because Microsoft can unify us on a best practice this just works for everyone so so it's fair to say that it's a new abstraction layer so to speak and taking some of the hassles out of absolutely aggregating your data and make it available for fun analysis I will I will put it differently um it's about taking all the external complexities that we don't really care about we care about a solution we care about going from A to B to generate value with our data so we have some sort of an idea and we have some data and we want to turn that into something that can make an impact in our organization so to go from that that starting point to that impact today requires you to go through a bit of a lot of Hoops of then you need to handle networking then you need to handle your Cloud infrastructure and then you need to handle proper governance and procedures because because everyone can yeah and and access and and everything and it just becomes a lot of work that are not directly tied to the solution you want to actually make the impact but it's just prerequisites and we're pulling away or Microsoft is pulling away a lot of these prerequisites and giving us powerful tools to do a lot of them as kind of point and click setups we saw it with the direct leg to do something similar beforehand we would have to use direct query in power bi connecting live to our our data warehouse import the data we want to import create some aggregations to make sure that some of the direct query was not too slow and even then we could achieve maybe 60 70 of what you get out of the box with direct Lake because Microsoft can go beyond the tools and make yeah just make something that sums up the best pieces of the different things so basically also do some optimizations and in terms of speed and access accessibility and I mean time will have to tell yeah okay that that seems to be the promise that I mean he said it pretty clearly that uh with for example with the direct query we could get real-time live data or we can choose import mode where it needs to be refreshed and loaded in and then it's blazingly fast and with the new dark Lake we get the best of both worlds and I think we will see more of that throughout the whole uh the whole platform it's it's about taking away the unnecessary distractions and focusing on the actual solutions that creates actual value because we will never not have to do those although you did mention it on top of that Microsoft fabric is infused in every corner with the with some promises or some AI capabilities about we saw in the video in the presentation of Microsoft fabric there's a video of a text that um we can show it here on the screen there's a video of a text that describes what kind of report we want to create and then co-pilot just creating that report for you in seconds so so actually I want to yeah say back maybe even the value-adding things will will become so much faster and then that is the benefit of the platform we will be able to move much faster we will be able to skip the complexities and we will be able to utilize the synergies that are in the things being in one place and tied together yeah um like we anything coming from Microsoft new stuff it's always a kind of easy or maybe not easy that's not the right word but to Envision okay now we've got this these new features these new products and then we can start using them but we also know from most of our customers and businesses out there they are already using a lot of these things what what is as you see the the path to using Microsoft fabric okay great great great questions because we have spent an entire day and three sessions telling all the amazing things about Fabric and there are a lot of amazing things but it's a preview feature so I think we should start by taking a little step back and acknowledge that it's a preview feature which means if we put anything in production within this service we will be um we will be at risk for outage outages I mean uh being the system being offline features that may appear wonky maybe we write some code and the next day it could disappear there's no there's no stability and robustness guarantee for Microsoft it's a preview it's a preview this is this is not an unusual concept we have those all the time but it's a preview with all that entails [Music] um but the benefit of us being able to participate in a preview is that for the first for first of all anyone can go now and sign up for the free trial so what we would have to pay for in a production scenario we can actually try out for free if we're an organization or even if we're in an individual business person who has a B2B account yeah you can go sign up for a trial today yesterday and you will be ready and play with it in 5 or 10 minutes um which also means if you're a bigger organization this is the time to start putting in proof of concept projects testing out if there are potential Savings in in a year from now or if there's potential for that even earlier than that and really try out and stress test the features in your scenarios so it's fair to say that you may have a lot of pockets of data already that you build whatever BI Solutions on top of that why not have the benefit of using this preview to kind of connect to these already existing and then try out the new stuff absolutely not have your business rely on it but but but see if it could benefit from it and learn from them absolutely okay so you're kind of building it alongside the existing systems you're not you're not intentionally upgrading existing systems with no road bike plan exactly and even better because all of the experiences are known experiences from the Assassin Azure synapse um ecosystem which means that you could even start proof of concepting in in fabric but also keep building in in Azure synapse and once you feel comfortable with the switch you have two similar platforms that will make porting and migrating easier than than if they were completely different so you can move your pipelines not one to one but almost one to one create the same same kind of copy activities and then use those in fabric you could do it with your your spark notebooks those are probably the one where you almost would be able to do a one-to-one one shift because it's uh it is just code that needs to be moved and and the same thing all around so migrating should be an easy task and I would not be worried if I started like a Synapse Project two or three months ago if this will all be some kind of uh in in vain because once you're ready and have a solution you're happy with pointing that from synapse to fabric um shouldn't be too difficult okay along those lines maybe touch a bit on it um what do you see the the the the training efforts are here how big of a hassle is it to get up the ladder and start using it or it's actually just downhill from here it's easy riding a great great question so let me start by by talking about how we collaborate in um yeah in um in fabric so fabric is Unified around this one leg so now we're back to some technology we don't need to go into to details with that but that means that every person in an organization whether they're a data engineer the data scientist data analyst today's Warehouse developer data administrator or is something else they will be able to collaborate around a joint joint file format joint storage but also a joint platform and user experience and that means that it will be easier to um for any individual to kind of elevate themselves from being a data analyst and then working with these scientists and data Engineers closely but also to take some upskilling training in some of the other areas and then suddenly being able to shift around and find the spot where they can add most value to the company so so so we have these clear paths where we can we can make better analysts then they focus on anything with anything around power bi and Microsoft made it easy they kind of gave her these experiences so we also already have the scope of well there are data list work data analysts working with power bi so they they the tools in there well if they focus on that they should be good to go for for any any future scenarios but maybe they want to move into the others well then they can can learn from them as well as an organization um the biggest things to consider here is that we may have a staff that is mostly in some cases mostly focused on power bi so those will probably want to train and learn some of the the more backend oriented Technologies and you could have a potential Clash of um someone wanting to stay within power bi someone wanting to move someone wanting to do a full stack and it it will be hard to stay up to date and to be skillful in in everything yeah so as an organization you need to be mindful of what are people's tasks and do we allow for more generalists and do we do we want to have Specialists you know okay in the different positions yeah um but anyway yeah so we collaborate around the the one leg and uh and there was a point so so for training also yes one scenario was you're already in power bi yeah of course you want to Branch out an upskill in all these other new experiences this this this may be uh this is bound to happen but because the other scenario could be that you currently have a data platform that is on-prem SQL based it could be an and a cloud a modern data platform in Azure synapse or or something else and so in those cases you're kind of getting a shortcut to learning because as I said before it's the similar Tools in a new platform so if you already have a person who's excellent at as at the Azure data Factory or Cinemas pipeline teaching them to use pipelines in Microsoft fabric will be a non-issue okay they need to familiarize themselves with a new new UI but that should be that should be it and new technical limitations and new technical features but but that that should be pretty pretty easy but then the whole new collaboration effort where they suddenly are able to hand off a a file or a day some data that the colleagues will be able to work on right away directly that will be probably the biggest training effort you have and will be the hardest one because this is all culture yeah it's not hard skills no it's not a new technology and you need to understand it's a new mindset of if we're able to work closer together well what does that mean for our day-to-day interaction should we accept projects where we don't communicate except every second month about what is the status of what I what I did then suddenly that makes a lot less sense yes so we need to think about how this whole collaboration needs to spread out to the culture of our whole whole organization sure yeah maybe it's just to to wrap up that uh about the skills uh you know sometimes when major shifts comes along with in terms of technology and and services to be used in the business and I foresee this is a major shift in in some areas and uh what what bloggers are you seeing in that sense or what are you you know the the Kill Your Darlings so to speak maybe you should put forefronts and and say hey guys we need to do something differently here to to Really harvest the benefits I think the the first things first thing that comes to mind that I want to really emphasize is and I think actually two of our speakers today said something similar and that is if it ain't broken don't fix it so if you have something that you're happy with that works that that is stable robust adds value there's no reason to tear down a solution tomorrow to build a new one in two days that's an important Point must say yeah there's nothing about Microsoft fabric that promises that we need to shift very fast so that we need to tear down our old work or that they will shut off the the button of our current work um but there is a promise of future potential that we can start looking into and maybe consider for any new projects if that should be placed in Microsoft Fabric or if it shouldn't so so it's more looking uh looking onwards but for for bloggers I would I would actually go back and say I like to compare what is happening now with what's happening now with fabric I like to compare that to what happened in the world of bi when power bi was introduced the first time yeah because we kind of have a similar thing going on when power bi arrived it was not Superior to to analysis service in any cases it had a lot of features that wasn't there so analysis service was the cloud cloud equivalent to power bi and the reason why Microsoft space in the Microsoft space and the reason I want to compare them is that what happened with power bi is that they had a strong Cloud offering they made an even stronger software as a service offering with power bi and well history kind of tells what happened there it was a very very dramatic adoption for power bi but people who were already happy on analysis service they didn't need to jump shift immediately because they again they had a robust working product they could steadily and slowly wait for the point where power bi was better in anything that they required and then they can can make that move yes and today it's a no-brainer for me I mean if power bi premium is better than analysis service in almost all instances there's so few Niche cases where it's not so now it's a no-brainer but there's been a long period where the the balance has been getting closer and closer whereas from the beginning spotting that probably would made sense even from the first day to to be used for new projects and I think we have a similar case here going on uh with fabric so I think it's about as organizations to think about what it means to move the data platform into a software as a service because inevitably we're moving in that direction surely yeah yeah so that's your prediction so I pretty much so let's see how long it will take are there any other Concepts maybe you would like to to introduce to the audience at this time yes so one thing that I find a bit curious and I really like it actually so my title is principal architect and when I read some of Microsoft's communication here they say well this has one architecture so there's uh there's you don't so you don't need to think about all the architecting anymore because that is being done for you okay and and it's a promise it's a prediction so in some in some ways I oh sorry wrong slide so in some ways I I kind of tend to agree because as I said before some some task of the architecting will be simplified we will not have to think as much about networking or security or how to set how to how to build systems of systems to make that work because in the future we will just have components Administration settings configurations that allows us to do this which means a lot of what was before architecture becomes best practices it becomes pure governance it becomes Administration it becomes culture even um so I think the the role and the concept becomes a bit more soft in a in a sense but that doesn't mean it disappears because we still have the solution layer where we need to go from an idea and some data to a solution and to be able to pick the right tools for that task bound to stay there and the the how do you say that the need to then structure what what are our governance how a governance how do we allow people to create a warehouse or to create a lake house and how do we allow them to move data from A to B and who should be allowed to do that how do they need to have access to this and all of these questions well that's only going to be higher Focus because we suddenly have a platform where something that took maybe three months four months six months creating a data lake house can be done within three clicks of a button and waiting 10 seconds sure which means if we don't think about governance and we don't think about control or Administration that means if you just give people full access to everything that means within the days we're probably going to have hundreds of thousands of warehouses and lake houses exactly and I and and that was actually also one of my thinkings in here and during this morning and afternoon uh I've been in the the Microsoft industry for what 30 years now and every time Microsoft introduced new technologies uh it's within a couple of years it ended up in me speaking about governance around that technology that service starting with SharePoint my usual 365 Azure for and and that's just uh blowing up right but you think you you actually also make the argument here that really applies here as well um yeah how to govern this and then you you can of course uh start a discussion about what is architecture and do we need the architect still but but still someone needs to think about this the role is going to change surely and at least and the work of the implementation part of it but but someone needs to sit down and think who is about how are the tools used not for one specific use case but how are they used across the whole organization and with collaborating collaboration with other organizations and all that yes that is that is definitely something that that will be yeah and inherent to talking about data is also talking about confidentiality authorization who who has access to what and so on I I understand that maybe the the technical part about authenticating and authorizing you just use your what Microsoft 365 account to access these reports are seeing the data accessing the the tools of the whole fabric but what are you allowed to see exactly and and I think I think we're yet to see what Microsoft is going to communicate specifically about this I think also we're yet to see what what will we get and how granular control are we getting in respect to security okay um the only thing I can kind of speculate or think about is well if we have data in one place it should inevitably be easier to control the access of that data so something a lot of people forget today is that they think that well we have one big data warehouse we have a very tight control of who can access that data so everything is good then they have someone who is maybe a bi developer who has or even a business analyst who has access to this the data warehouse and they're making reports maybe for management so of course they have escalated access this is fine This Is Not Unusual they built some reports a lot of organizations forget to think about that this means that this person will have a local copy of that data on their Hardware they could quit tomorrow they could I mean and then they probably they would have full customer lists they could bring it out of the house yeah so security and access is not just about um one one part of our organization we should actually think about this holistically as well and I think and this is a personal opinion prediction having all of our data in one place can only help to simplify this task yeah so so it is inherently that's your prediction fixing a problem that any organization is facing today because I think also from from a more technical background we've been invested and involved in many clouds migrations right and and one of the the inherent risks there is that you you take one of these say SQL servers just sitting there and it's been sitting there for years and and we we know or we get the documentation of who's using this one okay we can then plan a migration and we migrate and then all help Breaks Loose because oh someone else connected to some Excel sheet that use the odbc connector and apparently use this this SQL database for a very business critical report no one knew about it yeah but yeah you see it also kind of addressing that in in a sense then I I haven't seen it yet no I see see it collect I see it unifying on one storage solution and then I'm very excited to see what what that could mean for security yeah yeah because I I would also make the argument that in many organizations today you may have a lot of systems actually running out there that just sits there and it was set up and prepared and implemented correctly but who knows it uh that the guy so to speak that owned it left the company years ago and it still sits there so that that it's potentially a risk you have today if it's not a maintained correctly uh but but then moving into this new future of them the Microsoft fabric are we able to control that better will we announcer an owner of each and every data source in a logical manner that we we set up so much yet what we do know is that that there is at least one thing that's introduced which is domains so there will be a grouping of workspaces that can have some some more more overall controls but what exactly that will mean and how it can be used we don't okay we don't really uh that's that's one thing to look out for that's definitely one thing to look out for and I would I would say I I'm so excited but I'm also worried about the future where we will see today when we visit a an organization sometimes we will find what we jokingly can call like spaghetti data models data models that were built um in very good faith but then build and then something was added something like that and suddenly we have like a spaghetti bolognese of things pointing to each other so My worry is that will we in a few years time also see organizations that has spaghetti data warehouses with things just pointing uncontrollably to each other circular references even then I mean one lake house using one as a source but then writing yeah oops so we we do need to look into how do we how do we solve this in the in the future because today if we if we visit a project and some some organization may have 10 different Data Solutions that are kind of not unified and they're used uh sporadically well then we call that Messi but in a Microsoft fabric future 10 different data warehouses is really not a lot I mean we are going to see Solutions with hundreds even thousands across an organization so so uh so we do we are going to need tight governance yeah yeah I see a comment here from uh one of the listeners Klaus stoning saying live competent fabric administrators will be a blogger if you start using Fabric and one leg so start scaling up your power bi administrators absolutely absolutely it's a good one absolutely good um now um we said that it's now out in preview and you can start using it and there's no I I think you even said it there's no cost to doing that but uh we can't talk about this really can we without saying okay what is the cost of this or what good question I don't have I don't have an answer again but but but I have what we have so I visited the Microsoft official documentation yeah and what I did let's be honest I went to Google and I wrote with this driving what what does fabric cost and I got into a page that said buy fabric buy Microsoft fabric yeah and found in the documentation the actual purchase detailing of a fabric I didn't find any prices though not in not yet in the calculator not yet in the this page and not the not in the the Office 365 Center yet but what I did find was some info about the new capacities so in a in a new fabric future we will have a possibly some some new skus no some skus yeah um which I can see there's a little star so there may be a a sub note in Microsoft saying subject to change or something it could be but but going from F2 to F4 to f8 and this is based on a concept called capacity units yeah we heard about that before right and capacity units in synapse we also had synapse capacity units yeah okay but now it's a Microsoft fabric capacity on this and I don't think they necessarily compare one to one no okay um but it seems like they're aligning them with the power bi current power bi SQ so we can compare if we have a P1 today that would be an f64 tomorrow oh yeah so take a guess of why this makes me very happy looking at this charger you may not be able to but there's something here I'm noticing that makes me very very happy okay please the smallest power bi premium capacity today yeah is f64. okay yeah so all the times we had talked about how we are premium is awesome but it's too expensive to get started yeah looking at this chart so I still don't know prices no no but I do see capacities being at 32 like divided by 32 of the size of the current capacity yeah so I think we're going to see a lower barrier of Entry sure yeah that would make sense right yeah and I mean we've been adding for smaller premium capacities for a long time yeah and this shows me that we're getting uh getting that and you could basically hope that also the bundling uh user licenses also creates a pool of capacity to be used then maybe today that's that's the speculations we demonstrates that it is per speculation but but today for instance with the power bi workspace is kind of like you can buy a workspace where you buy the capacity so and then you can buy a per user right yeah but that's what you're talking that seems like isn't that what they're doing kind of in Dynamics so you get like I think if you have a power apps license you also get some data worth storage yeah yeah I don't I don't I mean I don't know that you use it across many of the the technologies that they're part of the the licenses especially when you're licensed on a per usual level it kind of brings along some capacity you can use uh just yesterday I looked at some Visual Studio subscriptions it brings something about uh these built the agents I don't know maybe we should stop speculating um yeah but I it to me it seems promising but it also it's it's definitely an area where we need more information sure and I'm not sure we're getting that information right right now because I understand also there's someone who wants to evaluate surely what should be the pricing model around this no I mean it is probably difficult also for Microsoft to figure out because it's it's the the inherent thing of also providing SAS services that that we kind of expect out there that you get a per user per month cost right and then you can just go nuts right so so what is the variable here uh I guess everyone needs to figure that out yeah yeah um um again the session to before this session um was really interesting also for me looking at from from the outside not being a may I call it the data data nerd yes but one of the things I've been struggling with um in in my day-to-day work is actually a setup where we would like to see some real-time data uh presented through Power bi um is that something you see improving in this next generation of a setup absolutely and you should have you should have caught Brian going outside out of the room here earlier before because Brian is in my uh in in my view he is Mr real-time data yeah um he's been digging deep into what can we do with these new tools and and everything but but one thing that do come to mind do come to mind is that we are getting a bunch of new easy streaming capabilities so we have a new event stream yeah which allows for point and click loading of streaming data and then delivering that data to a streaming database but also to a lake house and probably even more um but you should definitely have a chat with uh would seem so with Brian because he knows uh he knows all there is to know about the okay yeah about this thing um again at the session before we talked a lot about the power bi is is are there any other things that is worth mentioning in terms of uh the accessibility of power bi as a as a business tool I saw the announcement that the power bi desktop seems to be coming for the Mac OS now as a desktop tool that's no not not not to my knowledge so but but what is happening is and this is also some of the things that last demoed is that we can now create data sets in the power bi service or in the Microsoft fabric service um we can create power bi data sets there so step by so so just to explain it so instead of using the desktop tool you're able to use you can use it the online one okay yeah okay of course there are some steps along the way that needs to be some bloggers that needs to be eliminated we still don't have to my knowledge we don't have a code versioning or a solution control of what we are developing and changing in the service so if we're changing something but we had Version Control on our desktop well then we're kind of breaking breaking that that's CI CD doing that so that would live there's limitations but step by step we are getting an online service that can do more and more of what the desktop desktop can do and I think that is the answer to people on Mac wanting to do something in Microsoft fabric is that well the online service is becoming more and more capable and it should be their their preferred way of of working but but right now it's the limited experience I would definitely say install the power bi desktop and do what you can do okay but we're in a weird place because what we saw earlier we can actually do some things now in the service with for example directly on our lake house which is not possible in desktop interesting so so far it's been kind of a catch-up and now it's this mix again where some features hear some pictures there yeah yeah but yeah that's that's the whole notion of it it is a cloud services and maybe it's easier to work with it more natively and closer to the cloud service and trying to have this uh how could you say it disconnected uh setup that the uh the desktop apps absolutely inherently works on so one thing I also would like to mention is that um we need to think about going forward what kind of organization we see ourselves in in respect to our data and data management um I've I've seen three major patterns or three major archetypes of how we how we handle our data and that is the Enterprise and as we see on the slide that is the the Enterprise way of doing it with the uh centralized Data Warehouse close to close to it close to the core of the business with very tight control of how things are created how things are shared how things are used um but with no flexibility to give people out there in the The Edge in the business units the the actual last step transformations of the data that they may need or they may want so this is the classical data warehouse example where you have these Warehouse in the basement yeah and it's very tightly locked who can use it for maybe some reporting maybe you can they can create some reports but they definitely cannot create their own semantic data models with something like Microsoft fabric suddenly it becomes very easy to create a fully self-service data platform scenario um I think this often is also branded as a data data mesh um where one business unit can have their own little landscape of pipelines and warehouses and data and and and then they can subscribe to the data of other units that they can offer their data to other units and very much make it kind of a Marketplace of data more than a more than a modern structured unified data landscape so so very interesting and very promising but also for to me very risky in terms of of how do we govern that kind of that kind of thing and then finally we have the which also is on the slide the hybrid model where we where we try to take the Best of Both Worlds there is some robust refined data data sets and data pipelines and and things in the koi we can offer that out there in the periphery to the to the individual contributors and they can maybe take that change it a little use it for a specific reporting maybe push some of the data back but in a more controlled manner so I think as an organization we should also look look inwards and think do we want to be a very tight control organization do we want to be one where we have um very self self-sufficient units that don't need to follow strict guidelines or governance or best practices or do we have one to find some Middle Ground where we maybe have Center of LX lenses and best practice guidelines and stuff like that so there there's going to be a spectrum here and I guess that was uh some pretty good final words I guess you could be talking for many hours still but it's been interesting at least for me uh yeah and uh that was actually also my final question uh on the slide here how to get started apart from reaching out to us right absolutely to get started and they I would say three things shown here but four things you can go to to try fabric which is on AKA dot Ms Dash try Dash fabric yes um you can watch out for fellow my monthly power bi update which I have kind of a feeling will be via Microsoft fabric update in the future or you can connect with me and Brian our data platform MVPs we're more than happy to have a chat or a talk about uh what your what you're wanting to do with it if you have any questions confusion everything I mean we'll get data nerds if you want to know data with us reach out yeah and finally of course you can also formally reach out to to you Henry yes or to anyone in fellow mind if you want to do more of a formal collaboration about about a great proof of concept idea or something like that so don't be a stranger and reach out if you have any any questions and I've enjoyed having you here as a guest today I hope you learned something today absolutely and it's been a blast it has thank you for joining and thank you for joining and I just hope you enjoy your day and stay in touch bye for now bye