NATS Terminology Clarification

[gcolliso]

We will use this discussion to list NATS terminology that may need additional clarification and thus need additional documentation.

[nav9]

Thank you for this. Starting off:

These points are from the Request reply page.

1. It would help to understand more about “drain before exiting”, to explain it to newcomers. How do we explain this? What kind of resources does NATS drain from memory before exiting?
2. Similarly, “NATS allows multiple responders to run and form dynamic queue groups for transparent scale up” needs simpler language. Could we say "Multiple NATS responders can form dynamic queue groups. This allows the responders to scale up as per demand. ". What does the "dynamic" mean in "dynamic queue groups? Does each publisher and subscriber constitute a group? So when a subscriber replies to a publisher, is that message treated as being part of a queue? Or does the publisher maintain multiple queues to cater to the responses from each subscriber?
3. One would also need to know what "observability" is in “observability is as simple as running...”.
4. When mentioning “Try NATS request-reply on your own, using a live server”, one would want to have a link to the live server (if the NATS team intends to make it available), or the user needs to be pointed to a page which helps them setup their own live server (which page would you recommend?).
5. The diagram on the page does not capture all aspects of what is explained. We need to show the queue groups and show some way in which the scaling up happens. Any suggestions on what you'd like to present as part of the diagram?

[timove]

IIRC it was confusing for me at first that there are streams in Jetstream as well as in export/import.