# Flowable Modeler vs Activiti Modeling App

**URL:** <https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680>\
**Category:** Flowable Application\
**Created:** [March 20, 2020, 6:41pm UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680 "2020-03-20T18:41:38Z")\
**Posts on this page:** 10\
**Page:** 1

<div class="post-metadata">

**Author:** ![chrisw](https://avatars.discourse-cdn.com/v4/letter/c/919ad9/32.png) [@chrisw](https://forum.flowable.org/u/chrisw)\
**Post date:** [March 20, 2020, 6:41pm UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/1 "2020-03-20T18:41:38Z")

</div>

TL;DR

1. Why is the flowable modeler abandoned (2 years ago) in Angular 1 while the Activiti-Modeling-App is under active development at Angular 7 ([https://github.com/Activiti/activiti-modeling-app](https://github.com/Activiti/activiti-modeling-app))??

2. Can we embed the Activiti-Modeling-App in our modern (Angular 8) web app, and use it to edit processes we run with the Flowable 6 engine on our server?

Longer:

We are trying to upgrade our embedded engine and UI from Activiti 5.15 to Flowable 6.

We’ve got the engine working and are at the point where we want to add in the modeler.

We used to use KIS but deprecated it a while ago and want something modern that fits in our Angular UI (and has an open license).

Other posts on this forum discuss the option of runnign the Modeler _application_ or embedding the JS.

We don’t want a separate application, because we don’t use the task app, and the navigation, load/save lifecycle, etc with the separate editor would be clunky (among other reasons).

The only way we could embed the Angular1 UI would be to treat it as a separate single-page-app and intercept Save actions (assuming there is not a lot of other ajax we would have to intercept… this is basically what we used to do with KIS)

Of course it would be easier (presumably) if the Flowable Modeler were a more modern Angular 7 application, and better in general if it were being actively developed.

The we noticed that the Activiti-Modeling-App is going strong. What’s up with that? Can we get the best of both worlds and use the Activiti 6 modeler and the flowable 6 engine?

Is anyone else doing this?

---

<div class="post-metadata">

**Author:** ![Robinyo](https://avatars.discourse-cdn.com/v4/letter/r/dec6dc/32.png) [@Robinyo](https://forum.flowable.org/u/Robinyo)\
**Post date:** [March 21, 2020, 1:37am UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/2 "2020-03-21T01:37:01Z")

</div>

> [@chrisw](#):
>
> Why is the flowable modeler abandoned

 ![Screen Shot 2020-03-21 at 12.23.18](https://canada1.discourse-cdn.com/flex035/uploads/flowable/original/2X/1/10949a35a709de34fbe712b36a615cfe84ff7df7.png)

Looks like there was some activity as recent as 6 days ago.

**Note:** Flowable Modeler uses the [Oryx Editor](https://github.com/flowable/flowable-engine/tree/master/modules/flowable-ui-modeler/flowable-ui-modeler-app/oryx-editor-code).

> [@chrisw](#):
>
> Can we embed the Activiti-Modeling-App in our modern (Angular 8) web app

The Activiti Modelling Application is built using Angular and use the [BPMN.io](https://bpmn.io/) tooling (check the application’s [package.json](https://github.com/Activiti/activiti-modeling-app/blob/master/package.json)).

Ref:

- [SlideShare - Activiti Modeler Application - 2019](https://www.slideshare.net/AndrsPopovics/alfrescoactiviti-modeler-application-andras-popovics-2019)
- [An example how to integrate bpmn-js with an Angular application](https://github.com/bpmn-io/bpmn-js-example-angular)
- [Angular bpmn-js component with Custom Properties Panel Support](https://github.com/central-factory/mono/tree/master/libs/ngx-bpmn-modeler)
- [A more complete example of integrating with an Angular app](https://github.com/bpmn-io/bpmn-js-example-angular/issues/18)

---

<div class="post-metadata">

**Author:** ![chrisw](https://avatars.discourse-cdn.com/v4/letter/c/919ad9/32.png) [@chrisw](https://forum.flowable.org/u/chrisw)\
**Post date:** [March 21, 2020, 10:12am UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/3 "2020-03-21T10:12:50Z")

</div>

Thanks for taking the time to work up such a thoughtful reply Robinyo. The links are particularly helpful.  
Some followups.

On my abandonment issues: The Flowable modeler is clearly no being worked on. That 6-day-old change was not in the javascript app itself, which has barely been touched in 2 years if you see here ([https://github.com/flowable/flowable-engine/tree/master/modules/flowable-ui-modeler/flowable-ui-modeler-app/src/main/resources/static/editor-app](https://github.com/flowable/flowable-engine/tree/master/modules/flowable-ui-modeler/flowable-ui-modeler-app/src/main/resources/static/editor-app)). And it’s in Angular1! (aka AngularJS)

I’m not arguing this to be difficult, but I want some input from the Flowable developers as to what is the roadmap for the modeler.  
We don’t want to commit a serious lot of developer time to replacing our old abandoned editor (KIS BPM) with a more-recently-abandoned alternative, only to have to go through it all again next year.  
Again, Flowable people, what’s the roadmap?

I mention the activiti modeler, because I understood there was continued back-and-forth merging between the activti and flowable repos after flowable forked… are the flowable engineers involved in code used in activiti?

Also concerning: what does this imply for the two engines? Is it possible there is more development on the activiti engine than the flowable one? Are we making a mistake moving from Activiti to Flowable?

Lastly: is the activiti-modler UI fully compatible with the Flowable backend?  
Does anyone know of any advantage in using the activiti-modeler vs bpmn-js alone? It seems to have a lot of additional source code.

---

<div class="post-metadata">

**Author:** ![Robinyo](https://avatars.discourse-cdn.com/v4/letter/r/dec6dc/32.png) [@Robinyo](https://forum.flowable.org/u/Robinyo)\
**Post date:** [March 21, 2020, 12:10pm UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/4 "2020-03-21T12:10:25Z")

</div>

@chrisw

> [@chrisw](#):
>
> I’m not arguing this to be difficult, but I want some input from the Flowable developers as to what is the roadmap for the modeler.

No, I didn’t think you were 🙂

But in fairness to the Flowable team they are just making use of some existing open source tooling.

The Oryx Editor was developed by the guys that went on to found Signavio. The original [code base](https://code.google.com/archive/p/oryx-editor/) is no longer being maintained.

If you search “GitHub Oryx Editor” you will find a few git mirrors of the Oryx Editor project.

> [@chrisw](#):
>
> what is the roadmap for the modeler

Here’s a [link](https://github.com/flowable/flowable-engine/wiki/Flowable-roadmap) to the Flowable roadmap. It doesn’t mention the Flowable Modeler application (nor any of the other Flowable UI applications).

If you take a look at the Developer [documentation](https://documentation.flowable.com/dev-guide/3.5.0/175-flow-app-with-react.html) for Flowable’s commercial offering you will see that they favour React over Angular.

> [@chrisw](#):
>
> I mention the activiti modeler, because I understood there was continued back-and-forth merging between the activti and flowable repos after flowable forked…

Really?

---

<div class="post-metadata">

**Author:** ![PHH](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.flowable.org/phh/32/94_2.png) [@PHH](https://forum.flowable.org/u/PHH)\
**Post date:** [March 21, 2020, 4:53pm UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/5 "2020-03-21T16:53:16Z")

</div>

I think there was only one-way merging from Flowable 😉

And if you look at what’s changed in Flowable since we (the original Activiti team) forked, then it’s very clear which are the more advanced and active BPM engines.

We’re trying to work out what’s best for the modeling app, it mostly comes down to resource and time. Use of the bpmn io code has licence restrictions that prevent us from working with it. Would an embedable but non-open source option be viable for you? (just to help us think about options)

Cheers  
Paul.

---

<div class="post-metadata">

**Author:** ![tijs](https://avatars.discourse-cdn.com/v4/letter/t/bc79bd/32.png) [@tijs](https://forum.flowable.org/u/tijs)\
**Post date:** [March 21, 2020, 4:53pm UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/6 "2020-03-21T16:53:38Z")

</div>

Hi @chrisw,

The focus of Flowable is on implementing the engine projects with the Flowable 6 BPMN engine, CMMN engine, DMN engine and the new event registry. The Modeling, Task and Admin app have been developed with a lot of effort a couple of years back and now we maintain them when bugs are reported etc. Upgrading the Flowable Modeler app from Angular 1 to a more recent Angular version is a major effort and it would require also help from community members. If you would be willing to help us upgrade to a more recent Angular version we would be happy to put in effort as well.

Best regards,

Tijs

---

<div class="post-metadata">

**Author:** ![Robinyo](https://avatars.discourse-cdn.com/v4/letter/r/dec6dc/32.png) [@Robinyo](https://forum.flowable.org/u/Robinyo)\
**Post date:** [March 23, 2020, 1:18am UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/7 "2020-03-23T01:18:29Z")

</div>

Hi @PHH @tijs

> Upgrading the Flowable Modeler app from Angular 1 to a more recent Angular version

Could you set up a topic - “Reimagining the Flowable Modeler” 🙂 - where we could discuss this?

Suggested Features:

- Angular CLI
- Progressive Web App
- Authentication (e.g., OpenID Connect)
- Authorization (e.g., OAuth 2.0)
- Contemporary UI (Angular Material → Material Design Web Components)
- Custom Themes
- Code Splitting (Lazy Loading Angular Libraries)
- Dynamically Importing Static Libraries (e.g., oryx.js)

For discussion:

- Migrate Oryx Editor code base to Oryx Editor [Web Components](https://developer.mozilla.org/en-US/docs/Web/Web_Components) (so that it can be used by other [frameworks](https://material.io/develop/web/docs/framework-wrappers/))

I may be able to help out:

 ![image](https://canada1.discourse-cdn.com/flex035/uploads/flowable/original/2X/0/02374ab1bc13ce20384b494532941c3f32ec396e.png)

Cheers  
Rob

---

<div class="post-metadata">

**Author:** ![chrisw](https://avatars.discourse-cdn.com/v4/letter/c/919ad9/32.png) [@chrisw](https://forum.flowable.org/u/chrisw)\
**Post date:** [March 23, 2020, 9:13am UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/8 "2020-03-23T09:13:37Z")

</div>

Thank you all for the thoughtful replies and the clarification they bring.  
With these, and a good look at the github repos, it’s become clear to me that Flowable is the state-of-the-art where the engine is concerned, so we’re doing the right thing upgrading; while clearly Activiti-Modeler is the state-of-the-art where UI is concerned (even when compared with the bpmn-js layer it is built on). This kind of makes sense, since I agree (and have seen how) upgrading such a UI is a very costly undertaking, and it makes sense for Alfresco to invest in it, given their product and customers.

It still leaves me with the one open question:  
_Is there any experience out there - or opinion on the feasibilily of - using the Activiti-Modeler - with the Flowable engine?_

To @PHH’s specific question: non-OS would be a non-starter for us. One of the reasons we dropped KIS BPM was it’s restrictive license: that made it very hard for us to integrate and maintain it.  
Quite apart from the limitations in modifying and adapting code for integration, non-OS would raise security concerns.

Thanks again for taking the time to all weigh in - if anyone as any clear knowledge one-way or the other on activiti-modeler-with-flowable vs bmpn-js-with-flowable, be it from a technical, practical or legal standpoint, I’d appreciate it.

---

<div class="post-metadata">

**Author:** ![PHH](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.flowable.org/phh/32/94_2.png) [@PHH](https://forum.flowable.org/u/PHH)\
**Post date:** [March 23, 2020, 9:28am UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/9 "2020-03-23T09:28:41Z")

</div>

We’ve opened a new forum category to focus thoughts and efforts around getting an updated modeler. We do want to get something moving that is completely unrestricted.

[https://forum.flowable.org/c/modeler-remodeled/9](https://forum.flowable.org/c/modeler-remodeled/9)

Let’s see if we can get some momentum behind that.

Cheers  
Paul.

---

<div class="post-metadata">

**Author:** ![PHH](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.flowable.org/phh/32/94_2.png) [@PHH](https://forum.flowable.org/u/PHH)\
**Post date:** [March 23, 2020, 9:31am UTC](https://forum.flowable.org/t/flowable-modeler-vs-activiti-modeling-app/5680/10 "2020-03-23T09:31:04Z")

</div>

New category created and initial post: [here](https://forum.flowable.org/t/thoughts-on-updating-the-modeler/5689).

Cheers  
Paul.
