Hmm ... let's write some lines of code to restrict transaction size to let's say 1 MB instead of writing 500 lines of code for Segregated Witness that quadripples the complexicity of bitcoin and brings less capacity to a later point in time?
Stop being hyperbolic.
Are you referring to unbounded hashed bytes for sigops? Why not bound them? And are all these transactions isStandard()?
Example transaction.What kind of transaction could take that long to validate?
a large transaction with a lot of SIGHASH_ALL signatures, basically
The problem is that the algorithm used for SIGHASH_ALL is O( n2 ), and requires that you hash 1.2 GB of data for a 1 MB transaction.
Maybe you missed that highlighted bit. You are giving me a list of those in involved in creating the approach, and attempt to pass it off as a list of those who support it.
So let's take the development away from a group with several years of experience and assign it to people with much less experience or none? In what world does that make sense?