Title pretty much says it all. All I can really find is the paragraph here mentioning that the app replicates changes to the DOM model.
I have also been able to find that the application uses a web proxy to pass screen changes at this URI
https://cobrowse-v2.usw2.pure.cloud/proxy?url=
Is there a more detailed whitepaper that describes the process available
The page you linked describes the nature of how it works:
During a co-browse session, co-browse sends any changes to your website’s Document Object Model (DOM) to Genesys Cloud. The DOM describes the underlying structure of webpages. Co-browse then uses this information to seamlessly build the same page that the sharer sees on the viewer’s screen. Co-browse watches for any change on the page, such as scrolling, clicking, or text typed in a form field. When a change to the DOM occurs, co-browse sends the change to Genesys Cloud and updates the viewer’s screen in real time.
What additional detail are you looking for? Are you facing an issue?
Maybe a nerdy whitepaper/traffic flow diagram between client and agent. Does the customer forward the screen elements to the proxy, and then the agent subscribes to updates from the proxy?
We have one or two elements in our webpage that are having display issues still after fixing CORS etc.
The cobrowse offering is proprietary software and is not open source, so there's not technical details that expose the software's trade secrets. If you're having an issue with cobrowse, please open a case with Genesys Cloud Care for investigation.
lol, thanks Tim. Looking for something less than your source code and Github login credentials, and more than a quick paragraph overview of how it works
There's not much more to it than what the Resource Center page says that's not the proprietary part of how it's implemented. It gets the client DOM (i.e. the document object in JavaScript), transmits it through Genesys Cloud, and reconstructs it for the agent to view (i.e rendering the scraped DOM in an iframe). Discussing certain details of the implementation may be appropriate in the context of a support case about it not working properly, but that's not something we can get into on a public forum.
If you're just curious and want to learn more about the concepts of such a thing in general, you'll find lots of articles and posts on the public internet (places like stack overflow) about screen scraping. What Genesys is doing isn't fundamentally different from known approaches, it's just that our specific implementation isn't publicly shared beyond what's in the documentation.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.