I maintain a few Tor middle relays on different broadband connections, so I’m relatively familiar with running the Tor software on small Linux and FreeBSD servers.
My next project might be to set up a bridge relay, since I assume they’re a quite scarce resource. I want the Tor network and its users to get the most out of the bandwidth I can donate, so here's a thing I've been pondering.
To my understanding, bridges have a tendency to get blacklisted by oppressive entities relatively quickly.
So I’m wondering: would I provide a useful service to network at large by hosting a bridge relay and then changing its ip address every now and then, for the purpose of having a "fresher asset" to offer the network?
Specifically: it would be trivial for me to set up a bridge relay on a broadband connection. I could do something simple like presenting a different MAC address on the WAN interface to get a different ip address from the ISP’s DHCP service, and perhaps even automate this process to take place at random every 30-90 days.
I mean: Is there a sufficiently large variation in how sophisticated adversaries are in blocking bridges for users who need them, for the Tor network to make actual use of bridges with IPs that fluctuate within a single ISP’s regional address space?
Or is this a futile exercise that just would get traffic to and from my ISP’s entire customer IP address space filtered and scrutinized by the kinds of networks that already force its users to rely on Tor bridges?