With the benefit of hindsight, perhaps what should have happened is that the original 5 developers should have each maintained their own compatible fork of the codebase, cross-sharing patches they agreed on.
That's basically what the Bitcoin XT fork mentioned in the article is. The developers who thought it was a good idea to increase the block size limit increased it and released the client.
When enough people are using it it will end up being the official block chain and the versions that don't support these larger blocks will filter away as they become useless.
There are two problems with this approach. First, as mentioned in the article, people applied outside pressure such as DDOSes to force people to not use the alternate software.
Second is simple confidence. While the fork is an unsettled question you can end up with two diverging blockchains. This makes it uncertain whether someone has successfully made a payment or not, and introduces the possibility of double spending. All that uncertainty is likely to hurt the adoption of bitcoin, possibly causing irreparable damage to its reputation.
