FusionTalk

Have you ever been involved in a project and felt like you were going around in circles and not moving forward... well Anouck has a solution “Kill it or be Killed”   Listen to Anouck and Steve as they describe the situation where no solution appears to work... so... 

This podcast looks at the process of being blocked, stuck and unable to move forward.  How it feels and when a drastic solution is needed, throw it away and start again.

Using a real-world example of a connected workflow drawing content from a remote system, we look at the process around making the decision to stop, re-think and re-design the code to something more efficient (quicker) simplified with a focus on the goal and finding the courage to accept that “I have learned little” from the direction I was moving in and needed to return to Point A and take a different route to success.

Creators and Guests

Host
Anouck Fierens
MVP | MCT | 🎙️M365 | Blogger | Book lover
Host
Steve Dalby
Podcaster "Office365Distilled" Driving Collaboration Business Goals, Speaking about Governance, Whiskey taster and imbiber all round father and good guy.

What is FusionTalk?

FusionTalk is a dynamic podcast where technology, collaboration, and innovation come to life! Hosted by Anouck Fierens, a Microsoft 365 & Power Platform expert, and Steve Dalby, a SharePoint & Teams specialist with a knack for humor, each episode delivers fresh insights, real-world stories, and engaging discussions with industry leaders.

🚀 Topics include:
✅ Modern workplace & collaboration strategies
✅ Practical case studies & expert insights

Join FusionTalk and discover how to work smarter, faster, and more efficiently in today’s tech-driven world! 🎧 #FusionTalk #TechPodcast #Collaboration

Anouck Fierens:

Welcome to Fusion Talk.

Steve Dalby:

With a nook and Steve.

Anouck Fierens:

Hi, Steve, we are at episode four of our podcast.

Steve Dalby:

Episode four. Fusion talk. Episode four.

Anouck Fierens:

Yes. Last time we talked about live demos. Live demos. Yeah. But what if they go completely wrong with the customer?

Steve Dalby:

Ah, but not the demo.

Anouck Fierens:

No, the demo, but everything.

Steve Dalby:

Everything.

Anouck Fierens:

Everything is screwed up. What do you do?

Steve Dalby:

I don't understand. But yeah, we'll catch on that. That's what this is about. Today we are going to talk about. When do you restart? When do you go again? When do you throw it away? Everything.

Anouck Fierens:

Yes.

Steve Dalby:

So we'll get on to that a little bit later. So that is pretty cool. So we've got a few things happening. If you want to hear us speaking together, we're speaking at Bremen in a few weeks' time in Germany.

Anouck Fierens:

Where are we going to speak about?

Steve Dalby:

We're going to speak about business and development, how you can streamline the process of what the business needs.

Anouck Fierens:

More especially with power platform.

Steve Dalby:

Oh yes, we're restricting it to the power platform and how you ensure that what you're delivering is actually what's is aligned. We've got a great description of the session, so if you want to find out what we're doing, go to Colab Days Bremen and have a look what we are speaking about, which is be quite cool actually.

Anouck Fierens:

It will be. It will be the first time we do a session together, but it will be great fun.

Steve Dalby:

In public.

Anouck Fierens:

Yes.

Steve Dalby:

With real people. All two of them probably. So yeah, so that's gonna be good fun in Bremen.

Steve Dalby:

So yes, episode four is interesting. The con the concept of what we're trying to do still seems to be working. We have a number of ideas coming forward. We want to talk about how we actually get to the user experience, the UX. How we sort of make sure the business can use some of the ideas. But today, having had a real-world example of some work we were doing together kind of gets stuck.

Anouck Fierens:

That I throw the way totally and start from scratch again.

Steve Dalby:

I've got some news for you. Okay, that is a nook. What have you done? I've thrown it all away. What?

Steve Dalby:

Yeah, so we thought we'd just talk about that process and how that fits into play.

Anouck Fierens:

I wonder how many people do it, throwing their stuff away if it's not working and start from scratch again.

Steve Dalby:

Yeah, I guess I don't do it very often. I have done it on silly things like PowerPoint presentations when the concept's not working out right.

Anouck Fierens:

Yeah, but I do think many people are just afraid to say, I screwed it up, I need to start from scratch again.

Steve Dalby:

Hashtag screwed up. I did a whole series of podcasts and blog posts on hashtag screwed up and had a whole series of social media on it all. So yes, I am a big believer in fail to gain, and it's a very agile kind of thing, you know. Better to move it forward and fail and learn. Then not at all.

Steve Dalby:

But in this case we're talking about the process and about actually cutting code and trying to fit it in. And we're talking in two respects, I guess, as an individual developer. And we'll also touch on some of the agile stuff as well. But I don't know, I wonder how many people do that. How often do you do it?

Anouck Fierens:

Not that much. But when it's really needed, like the example we are using in this case, I do it because there was no progress anymore. I didn't see any new results or anything else and I was stuck.

Steve Dalby:

Okay. So is this because you're so perfect normally you don't have to, you don't fail. Is that what it is?

Anouck Fierens:

Everyone fails.

Steve Dalby:

Everyone fails. That is very true.

Anouck Fierens:

And that's the nice thing because you learn from it.

Steve Dalby:

Yes. So what did you learn this time?

Anouck Fierens:

To do another way of thinking.

Steve Dalby:

Yes. Take the box, think left, think right.

Anouck Fierens:

Yes. You don't need to think out in the box, but go out and see what you can do and it's not because you know a lot of those things, you cannot combine things and you need to search out how they work together.

Steve Dalby:

Yeah, I think in this place you simplified it. So your first instinct was to, hey, let's do this en masse and do it all in one go.

Anouck Fierens:

Yep.

Steve Dalby:

And that was really a timing thing. We'll get into the problems with it a little bit later on. Well, no, actually, we may as well talk about that now. So we'll go through the process and where this fits in. But what you actually had is a scenario where we're running some content from an external source, we have a connector in place running APIs, then putting it into a SharePoint list and to manage metadata. And of course there are tens of thousands of items in the original list.

Anouck Fierens:

Yes, and the update process, when there was something changed in the data source and it needs to be updated in SharePoint, it was horrible. It was taking up to three days before it was updated.

Steve Dalby:

Yeah.

Anouck Fierens:

And it should be updated every single day, once a day.

Steve Dalby:

Yep. And also you were hitting the 5,000 limit in the views when you were trying to iterate some of the issues. So that's an that was definitely a learning.

Anouck Fierens:

It is. It was a nice learning. We find the solution that is working now, the customer is happy, we can give him some he is what he asked for, and that's the most important thing. But yes, throwing it, throw that away.

Steve Dalby:

In this particular case, our scope didn't change. So we actually were able to deliver to the scope. But there are times, I guess, when you just cannot do what the customer wants.

Anouck Fierens:

Yes.

Steve Dalby:

And it's interesting you try and hit and understand that.

Anouck Fierens:

Yes, and if you can't deliver what the customer wants, you need to talk again and you need to go back to the previous process which we already discussed as go to the brainstorm. Explain why you can't do it and see what other solution you can have together.

Steve Dalby:

Alright, well let's do that. So one of the things that we're putting together and we'll talk about it in a in a future podcast is this concept of a design life cycle. So we're kind of building that out, thinking about where that's going. And this design life cycle is obviously including the whole process from idea generation and ideation and brainstorming and everything else. So, yeah, so some of the steps we're gonna talk about are part of that design life cycle and we promise you we will eventually have the whole thing sorted out and licked.

Anouck Fierens:

Yes, and we will make some nice diagrams of it that they can find on our socials.

Steve Dalby:

Yes, and we'll also do some nice podcasts about it and we'll put some presentations together about it as well. So I think it's always important to have some kind of structure when we're putting these things together. We've talked about structure within brainstorming. but brainstorming itself is a step in a design life cycle.

Steve Dalby:

So in essence what we have is a business which has a need and a requirement. And so we've already talked about how we taught get that business need out into a sort of technical need and requirement. So we have the brainstorm, we've covered that, we're done. So now we know what the scope is of the customer. Alright, good. What do we do next?

Anouck Fierens:

After the brainstorm, we are going to see how we can get this in a solution, what we can do technical before starting to develop or anything else.

Steve Dalby:

Yeah, we're gonna refine the solution, the technical solution. I mean, it refine is an agile term, of course. So if you're in an agile, your team will get together, you'll have your own technical brainstorming and say, okay, guys, this is what the customer wants.

Steve Dalby:

In this particular case, the customer wanted I've got all of these records on my ERP system and I want to be able to bring some customer data from the ERP system into SharePoint so that I can use it to trag everything within that customer in my collaboration platform, documents, lists, contacts, everything else.

Anouck Fierens:

Yes, and not only into SharePoint, but also in the managed metadata that they can start tagging things about it. So we need to go move way farther than only the development of the power automate flow but also the PMP search where parts are in place in here.

Steve Dalby:

Correct. So we know what the bigger picture was of the requirement, but the first thing, the first stage was get that data from System A into SharePoint Online. And that's the bit that we'd started working. And the idea is that you would get your team together. In this case though it was you.

Steve Dalby:

But you know, you whatever, the refinement stage is either I sit down and I work out technically what I want to do. Or as a team, we'd sit there and talk about it and work out there's different little stages that needs to be done. Everything has to be done in small iterations, whether you're doing it yourself or whether you're doing it as a team.

Anouck Fierens:

Yes. The little sprints. Little sprints.

Steve Dalby:

Oh my little sprints, little darlings.

Anouck Fierens:

But that's what we did. We first said to the customer, okay, we can do we can. But we need to know if we are able to get the data out of the SEP system. If that's not possible, it stops.

Steve Dalby:

Yep.

Anouck Fierens:

So that's where we started with.

Steve Dalby:

And we managed to do that?

Anouck Fierens:

Yes.

Steve Dalby:

We got data.

Anouck Fierens:

We've got the data and we know how the structure was of the data. So then we start up with building the SharePoint list.

Steve Dalby:

Yep.

Anouck Fierens:

And then the next step was in getting all the data from A to B.

Steve Dalby:

Which we did.

Anouck Fierens:

Which we did.

Steve Dalby:

And it took us about eight or ten hours or something silly.

Anouck Fierens:

Something like that. But that was with the amount of data. So you can't change anything.

Steve Dalby:

So we assumed that it would now take a lot quicker to do just the updates?

Anouck Fierens:

We assumed.

Steve Dalby:

We assumed. I assumed, yes. Yes. The development team, you, assumed And it makes sense. I mean, as we talked about it, we worked through what some of the things were and we both assumed, okay, so if we assume 10% of them get changed or there's an extra four added a day, then we know that we should be able to do it overnight. Because the idea is that you only have a window to do this kind of stuff when the business is not working.

Anouck Fierens:

Yes.

Steve Dalby:

So we kind of arranged a window as part with the client that says, look, we're gonna kick this thing off at 11 o'clock at night and it should be done by six in the morning. So we know that's what a full update was. So it was a no-brainer.

Anouck Fierens:

That it yeah, it for us it was no-brainer until we have developed it and until the workflow was starting to kick off and it was, oh, it takes up to three days. Where did we mess this thing up and what did we do wrong?

Steve Dalby:

And we won't go we'll go into the details at the end as to how we actually did fix it. But we spent an amount of time trying to, or you spent an amount of time, because all I can do is support you by saying, oh, it will work, don't we? You can do this, you can do this, and look, it's fine, it's fine. Because I have no idea how you were doing it or what you were actually doing it. And of course, it's a it's an unknown environment this is happening in.

Steve Dalby:

Yeah, you've got your you've got stuff in the cloud, which is your Microsoft 365 SharePoint sites, you've got a server on premise, you've got different networks in between the two.

Anouck Fierens:

Yes, so it's just trial and error most of the times.

Steve Dalby:

Yeah, of course it is. Run a bit of code, see what happens and everything else. And then it just went on and on and on. I kept getting messages saying, it's only been running for 12 hours now. Twelve hours, we don't have 12 hours.

Anouck Fierens:

So I know. I've keep you updated about the process, but yes, it was a learning It just we stopped it at one point. We just stopped the workflow, we get it offline, and we were t I was thinking about how to fix this. So getting all the data at once from inside system A and B and looping over them wasn't working. That was for sure.

Steve Dalby:

But if you knew that, why did you not just change it immediately? Why did you spend two weeks sitting there trying to tweak it to make it work?

Anouck Fierens:

Because we need to have something to compare if the data was already in the SharePoint list, yes or no. So we need to have the comparison between the two systems because we can't have the update table in the other system available. So we didn't know which one was updated or which one was new in the

Steve Dalby:

So what you were trying to do is to find a way of sort of reading the data, doing a comparison. and then making a list of those things that needed updating and then running the update on only those items. So that was the logic and that makes all the sense in the world. So you tried it and tried it and tried it and it didn't kind of work.

Steve Dalby:

So one of the things that we were talking about earlier when we was putting this together was at what point do you decide enough is enough? And who has to make that call? So from an from I mean this is your case it was you so we'll come on to you making the call in a minute. But in an agile perspective or a project perspective, there's more than one person involved in the team.

Anouck Fierens:

Yes, and I do think in agile or project it's more the team that decides at some point in time. You do provide updates. In Agile you have the sprint updates every two weeks or depending on what you decide to do. And in there you can say, hey guys, I'm not making any progress. I'm stuck at this. What do we need to do?

Anouck Fierens:

And you start discussing it and brainstorm over it a little bit within your sprint team then and then you can decide from alright what you have go back to the scope see what they want and see if there is another approach

Steve Dalby:

okay of course within a agile perspective you also have the fact you have that end of sprint point where you're supposed to go back to the customer and say hey we've done what we were planned to do of course now you have to go and say We have not done what we were planned to do, so there's a delay being brought into it.

Steve Dalby:

So from an agile perspective in that sprint, it's very there are marks and points when you can say, okay, just a minute, this maybe we're doing something wrong here. The other thing is if you've got a PM involved as well. So a PM could also turn and say, guys, you keep saying you're gonna deliver this, but you're not gonna deliver it.

Steve Dalby:

So eventually you get enough hints. And I think in a team it's easier to do. If you're a stubborn development woman, I don't know what I'm talking about here, that is tenacious, that kind of wants to make sure everything works, you just want to keep tweaking it, don't you? You just want to keep fixing it because you believe inside yourself that I can fix this.

Anouck Fierens:

Yes, because you believe that the way you are working is the way of way to go. But there is a time that you are tweaking and tweaking and tweaking and you don't see any progress, you just need to stop and need to go back to the scope as well and say, alright, what do the customer want? I tried this approach, that's not working. What can I do? What's another approach?

Steve Dalby:

Okay. So what actually happens you derive this piece of code, you it's an error, so we hit the 5,000 limitation error to start off with. So that was something you worked on. that spent a little bit of time. Then we had the time problem.

Anouck Fierens:

That was fixed.

Steve Dalby:

Yeah, that was fixed. Well done. All cool. And then you have the time error. So do you actually think about whether I go back to the point, okay, let's just go back to the version of code that had the 5,000 bit fix, for example, and then try and work on a different direction.

Steve Dalby:

In but you made the choice this time Which you sit there and go, no, I fixed it, but actually I started again from scratch and of course I was thinking, oh shit, that means we've got to redo all that code, we've got to redo this, we've got to redo the other. But you made that call and it didn't panic you or it did panic you?

Anouck Fierens:

No, it didn't panic me because I knew it was the best solution we had for d fixing this issue at this moment in time, because everything I did for tweaking and or changing made me go nuts because it didn't work.

Steve Dalby:

Made you go nuts. Yes. Well, how do you go nuts? You sit there and you throw things across the room.

Anouck Fierens:

You don't wanna be there then.

Steve Dalby:

Yeah, I can well imagine. I know what it's like when I'm trying to sort of do something. I mean, I don't do dev work, but when you're trying to kind of I don't know, get a search page set up, you know that, because I keep coming to you saying I've tried this and I can't get it to work. So I do get that frustration, but d but coders are supposed to be purely logical and, you know, not emotional, isn't that correct? It's all about the logic.

Anouck Fierens:

Who've told you that?

Steve Dalby:

Nobody did, but it's an assumption on my part, a wrong one of course.

Anouck Fierens:

Yes, most of the time people assume that coders are logical, but there is nothing logical about it sometimes. When you are going in ESP.net code, there is, but because it's the power platform and you need to think about the limitations we had, there is nothing there is logic that we need to implement, but there is nothing logical about how you do it. I mean There are thousand ways of doing something.

Steve Dalby:

Oh, I guess that. So I mean that's what we're talking about, of course. It's just that you're choosing one direction and you want to go into a different direction, which we'll get back to in a second. But so is I hear people talking about low code and I know what low code means, but I don't know what low code feels like.

Steve Dalby:

So I know you drag and drop your workflow stuff on there, and I've tried it, thinking, hey, apparently a dumb bastard like me is able to do power automate workflow, but I'm not sure about that. So you drag them in and then of course the real annoying to no the real thing is in the detail within each of those blocks isn't it so that takes a different kind of approach than if you were actually coding

Anouck Fierens:

yes for me it does okay it it's a different way of working different way of thinking if we do an API call in the power out power platform it's different than we than you need to write the entire API call in code Okay. You have other things to consider about. If you do it in code, you have more options you can give for your API call than in the power platform.

Steve Dalby:

So the power platform is providing that kind of, if you like, some kind of operational structure for the software to plug into so you don't need to worry about that framework. Wow, learn something new every day. Every day. But that's the whole point of these podcasts, which is quite good. So then going back to this problem we're having.

Steve Dalby:

Your problem, the you the tweaking that you keep doing, I guess is you go back to this structure of power platform and look, what can I tweak, but you've got different relationships between the different blocks of the workflow.

Anouck Fierens:

Yes, and with the problem we had, there was a structure that I created. Okay. And I was keeping stuck in that structure.

Steve Dalby:

So is the structure the correct one? Yeah, get it.

Anouck Fierens:

And that's why I decided to throw it away, because I thought at that moment in time it wasn't the correct structure.

Steve Dalby:

It's a bit like those silly games you get, you know, like the things where you have these balls and you have to try and put them into a pattern, okay? And if people know the car parking one, you kinda where you have to get the car out of the car park. But yeah, you're kind of given this shape and you've got these patterns zigs to put into and if you start it wrong you're never gonna get there.

Anouck Fierens:

It's a little bit like tetr playing tetris.

Steve Dalby:

Tetris, yes, I guess it is, a little bit like that. So you've kind of got to understand the basics and if the basics are not giving you what you need you go back to basics.

Anouck Fierens:

Yes, indeed. And then you just need to check do we keep with basics we had or do we going a total different direction.

Steve Dalby:

And in this case, your basics were the structure you'd put together in your mind when we did the initial refinement about what you needed to do.

Anouck Fierens:

Yeah.

Steve Dalby:

Because this particular job was an interesting piece of work because we everything stacked up. Yes, they've got the API call capability. Yes we can talk to SAP from SharePoint. Yes, we've got the connector in there. So all of the that was in clear in your mind. Then you needed to go away and make sure you can make the right API calls, get the right data back, and that test we worked. And I guess at that point you start to build that structure.

Anouck Fierens:

Yes.

Steve Dalby:

But the only thing you your scope at that point is just first iteration, get the data in. So you then carried on building on that structure.

Anouck Fierens:

Yes. Then we built the other requirement from the customers. So the first requirement was getting the data out of SAP. We managed doing that. The other requirement was saving the data into SharePoint. That's the second step we did. We managed as well.

Steve Dalby:

Built on the first structure.

Anouck Fierens:

Built on the first one. And then the other one was the update process. And we built that on the first two steps and there we failed.

Steve Dalby:

Well, there it was not working the way you expected it to do.

Anouck Fierens:

Yeah, it takes too long. Yeah. So And we failed also a little bit because it was updating my data more than once in my SharePoint list.

Steve Dalby:

Ah, so that was a clue.

Anouck Fierens:

Yeah.

Steve Dalby:

So we tried to talk about the things that identify the decision point where you throw it all away and start again.

Anouck Fierens:

Yes. For me as a developer perspective, it was the tweaking didn't work anymore.

Steve Dalby:

The tweaking, is that a technical term?

Anouck Fierens:

No, I don't think so. I just love the words of tweaking.

Steve Dalby:

Tweaking, it's a good word, tweaking. Isn't that supposed to be some kind of erotic dance that people were having on TikTok? No, it's twerking. Is that twerking?

Anouck Fierens:

That's twerking. That's something else.

Steve Dalby:

That's something else. Okay, I get that. So we don't have developers twerking the code. We sorry, tweak it, but yeah, I know I know what tweaking is a brilliant English word, so I do understand that. Yeah, so you're basically changing little bits.

Anouck Fierens:

Yes.

Steve Dalby:

But you're there there's a dependency then on those little bits that you're changing and you've got to start thinking about a little bit like those silly shapes things again. You've got to start to take them out, change it around, put it in like a jigsaw.

Anouck Fierens:

Yes. That was one of the things for me that did because it didn't work. I keep keeping heavy.

Steve Dalby:

It was just taking too long that we had a problem was taking too long. So the test was, well, actually we don't know because that did we did we ever run it to the end?

Anouck Fierens:

We did once.

Steve Dalby:

Okay. Oh no, we didn't quite. We did 55,000 or something and we'd stopped it a few thousand short, but it had been running for nearly whatever.

Anouck Fierens:

I thought it was running for three days. So another decision for me to throw it away was the changes. I've lost track of the changes. I didn't know anymore what I have already tried and whatnot. And I was like, alright, I can try to s write them down and to keep track on them or just try to retrink the entire process.

Steve Dalby:

Okay. And this low-code approach, this structure approach, does that help the old sort of working out what I've done and what not done? Or does it add confusion? Because you go into the structure, change a bit in this part of the structure and change a bit in the other one.

Anouck Fierens:

It's then For me it's adding conf it added some confusion. Because when you do something in code you can put in code just in comment.

Steve Dalby:

Gotcha.

Anouck Fierens:

And then you know what you have changed. But in here you don't see it anymore.

Steve Dalby:

Yeah, I guess because the idea is it's trying to simplify the overall approach.

Anouck Fierens:

Yes, so that was a little bit of the confusion. But we managed that out, so we did the changes and

Steve Dalby:

we always manage these things, don't you? You work out what you need to do, so that's fine. So we had a couple of things then, so tweaking. Well when I think I've done enough tweaking. Tweaking? Yeah, that's good. And of course changing and unsure un unsure of what change effect it has on something else, of course. Because very often, again, those silly bloming games, you think there you've done one part, but if you haven't got that part right, you you're just compounding the problem, moving on.

Anouck Fierens:

Yes. And I do think in developer perspective, there is one more. thing one more item that you need to keep in mind when you need to throw away code when a developer lost the track of the scope he doesn't know what the scope is anymore and he just tried to do something.

Steve Dalby:

Well he knows what the scope is but his code is not delivering to the scope. So I guess there's a constant testing and just making sure you know what the goal is.

Anouck Fierens:

Yeah. That's also in the previous podcast we had a little bit of the developer lifecycle.

Steve Dalby:

Yep.

Anouck Fierens:

That's why the developer also do the first tests. So you need to test your code to see if it is work working, yes or no. And when you have deliver and when you see in the test they don't deliver what they need to deliver, you can stop the code and redo them or you can see how long it takes if it is depending on time. But that first developer that first developer testing is very important in doing this.

Steve Dalby:

Okay. So to go back to how you were working with this code and deciding to restart it, it must be incredibly frustrating. So you go, I fixed the 5,000 item limit, so I'm all good now. So I'm just gonna run it now and see what happens. And then it takes its time and slow. You kind of hope it'll resolve the problems, you stop it, you make sure you everything is still, and then you redo it again.

Steve Dalby:

So there's a level of frustration each time until you get to the point and go, Crap, I have another problem to deal with now.

Anouck Fierens:

Yes, it is. And it's frustrating. It really is, but that's also you can have it with other items as well.

Steve Dalby:

It's a challenge though, isn't it? That's what keeps you ticking over.

Anouck Fierens:

Yep. That's the developer challenge. You can have it in every single step.

Steve Dalby:

The developer challenge. Okay, alright, good. So you made the call. I'm gonna start this again from scratch. So did you throw the whole structure away?

Anouck Fierens:

Yes. I actually went back to just open my notepad and design the workflow, another workflow, just on paper without doing anything first in the power platform. I just say, alright, I first need to do this, then this, then this, and build all of those the little steps on paper and there before going to the power platform.

Steve Dalby:

So you went back in the dis the designer work life cycle. Sorry, the words went straight out my head then, and basically went to the refinement stage again to say, look, I well I know what the scope is and that's fine. and literally brainstormed it in this case, drew up the logical flow of content.

Anouck Fierens:

Yes. So before I threw it actually away, I've compared what I designed there and what I was having. And it was completely different. So I just throw it away, start from scratch, and that's what I did.

Steve Dalby:

That's cool. Actually it's very, very cool. So what you basically had the first stage and the second stage because that's what you'd had in your first drawings and then you'd added the third stage.

Anouck Fierens:

I even changed those two stages.

Steve Dalby:

No, I know you did. That's what I was trying to get to. So, but then that was the first time you got to see it in the bigger picture, the whole thing, and realized that, hey, this is building, I'm building a monster here because I can, I don't know, make calls for my stage three without thinking about one or two.

Steve Dalby:

So you were able to move across. So is the lesson here then that you when you do the first iteration. So I in this case I can get the code, the API works. Second stage, I can copy it into the to the appropriate list and I can get the data and I can go to the end. Second stage done.

Steve Dalby:

And then the third stage which we failed at which is where you were trying to do the updates and changes. Is it that we need to change the design lifecycle that says actually there's a multiple number of loops here. So after I've done the first test First iteration. When I approach the second one, I just throw it all away and I just rebuild from scratch.

Steve Dalby:

And then you get to very good question. And then you get to the third one and instead of thinking about one and two, you go, okay, I'm just gonna start the design from scratch. So you do mini design iterations ignoring what you'd previously done. That would have saved you a lot of frustration.

Anouck Fierens:

I think it does, but I never thought about it like that.

Steve Dalby:

This is where you say, Steve, that's brilliant.

Anouck Fierens:

You know, you say it from yourself, so I No, no, no,

Steve Dalby:

this isn't look at that. You're a hard woman. You're a hard woman. And no, it just struck me that

Anouck Fierens:

But what if you have a workflow or something else that is divided in four or five different workflows? You cannot throw them away every single time.

Steve Dalby:

No, but I guess you would, again, just purely from a non-technical perspective, thinking about the flow of content, you would know it's a bit like having multiple flows, isn't it? Alright. So I need to I need to approve this document. So you approve a document, flow completed. I need to now have the document signed. So you start a second flow that then does the signing.

Steve Dalby:

And then I guess if the third one is that I need this document approved and signed by the same person, you would make a decision whether you actually build one flow to actually do the approval in signature or whether you kind of just push one flow into another. So each of those sections, I'm guessing there must be an endpoint and a trigger point to start the next one.

Steve Dalby:

So I guess it depends. If you've got five of those to do, it really is a big redesign, isn't it?

Anouck Fierens:

It is. But it's actually a quite good way of thinking. So if you do it by the small steps and you go to have a one note or what else or a whiteboard and you just put in the design there every single time you can, then you have the complete review of how it would look like. And that was what I was missing in the first time. So it was actually a quite good idea for you to do it like that.

Steve Dalby:

There you go, that's what Moraine would call a golden nugget. Alright. No, no, that makes sense. That makes sense. But of course the thing is, it's the same with anything. If you're in the middle of something at the time, you don't see the bigger picture, do you? And what you did was you stepped back.

Anouck Fierens:

Yeah. And that's sometimes just needed to take a step back, to rethink over all of the things you have done, what you can do, and that delivers most of the time the best results.

Steve Dalby:

It did this time.

Anouck Fierens:

It did?

Steve Dalby:

So far.

Anouck Fierens:

Yes.

Steve Dalby:

But what this also highlighted in our particular case, because we t mentioned earlier about the idea that there is a update table that we can potentially look to use, but we couldn't get access into it and everything else. And that was also because of timing, because it's such a big table, it's got all the updates from the last 55 s years or whatever. Might maybe a slight exaggeration.

Anouck Fierens:

It has the updates from the last hundred days.

Steve Dalby:

Yeah, so effectively stepping back and relooking at it also and dealt with a few other issues that so it's not finished yet because there's some more tweaking to do.

Anouck Fierens:

Yes, there is some more tweaking to do, but the basic is right now and the tweaking is now just which column needs to be updated, which not. So these are the small things to change, but The basic is there.

Steve Dalby:

Okay, cool. So in this particular case there was real value in taking a different approach, rolling it over, throw the tweaking away to and basically rethink it, get the old surface pen out and whiteboard and away you go. And was there a ha moment?

Anouck Fierens:

No, not really. There was a moment from yes, I made it. That moment there was.

Steve Dalby:

Yeah, it was very rewarding. Was there was there a moment where you went, what a bloody idiot. Why didn't I do that to start off with?

Anouck Fierens:

Yes.

Steve Dalby:

Yeah, I get.

Anouck Fierens:

Of course.

Steve Dalby:

That's the learning process of it.

Anouck Fierens:

That's the learning process.

Steve Dalby:

Hashtag screwed up. Now let's move on and waste on what we've learned to move forward.

Anouck Fierens:

And there was also a moment that you just want to throw away your computer. Because you are sick of it.

Steve Dalby:

Time for a new one. No, no, I get that. No, this is cool. This is quite interesting. So if we if we was thinking about our design lifecycle, we need to think about refinement of the solution, but we also perhaps need to think about re-refinement.

Steve Dalby:

So when you do an iteration of code, that's fine, but when you do a second iteration of code that maybe is the next stage or the next solution is to actually go stop. Can we be efficient by combining those two together into one structure so that when we build the next one I'm down. So I guess a golden rule you've learned here is, you know, don't have too many make sure you review the structure that your low code is dropped into.

Anouck Fierens:

Yeah, that's something we definitely learned from it.

Steve Dalby:

Yeah, and we'll put that I think into this design lifecycle. It's all part of the testing process, of course. And I think you're right. I think As an agile perspective, your team would say, hey, we're getting to the end of the sprint and this is not working. And so there's an inherent trigger point where you go, guys, we are not going to be able to kick this into the neck to this sprint.

Steve Dalby:

So either we it's legitimately much more difficult than we thought, or we take the process of rethinking this through, even if it so the next sprint we can redo it. So that's a team. But on your own, It is a personal

Anouck Fierens:

It's a personal decision and you need to be ready to throw things away. It's not easy to throw your work away, what you have done, because you spend a lot of time on it.

Steve Dalby:

Yeah, and you can't charge the customer for stuff that you've thrown away necessarily.

Anouck Fierens:

Yes, but yeah, we you really need to think about it and you really need to consider is this where you want to go? And before you do throw it away, make sure you have another way to go and have another view on it.

Steve Dalby:

Neat. I like. I like it. I mean we talked about this. We thought it'd be a great subject for people. Hopefully people have learned from it. I mean, they're probably all sitting there going, well, of course you throw it away and start again. But, you know, it's not always easy. And we but as human beings, in, you know, in IT we tend to have egos. You know, some people bigger than others, software development egos are huge.

Anouck Fierens:

And of course, also telling your customer that you've just thrown away the work you have done because you don't see a solution anymore. It's not easy.

Steve Dalby:

No, but hopefully you're open and honest enough with your customers. And the agile process is about being honest and open. So maybe that's something that can be done. But then but you are right, you know, at some point you have to sit there and not just say, no, I fixed this, no problem at all. You kind of need to be say, look, this was a pain in the butt. But and I used a different, completely different approach and that got us to this point.

Steve Dalby:

So we're better off now, ready to move forward with a stronger platform. Good, good, good. Well, I thought that was great. It's a very short podcast this time, this episode, nothing too major, but I think a great learning point.

Anouck Fierens:

It is, it was a great learning point for myself. And also making people aware that there is nothing wrong with doing things like throwing it away and starting again. There's nothing wrong with it. No, it's too long a

Steve Dalby:

title though, throwing it away and starting again, but we have to come up with something of a similar ilk.

Anouck Fierens:

If you are stuck in something like this, you can start asking other developers to get involved. They have another way of viewing things. But you just can rethink it first yourself.

Steve Dalby:

No, I think that's a very valid point. It's always worth getting a second set of eyes on something to just clarify where you're at. So that's good. Alrighty. Good. Well, yeah, Steve's Dolby, Steve Dalby who has no idea any of this code's talking about, but as you can hear, there's a real value in bringing that kind of just logical flow approach to the solid, you know, development life cycle of code.

Anouck Fierens:

Therese, and you do know the value of this because you were one of my discussion points. So I was I was telling you I'm going to do this, this, this, this, this, and you just stood there and listened and asked, oh no, that's not right, I need to do this.

Steve Dalby:

So No, that's good. It's all part of the fun, it's part of the sort of collaboration and aiming for the scope, of course, that sort of end result. So again, if you want to hear us talking about business and developing in one space, we're gonna be at Collab Days Bremen.

Steve Dalby:

where we're gonna be literally talking about a it might even be the same example but a similar example how we took that business approach integrated it into the design life cycle of coding and cutting it and how we then actually delivered the final solution so yes that's at Colab Days Bremen so maybe we'll see you there.

Anouck Fierens:

Yes. And if you they have any questions about this they can just ask us there.

Steve Dalby:

How can they get old of you?

Anouck Fierens:

Follow us on socials.

Steve Dalby:

On the socials. Actually, what we did try and set up the we've set up a WhatsApp group for the podcast. So if you search for FusionTalk in the WhatsApp chats, is that what they call it? I can't remember what it's called now.

Anouck Fierens:

Chat of communities.

Steve Dalby:

Communities. Well done in the WhatsApp communities. So yes, FusionTalk, join there, leave us some messages, we will keep an eye on that. And yes, you know what the names are and it's in the chat anyway, so that might work out.

Anouck Fierens:

And if they don't want to go to WhatsApp, they can find us on X and LinkedIn as well.

Steve Dalby:

All over the place. All over the place. Good. Alright, Steve Dalby saying hi and bye and Goodbye, everyone.

Anouck Fierens:

Welcome to FusionTalk.

Steve Dalby:

With a nook and Steve.