A one-time perpetual license with 100% source code sounds like the ultimate escape from the Microsoft 365 tax.
But real-time audio, video, and messaging aren’t static tools—they are dynamic infrastructure. Operating WebRTC gateways, mobile background states, and push sync across OS updates never stops costing engineering hours.
This raises an interesting question:
At what team size does owning and operating the software become more expensive than paying for a SaaS solution on a per-user basis?
For example, if a self-hosted solution requires a dedicated engineering/DevOps team to maintain:
-
Real-time messaging infrastructure
-
WebRTC/audio/video services
-
Push notifications
-
Mobile SDKs
-
Database and storage
-
Security updates
-
Monitoring and observability
-
High availability and disaster recovery
then the perpetual license is only one part of the total cost of ownership.
I’m particularly interested in the economics of commercial self-hosted communication platforms such as MirrorFly, which promotes full source-code access and perpetual licensing.
Most discussions around self-hosted team communication tend to focus on open-source platforms such as Matrix, Mattermost, or Zulip. I’d like to understand how commercial self-hosted offerings compare when we include engineering salaries, infrastructure, maintenance, upgrades, support, and opportunity cost.
I’d be especially interested in real-world numbers and experiences rather than vendor marketing claims.