Interesting thread. Myself I am thinking much about this question. I already suffered many delayed transactions - as a Bitcoin "veteran" I know what I have to do, but for a newbie it must be hell. So I think adoption could be severely hit by a continuation of the "blocksize wars".
Actually this thought was also on my mind, though in less radical version - maybe Litecoin implementing SW+LN will become a go to currency when it comes to actual buying and selling things, however as Litecoin network is not (at least yet) secure enough the coins won't stay there for long and going back to Bitcoin [...]
If LTC manages to become the main "payment network" thanks to Segwit & LN then also its security will increase drastically as there would be a massive price increase and many miners switching to it (although they then need other hardware, as far as I know).
But then it comes back to what you originally proposed - even though LTC might be less secure in overall, are people going to bother themselves to exchange it twice? Why not just stay on the LTC?
Exactly - they would stay inside LTC and the security problem would go away.
I never really fully understood why there was so much resistance to 2MB block or 8 MB block or other blocks (the technical constrain I don't really buy - storage as well as bandwidth is cheap and if it goes faster than Moore's law then the storage technology will followup),
The problem is bandwidth, not storage. Synchronizing a 1 TB blockchain would be only possible in a flawless way for datacenter users - home users and probably also many users living behind China's "Great Firewall" would get serious difficulties to maintain full nodes.
Do the miners actually understand that if some 2nd or 3rd most popular crypto pickups this technology then BTC will put itself at serious disadvantage? Only then we can expect overnight switch in sentiment?
The problem is that some miners fear that Segwit would be the "final" blocksize increase on the Core roadmap and when the new 1.7 MB in capacity is filled (could happen in 1-2 years if transaction volume continues to grow like until now), we'll have worse bottlenecks than now. And some of them think Lightning would not be able to replace on-chain transactions, like seen in this
interview with a ViaBTC spokesperson.
In part, I agree with this view. I would like to use on-chain transactions for middle-to large buys and sells, only for real microtransactions (like buying a coffee) LN seems acceptable to me.
I have returned to be a bit more optimistic because of
this article by Rick Falkvinge about the last Satoshi Roundtable. And my view of the most promising outcome would be a multi-blockchain world with BTC in a leading role. But for this to happen, there _must_ be a block size increase, at least a continous 10-20% increase per year, as well as off-chain methods like LN for microtransactions. If this doesn't happen, forget about BTC's leading position in some years.