Hehehe. This is absolutely true and trying to explain it to anyone around me makes me feel like a Cassandra quoting battlestar galactica: "All of this has happened before. All of this will happen again". I'm not even old, I just happened to witness the tail end of SOA when I was still a student, followed by the rise of micro-services afterwards. Your service mesh is just someone else's message bus. Sure, it's different. Except not really. Also, bonus points for younger developers reinventing features for json APIs that SOAP already had.

I agree that it does among some. But the semantics of "MICRO-services" is that smaller is better and you can find those who take it to be smallest is bestest. You can't just assume the word "micro" is just a hanger on and isn't doing work convincing people of the virtue of super-small. Lots and lots of orgs and people fall into this trap. I've worked at them. I've argued with advocates in the professional sphere. All the time.

If it was literally just SOA, I wouldn't have issues with it as you can reasonable have conversations about where divisions should be placed. Maybe you are surrounded with more reasonable advocates, but that is not the norm in my experience.

When you have orgs assigning a team to build and maintain 20+ services... It's gone into full self destruct mode.

When concept X becomes hot as an alternative to Y, everyone absolutely have to do X or be square. But for large enough companies it's cheaper to influence the industry so that X = Y and we relegate Y to Y', where Y' only contains our bad memories of Y.

This is the "Enterprise Technology Adaptation Strategy".

SOA was all XML at the API level.

Serverless is RESTful, lower-level, and coupled to the cloud-provider's menu of containers, key/value store, authentication, &c.

"Synonymous" seems a stretch.