Which poses the problem of building your user base on a clone site. Where all your first movers would be exactly the kind of people that no one would want to join. Just look at voat, it was a Reddit clone and many people went to see what it was like, but left because of how disgusting the regular voat users were, and since 99% of us aren’t going to get banned from Twitter, why would we leave it to join a bunch of degenerates?
1. Make it work. This sounds easy with open source tech, but really hard if it includes effective content moderation from day one. That's a tough product problem to solve which only the largest consumer companies have addressed with their own proprietary technology.
2. Make it scale. This is hard, especially given the potential large userbase and the fact that none of the major public clouds like Google, Amazon, or Microsoft will likely host him without very strict content moderation policies. He's going to have to find some esoteric hosting providers that are willing to look the other way and potentially lose every other customer over it.
3. Make it cheap. Related to previous option, there are few viable hosts so he will have to pay a lot due to having little to no leverage on pricing. It could turn into a money pit very quickly without a rich corporate benefactor (that also has the tech) willing to take it on. Even friends like Peter Thiel at Palantir may not be able to do it with the potential risk it would pose to their other contracts.
Unchain yourself from dorseys and zuckerbergs.
Honestly I hate Trump but I think they've gone a bit too far. They should just filter individual tweets if needed (like they have). It comes across as divisive and heavy handed.