Name:
Optimizing Scholarly Publishing Workflows: A Deep Dive into BPMN for Journals
Description:
Optimizing Scholarly Publishing Workflows: A Deep Dive into BPMN for Journals
Thumbnail URL:
https://cadmoremediastorage.blob.core.windows.net/4444eabc-e5b4-47c1-ad50-be3a0d32ab53/videoscrubberimages/Scrubber_1.jpg
Duration:
T00H31M48S
Embed URL:
https://stream.cadmore.media/player/4444eabc-e5b4-47c1-ad50-be3a0d32ab53
Content URL:
https://cadmoreoriginalmedia.blob.core.windows.net/4444eabc-e5b4-47c1-ad50-be3a0d32ab53/industry_breakout__sitefusion (1080p).mp4?sv=2019-02-02&sr=c&sig=4ZZX7%2FoXbAc1rcml8z3X45%2F8Mi9Vp8Fe%2B2lcP5cGkWw%3D&st=2025-04-29T21%3A48%3A14Z&se=2025-04-29T23%3A53%3A14Z&sp=r
Upload Date:
2024-12-03T00:00:00.0000000
Transcript:
Language: EN.
Segment:0 .
Good afternoon. My name is Colin O'Neill. I'm a content. Can everyone hear me OK. Do I need to speak into this more. My name is Colin O'Neill. I'm a content solutions specialist with site fusion pro consult, a full service content management solution provider, and I'm joined by Mario Candler, who's a CEO and founder for site fusion.
CMS site fusion is a CMS solution which uses a bundled solution of Camunda process modeler to handle all. Louder to handle all sets, all parts of the publishing lifecycle. Has anyone here heard of a business process. Thank you. BPM OK. One, one or two.
So what we'll do is we'll start off and we're going to explain what BPM in is and then how you can use it as part of your academic publishing needs. So I'll introduce Mario Candler, who's going to talk a little bit about the background of BPMN and first. Thank you very much, Colin.
Mark Handler my name's CEO and founder of site fusion, and I would like to introduce you today a little bit about BPMN and how we use it, creating processes for all our publishing publishers, publishing companies. And Colin already asked you, you heard from someone who has already worked with BPM or no one. Well, OK. Already a little bit worked, so.
Yeah, I expected. The next. Doesn't jump to the next. In the space, although not. Yeah, that's not working.
Just the tick tick guy. All right. Yeah usually I can use PowerPoint. It's not my first time using it. Yeah we have a little problem again. Jumping to the next page. Doo doo doo doo doo doo doo.
When in doubt. To do. Just make sure I have that folder up and to do. Slideshow from beginning. And you should be able to advance just fine there. Thank you. All right.
Thank you. OK two or three words about BP and what's behind BPM. BPM is a standard and part of the OMG. The OMG is a company founded already in 1989 and develops a lot of different standards like BPMN and much more. And about 800 companies working on with the OMG with the standard company and companies like IBM, Apple, Camunda, Microsoft and we site fusion are already a year part of this community working on this standard and try to bring in ideas for our projects and publishers.
BPM from 1.0 to XML. This means the first version of BPM was only like drawing images. It's like Microsoft Visio. You couldn't do anything, only visualize the diagrams. But with the version 2.0 in 2011, the BPM was an XML standard. This means that what you create in the model is XML and based on this, XML machines can run on it and do some automation with this XML standard.
And since the year 2013, it's an ISO standard. That means what you're doing with BPM can. It's not an idea what we thought to create. It's an international standard. So you can create diagrams using this for a lot of different engines and to run your processes based on this standards in a few slides. How it works will be introduced by Colin and when I will show you how we use it in a sample program.
Progress for Yeah, scoring content. So BPM. BPM stands for business process model and notation and is made up of four main components. These are events, activities, gateways and flows. So an event would be something like start intermediate or end. It is not.
This would be like a signal coming in at the beginning of a process. So if we're talking about submissions or something like that, an author submits. A manuscript and this would kick off the process. And then we set up the different pieces of the nodes. So the activities, the gateways and the flows to follow in line after that.
So sorry. Just can't scroll. Sir so sorry. An event represents something that happens during a business process.
Events are critical elements with BPM as they indicate when why, and how the process flows should be started. An activity represents a task or a work that is performed within a business process. Activities are fundamental building blocks to BPMN diagrams depicting activities is generally classified into two types.
A task or a subprocess. A task is a basic unit of work within a BPMN diagram. A task is a single atomic piece of work that cannot be broken down further. It's a approve invoice, send email process payment, something like that. A subprocess is a more complex activity that can be decomposed into a set of other activities or even an entire process itself.
So processes allow for modularity and reuse within the BPMN module model. They can be either collapsed, shown as a single activity or expanded to into their own process flow. The third thing would be a gateway signified by this diamond here or diamonds in typically a gateway is a modeling element that uses the control to control the flow of a process based on certain conditions or events.
Essentially, gateways facilitate the branching, merging, forking and joining of paths within a process. And then we have flows. A flow represents the movement and the sequence of activities, events and gateways within a business process. There are two primary types of flows in BPMN and sequence flow. Which is symbolized by an arrow, a solid arrow.
The function of a sequence flow is to indicate the order in which the activities, events, or gateways are performed within a process. In essence, it connects the elements of a process to show the direction of a flow from one step to the next. And then you have a message flow, which is the dashed arrow. This represents a flow of messages between different process participants, typically between different pools.
So this is different departments, different organizations. So not only are you passing things on to a user, you're passing it on to a pool of users, and out of those pool of users, you can set up rules that either someone needs to manually take the task or it automatically gets assigned. These are all fundamental pieces that you can write in with any type of code.
This could be PHP, JavaScript or even low model code to do stuff like that as well. So one of the reasons that we wanted to talk about it here is that there are a lot of challenges that come in with academic publishing and publishing in general. So with Fusion pro console, we do work with academic publishers. We also deal with education and legal publishers and stuff like that.
But all publishers kind of have the same challenges that they want that we need to address. And we want to address these things in a programmatic way that is also flexible enough to adapt to not only error handling, but just kind of like bespoke things that happen as part of the manuscript, as part of the publication process. So one of the things that we're trying to focus on with BPMN and authoring with BPMN is that we can.
Sorry so. Sorry So it has the flexibility and the customization. So as we go through and we have different aspects that are coming about, like I said, we can develop the process. So that based on anything that comes in. So if we want to handle an error, can also if we go to the next slide.
So this is a bigger process tree where we have different decision points and different actions that are handling here. So one of the things that we can do through automating with BPM so we can handle copyediting tasks. So if an author submits a manuscript, one of the first things that we do is we want to check it for grammar so we can either have a copy editor or a proofreader or someone like that go in and do this.
We can also connect to third party systems where we would go out to Grammarly or one of the other third party systems that does this. We can also, if we have pictures coming in as part of the manuscript submission process. But we know that in all of our publications we need PNG or SVG or whatever it is and someone sends in a JPG, we can take a JPG, we can send it out to an external service, automatically, convert it, change the format, change the dimensions, do all that stuff and the whole time we are the backbone of this is all through the workflow.
So we can set up the processes that communicate with these different things. One another one another piece that often gets used is citation validation. So communicating with Crossref or PubMed or anyone like that to make sure that all of your cross references and links are all not only accurate but up to date. And then we have task alerts where we can send out messages, email messages, or it could be messages within a system.
So for us it's site fusion. And we have an internal messaging system, but we can also do it with text messages, whatever. You just have to sign this up to a different service and have this communicate that way. Now, because I kind of went into that a little bit here. But so. Because we can send it out to third party services. We can also just build bespoke applications within the application suite.
So if you do have internal dictionaries or anything like that, it doesn't have to go out to a third party system. You can just send it in here and build something to go through it like that. I'm going to go back to Mario, who's going to talk a little more hands on about how BPM gets used. And for this, I have a little simple and simple article workflow process designed so it's only a short process with which shows us some loops and steps for creating and checking an article.
This workflow is here a little bit bigger. You have checked the metadata. Then we create an article by an author. We have an XML review by an editor process. We create automatically a PDF out of the article. If we decide that the author should prove or check it again, we give them the task him the task. And if the author is proving it, it gets, for example, to an external service like PubMed.
Create again a PDF document with the enriched document. And at the end we create main versions and can publish the content. It's a really short process, but the cool thing is all the tasks we create, we can reuse, we can change them in wherever you want and however you want. And for this to create such a process. You have this here, a user task check metadata and already within this user task, by creating the BPM standard, you have the possibility to define fields and variables.
So if you want the interface later in the application has some metadata values, you can define them here on the right side. I wanted to show it in but I have a video now, so if you go to the check metadata in your BPMN editor, you can add on the right side some more detailed information about what kind of fields or publishing date. It's a date format.
So you define in your process already all the information you want to get in this step in your workflow process. So this means that. Define defining a process already is part of a development of a source code so you don't have to write source code and deploy it to application. You can do this all in this international standard.
BPMN so for this we have user task, create metadata or a user task, create article XML review and. Each task can be handled with an editor's, for example, like a word editor. So if you still working with your documents in Microsoft Word, a user gets a task. He can open Word and edit it. The better thing, of course, will be an XML editor like the font editor, which we had have integrated in the system.
So if you get a task and say this article should be edited, you have the rights to edit this task. The editor is opening and you can add all the information within the editor like WYSIWYG. So in Microsoft Word, but only already in XML definition. Another part is not only the user tasks. So user tasks are very easy, very simple to create because this always means that the user has to do something.
But the cool thing from BPMN is for automate automating tasks like the service tasks. So you see the sign. This means that each task can connect with an API. So if you have some services and the services can be reached by an API, you can integrate it in your process and call them, give them your information. This can be a document, this can be a digital file, an image, a video, this can be all kind of information and you get back from the API the result and work with this information in your process.
So for example, we have tons of APIs to Microsoft Cloud, Amazon cloud, but all other tools like ChatGPT, for example. So you can call a ChatGPT, give them your article with the information, what you want to get back, like a summary, like the keywords, and you get back the information and store it in the database. If you create it once this task.
So like such a service task, you can reuse it only by copy, pasting it in another workflow diagram. So you don't have to create or Yeah, a lot of times the same task in your application. This is the XML the result of. But don't worry, no one ever will see this XML and will use it because usually if you have a BPMN diagram and this defined your process in BPM, you have an interface which is working based on the BPMN diagram.
And for example, with site fusion, we have one of our developers said long time ago said fusion is only an interface based on a Venn diagram and showing all the information which you defined in BPMN and that's it. So you have here a list of tasks. So like an XML review, check metadata, create article and check metadata again. And if you claim a task.
So you click on it, you have, the more information. So in XML review tasks, you see which kind of asset this is. And you can see in this step the PDF preview of the XML document. And if you say, OK, I want to edit it, show it in the short video, you click on it, you can update it. These are the metadata fields I've shown in the diagram.
This means if you add something, it's automatically here in the application. And for example, if you go to a review, you can open the task. See the PDF which was automatically created out of the XML content in this step. You can prove this and if it's not OK, you can open the editor and the editor opens the content, which can be edited directly here in the system with all the information.
For example, some notes, some comments. You always see what you changed in the document. If you integrate something new, like here, some new, more things changes. These are all information which you can see on the right side. What has been changed. Who has this changed. You can accept this changes, you can remove this changes and work here in XML like in Word editor.
Of course, you can add comments and answer the comments in the system and. When you've done it, you click on the Save button and then you say, OK, the document is saved, my task is done. And I click on Update and the next step in the workflow process is and that's the next task.
And here the question proved by over. You have seen in the diagram there was the question proved by over Yes or no. If no overjumped this task, if Yes, give it to the over. So all the decisions you define in your diagram are displayed here to decide so you don't have to create or to develop for each task or for each diagram, new interfaces. You only change the model and that's it.
And then you have here your interface for this. A shot few in the back end. This is usually only for administration. You see here the diagram, and in the diagram you see how many documents are on which part. Three documents are at the moment. By create article, and you always have an overview of all your content in which workflow task is the content at the moment, what's the next step.
And so on. So this diagram is created, uploaded to the system and the whole application, all the interfaces working with this diagram. So you manage your workflows, you define your workflows and the result is the application based on your processes. So this is. Yeah in our opinion, the basic for the next steps what we can do with it.
Because there are a lot of applications. A few weeks ago, I've been in a webinar about scoring, scoring information and how process and what's very important. And a lot of people said, yeah, AI is the future. And if you want to integrate AI applications, you need technology like this. So you can very, very easily integrate each kind of API in each position on your process.
Only creating one Swiss workflow task and define where it should be placed. And you can use all this technology in your processes and you don't have to develop all the applications your own. So it's like using a lot of microservices in your environment. Existing services all integrate in one workflow process. Yeah so also sorry.
Also one of the things. So I should have asked at the beginning. Also is anyone working with content in a CMS or in a cms? So if you are storing content in a CMS, typically what this system has done is they have a workflow engine on the back end and it's usually something that's just proprietary to the system and it's very difficult to attach code to it, to it, to a number of those things.
And one of the reasons that we use BPMN is because of the flexibility. Every time every one of those nodes, you can just delete really quickly through a commenda or through some other interface and quickly swap things, swap the order in and everything like that. So one of the things that we wanted to talk about with when we're talking about the future of publishing with BPMN and like the most everyone's talking about AI in some way and the most common way that we're seeing AI being implemented with the BPMN process is we have these.
So like with WCAG standards. So one of the most common things that we need to do is we need to add Alt text to every image. So there is an editorial control where you do want you do want a user to probably have relevant information where they are talking about the information in an image. But what we can also do is we can send it out to a third party service and working with customers and doing this.
What we've noticed is that Yes, you can send it to Google, but sending it to something like Claude, which is the Anthropic or one of the new ChatGPT models, what we have found is that these LLMs are actually much better than the Google APIs for adding this Alt text to images and stuff like that. So it's one of those things where we didn't have to rewrite an entire system to say stop going out to Google for this.
Start using ChatGPT. All we need to do is update a script, put a new, what I'll say is a rectangle into the workflow, into the flow chart. And it was just that quickly. So I think that's really the benefit is just the flexibility of being of using a service like this, which is outside the realm of a straight vertical CMS where everything is decided for you.
Anyone can come in and create new process models and you can connect this not only with your publishing lifecycle, but you can also connect it directly to something like SQL or one, or if you have some sort of interface that's working over SQL or one to do all of the submission handling and stuff like that. There's a very low learning curve with coming in and building out stuff like that. So did you have other stuff.
OK I think this is the most interesting thing to change the services you're using, for example, to creating this PDF document using the standard antenna house for creating out of XML PDF documents. But if you say no, I want to use other applications like typify creating PDF documents and background using maybe InDesign. It's also possible only to change this task in your process and switch it to another.
So you're really independent if you have your processes defined and to change some tools to other tools. And you don't have to change your software application because the complete process defines the way you can work and you're very flexible how you use it. So not the system says you have to work like we think, like the developer developed the program, your own workflow decides and defines how you work and what's the next step in the workflow process with all the information, like a checklist, like a help information, these are all information which is supported by the workflow.
This means you can have a description like I have a task. What should I do. What are my next steps. Like a short description checklist so you don't have to hand out in documentation or other things because in your task you always see what you have to do and what's the next step you have to solve. And that's the easy thing. All in one place.
So questions. Some questions here. You have 50 seconds like or a minute later or more. I think. No, no. Someone has a question.
Yeah bullet states analytics that can also identify. Yeah Yeah Yeah. Of course. The workflow engine we're using is Camunda. Camunda is one of the favorite workflow engines at the moment and we have in the workflow engine all the information, all the metadata.
I always have the information how many tasks. This can be. Maybe documents or other assets like images, how many tasks are at which position. How long are these tasks already there. And we have the heatmap functionality. So you see. Which path is the longest path. Why the overall race needs much more than when expected, and you can combine each task in the workflow with a time.
So a time how long Usually a user needs for this task and which time frame he has for calculating the whole process. So if you get a task for reviewing a document, depending on the length of the documents, you have two weeks. But the time you really need for it may be are 20 hours. So you have two timestamps. And based on this, you can plan all your tasks and see how is the work load already for different users in the company.
More questions. OK I just wanted to add, we do have a booth over in the exhibition center where 107 where against that far wall, if you do want to come by, we can do a more interactive display, one that's not going through PowerPoint. We also have a poster that is available.
So we'll be talking about that at like 130 tomorrow whenever that whenever the poster is going on. I just wanted to Thank you again for your time and hope everyone enjoys the rest of the conference. All right. Have a great day.