What are the origins of segwit? Was it long proposed as a nice idea by multiple people or did it arrive out of nowhere from a single source and tickle everyone pink?
SegWit was presented by Blockstream's Pieter Wuille to the world at the end of the second Bitcoin Stalling conference in Hong Kong. The video of his talk should be on YouTube. Apparently it was a surprise to most people there, except Blockstream folks of course. Indeed I would say that Blockstream planned the conferences to be just a stage for the SegWit announcement.
According to Pieter himself, he thought of SegWit some time ago, but put it aside because he believed that it would require a hard fork. But then Luke Dash Jr. found a way to make SegWit into a soft-fork type of change, by using a script hack and redefining one of the NOP opcodes. That made it possible to deploy it in Blockstream's favorite "stealth mode". (That is, the change is effective as soon as a miner majority adopts it, whether full nodes, users, and businesses like it or not.)
AFAIK, the only significant improvement that SegWit brings is to fix various malleability problems in one go. Even that benefit could be obtained much more cleanly by other means, without changing the block and transaction format; but this cleaner solution would require a hard fork, and also the discarding of Pieter and Luke's ingenious hack -- so obviously it could not happen.
I haven't heard of the "fraud proofs" in a while. In the initial description, they seemed to be more "hints" than "proofs"; and it was never clear how they would be used, and for what.
One interesting "benefit" of SegWit was to make people aware that soft forks are actually more dangerous than hard forks. With SegWit's "extension record" trick, a soft fork can achieve many of the taboo changes that were thought to require a hard fork; such as increasing the block reward (and therefore the 21 million issuance cap) or confiscating coins. As soft forks, those changes would require only the agreement of a mining majority, without the consent of the rest of the community. However, for that same reason, there is nothing that the community or the developers can do to prevent non-consensual soft forks.
I find your above post a bit confusing.
Are you talking theory or is there anyone of import (besides you and some members of the bitcoin community who actually bought bitcoins) who is actually opposed to seg wit?
My understanding is that even "developers" Gavin and Jeff Garzik are in favor of seg wit, and Gavin and Jeff Garzik are the only two "developers" who had been proposing a need to hardfork (through XT and Classic).
So, what's the deal, is there someone else who is notable who is opposed to Segwit?
In other words, if there is no meaningful opposition (besides some whining voices of non-users or even users), then it seems that a softfork would be preferable to a hardfork, no?
In other words, are you whining for a hardfork when a hardfork is not actually necessary? how could a softfork be more dangerous in the deployment of seg wit and in such actual real world circumstances, if there is actually no opposition seg wit?