Navigating the New Features of Camunda 8.5 (S3 EP1)

Ready to revolutionize your business processes? Camunda 8.5 has arrived and it's packed with groundbreaking features set to enhance your operational efficiency like never before! Together with the savvy minds of Max Young, Mark Lucente, and Dmitry Molotchko from Capital BPM, we dissect the latest multi-region capabilities, zero-downtime updates for Zeebe, the trailblazing potential of multi-tenancy, and more!

You'll discover a treasure trove of user-centric advancements that promise to empower business users with elegance and ease. Plus, the revamped task list now featuring candidate groups will have efficiency enthusiasts buzzing with excitement.

Transcript

0:06

So I'm here today with Mark Lucente, uh, Dmitry Molotchko and myself, Max Young from Capital BPM, and we want to talk about the new release of Camunda 8. 5. Uh, Mark and Dmitry, would you just, uh, introduce yourselves briefly, please? I am CTO Camunda, of, of Capital BPM, uh, and I'm responsible for technology. Mark Lucente, I'm one of the principals in the company, and I spend a lot of time Across, uh, multiple project teams, methodologies, and, um, some of our accelerator assets in our consulting group. I'm Max Young. I'm also one of the principals. I'm a Camunda champion and a certified Camunda architect. So, guys, I guess the, you know, the thing that I was really interested about today was, uh, the announcement that Camunda has released the 8. 5 version. And, you know, they're calling it a minor version, but it doesn't actually seem that minor to me. There's a lot of changes that have gone into this versus the last one, you know, eight, four, three or eight, four, six, which had, you know, fairly small steps, what are some of the things that you're most excited about? All right. I'll start with a nerdy topic. And my summary on this would be that, uh, in the eight, five release documentation, they talk about being able to go multi region, uh, with the run times and. Most software vendors really struggle with this. Therefore, I'm, I'm interested to see how they do it, how well it works, how it works in different Kubernetes engines, you know, cause each cloud provider tends to have a different Kubernetes engine without it becoming complex stretch cluster kind of conversation. Yeah. A lot of times these conversations will devolve into. Block storage replication and all these super technical cloud architecture terms and, uh, at least the way that it's being described early on in the 8. 5 docs. It looks like it's going to be an inboard and built kind of capability. I haven't got a chance to look at the helm charts yet, but I'm looking forward to seeing what that looks like. Yeah. You know, kind of building on that, something Dmitry and I talked about a while ago was ways that you could deploy like different versions of the process without it becoming the official version so that different people could work on a process diagram. And, you know, with multi tenancy that becomes possible, like you could potentially like deploy your version of a process and, you know, to a, to a tenant and display it and deploy it over there and get feedback from it. Right. And that kind of eliminates the need for everyone to single thread through a single deployment at a time. So I thought, I thought that was kind of interesting. Yeah, I agree with you guys, uh, but I want to highlight, uh, really great features that we have, uh, and this feature connected to this, uh, delivery and updates. If you, uh, as you know, uh, the ops acquired, uh, let's say This is a big area of knowledge, right? And, uh, it seems for me that Camunda starts to follow knowledge of DevOps strategy and methodology. Means that, uh, they added roll up feature for Zeebe and now, uh, it literally doesn't take any time. Yeah, yeah, I, I, I'm really excited to get our local installation and start playing around with it. I think this is going to be really sort of powerful. I'm also kind of excited about the security aspect, the OIDC provider stuff. You know, one of the things that has been hard is that at least with the Camunda 8 SaaS, there isn't really a concept of like roles that you can access readily and that you can sort of tie candidate groups into and all that kind of stuff. And it seems like they've made that possible now. And I'm pretty excited about that. As I'm looking through all the things that they're releasing here, my mind is sort of drifting to the lighter side of the news. Actually, there are 2 things that are in here that I can I'm personally going to eat up. As you guys know, I'm always joining on about. Can we make these things really useful for business audiences and we have different soul searching conversations about. How much a business person can do versus a person and I love the. It seems like like eye candy, but I love the expression component and the forms people like me eat that stuff up. I just don't spend enough time writing expressions to be really fast at it. As you know, I spend a lot of times and I'm in product and project delivery. And that's just not enough repetition for me to be fast with in the modelers. Right, so that kind of stuff is, I know it's probably a lightweight thing for you guys, but for me, it's kind of nice and it's really a time saver. So that's one of the two. The other one is Optimize. As you know, I'm always droning on about Optimize and how we could be using Optimize more, not just we, but our customers in particular, to get more out of it than they often do. And Optimize now has a wizard as well, because if you tried to create collections and metrics and analytics in the previous version of Optimize, Um, it was a little bit mind bending at times, and, uh, it's it was due for, uh, an improvement in terms of the user experience of developing the metrics reports, and they seem to address that, too. And it really does not seem like a small release. I'm just going to echo what you said before. They really covered a series of topics all the way from infrastructure for security through task lists, wizards. I mean, this is a pretty comprehensive. Release and I think, you know, we've been following it even in the, in the sub releases all the way up to here. So I know that our team has been readying for this, but this is a pretty exciting, pretty good size release. Yeah, you know, you talk about sort of the the eye candy stuff. Um, I'm excited about what they're doing with the task list, like the introduction of candidate groups. I, I'm curious to know if the candidate group visibility will be enforced or just possible, which is to say, you know, currently when, you know, we get a task to get all the tasks. But if it's going to be enforced, if you're only going to be able to see tasks for groups that you're a member of, I think that's going to be really compelling. I think that's going to be really powerful. So I'm pretty excited about that. Yeah, I noticed in there there wasn't a sign. There was an unassigned. There was back to a group I'm part of. It seems like it's a lot of the things that we always do. Talk about and lament not being in there. We've always got an early in our methodology. It's about addressing who can do what at what step of a process. It seems like very few companies just think everybody can do anything, right? There's going to be some level of this. I don't care if you're a giant company or a small company, there's some necessity to know who is allowed to do something at a certain point in a process. And so again, this is a pretty big deal. Like we spend a lot of time doing this in projects. And this now gives us. a common approach for that rather than, you know, having to roll each one for a different customer based on how they do their own identity solutions and stuff like this. Now we can start to get into who really should be able to, uh, execute a task and have that be tied into the core runtime engines and security. Yeah. Yeah, absolutely. Absolutely. I guess we could talk about, uh, open source support. Yeah. Uh, you know, that you have experience with open source, uh, with our clients. And it looks as a really good option here if you don't want to use Elasticsearch for any reason. I, I agree and you know, I, I don't think it's binary. I don't think it's Elasticsearch or OpenSearch, right? That can be, it can be both. And I think that's a really compelling thing because you don't have to tear out Elastic to put it in OpenSearch. It can just be a second or a third or a fourth, right? And I think that, that can be pretty powerful. I think the exporters and the, and the targets with Open and Elastic or something that we talk about customers. We have this broad range of customers across. C8, C8 SaaS, and of course on C7, and when it comes to this topic, we have a lot of customers that don't use things based on, like, maybe they don't use optimize at all because it's just, it's too much to deal with. Let's say you're an AWS customer. You want to use open search. You can't use it right now. So you sort of forgo optimize in terms of your thought process of even how to own. And analyze and do metrics for Camunda. And now you're kind of missing out. And all we know in it, once you start a project in motion, a program of usage, a pattern of usage, it's hard to reinsert it. Sometimes, sometimes you don't come back and revisit it. That's like, Oh, well, optimize wasn't part of the project eight months ago. So now we're going into phase two and we're just not going to, you know, bring up the mic into the project. And we it's on over and over again. Yeah, you're totally right about that, Mark. That, that's a really subtle point because a lot of times you architect in or you architect out based on an assumption of technology, right? And then even when that technology becomes available, you don't often have the time or the latitude to go back and sort of reconsider everything and start pushing it back in. That can be its own sort of insidious form of technical debt. Yeah. And I know it was a rare circumstance'cause there, there was this, this moment in time where open search on AWS was challenging for certain customers, but it was just several projects have sailed without the benefit of o of optimize, for example. And, and that's, uh, you know, it's unfortunate it's hard to call back. I got one for you guys, Play in 8.5. Being able to do things like rewind, being able to do things like collaborate with your co workers. This sort of feels like more back in my, uh, area of what I call eye candy. Um, I know it seems like a, a minor thing, but when you're in a sort of a project delivery role where you're looking at collaboration among team members in the sense of increasing the wheels of progress and trying to get things done. Think about how much time do we spend trying to generate, you know, synthetic data payloads. That will work with processes. And then you have to go ask a business person, what is this realistic? Or, you know, it's enough to get it to go through, but it's going to branch the process differently. I even noticed now that you did, there's an API, a Zeebe API rest API that you can, you can affect the data to branch the process on its way through. So in this new world, we could actually be collaborating on input data to a process, paid it, and then actually not break it, but affect its path. What the later A. P. I. Call, which we've talked about doing many times in our processes where we go say, Oh, well, that process has already sailed. We can't call it back. The data that's with it is with it. And now it looks like we can come in the side door with another. If you had a long running process where you had to inject a new piece of data into a variable that would then alleviate or affect or approve its behavior, improve its behavior. Now you can. Well, you can do that and then, you know, there's the migration stuff, which is new, which I think will be very powerful because you can basically migrate it to like an entirely take your live data, sort of freeze time, teleport it over to a new process instance and start running it from there. I mean, that's crazy. The power of that is like telling people about email in 1989, like it's, it's sounds simple . It's really revolutionary. I think this is going to be a big change. Um, I'm, I'm actually really excited about that. I think that's going to be really cool. And I'm also excited about the fact that they're making like documentation easier, right? So in the modeler now you can go and like, it's just more clear and you have a better editor for documenting where your steps are and what they're actually going to be doing. And I think that's really compelling. So I'm excited to see this. I, you know, I think that over the next couple of days and weeks, I want to do little videos, little meetings like this, where we actually bring some of this stuff up and sort of click around in it and sort of show it at a kinetic level. I think that would be a lot of fun.