Weren't you supposed to be some sort of a comp-sci professor? I don't know how to be kind about it, but your students are getting screwed...
Indeed. Students who are used to uncritically memorize and repeat what the textbooks say often get screwed in my courses.
Note that to a satoshi-based cyrpto-currency solution the proper terminology is 'longest valid chain'. Bloat-blocks or any other illegality is simply ignored.
No. Chains that cointain illegal blocks are ignored
only by software that considers such blocks illegal. But, for players that are running software that accepts big blocks, big blocks are legal. Is it clear now?
Core would keep right on chunking away even if it got 1% of the hashing (excepting problems with difficulties and absent attacks.)
No, it is not "Core" that will do that, but "any miner who is running software that does not accept big blocks". That software can be the current Core version, or XT minus the 8 MB patch, or any other independent code with a 1 MB block size limit. Miners that accept big blocks may be running the current XT (with or without the other patches), or the current Core with an 8 MB patch (by repentant Blockstream devs, or by someone else), or any other version that accepts 8 MB blocks.
That is the main point: the "Core vs XT" and "Blockstream vs Hearn&Gavin" issues are red herrings. The only thing that really matters is
how many miners will accept big blocks at some point in the future.
Ideally, whenever a big block is actually solved and posted, either almost all miners (counted by hashpower) should be accepting big blocks, or almost all miners should be rejecting them.
If, at some point, 75% of the miners suddenly start accepting big blocks, the other 25% had better start accepting them too. Ditto if, after every miner is accepting big blocks, 75% suddenly decide to start rejecting them.
If the miners are mostly in agreement about accepting or rejecting big blocks, but a majority changes their mind, in either direction, they had better all change together, before someone posts another big block.
As long as all miners are mostly in agreement, it is safer for all clients to accept big blocks than to reject them.
If the miners ever get into a messy state, with significant percentages of both big-block acceptors and big-block rejectors, all clients had better stop using bitcoin until the mess gets cleared up, and almost all miners have again converged to the same policy.
the mining effort will be primarily driven by the market value of the product.
The preferences of the "economic majority" will surely influence the miners' decisions. But, once a significant majority of the miners has decided that they will accept big blocks, after the first big block gets mined the "economic majority" had better be already accepting them too, or start accepting them right away. While the miners can wait a few weeks before selling their mined coins, the "economic majority" cannot stop using bitcoin for a week, in the hope of winning the strong-arm duel.
Moreover, the "economic majority" does not have any good reason to reject big blocks, and some very good economic reasons to accept them.