> It ISN'T low right now...

I agree, but I don't think it's a good idea to softfork it to lower than 4M WU though, and I don't think we need to;
hopefully when exchanges start using Lightning or Liquid, avg blocksize will go down.

> Extension blocks are not softforks, and are unreasonably convoluted for no
real gain. When the time comes, the block size should be increased only using
a hardfork.

It depends on how you define soft and hardforks, I suspect you don't see extension blocks as a softforks because old nodes won't maintain a correct UTXO set.
I think an extension block is a softfork because old nodes will still be able to follow the mainchain.

I don't know if a blocksize increase hardfork will get consensus as the idea has been ruined by all malicious hijack attempts we've seen over the years.

Hampus

Den mån 11 nov. 2019 kl 17:47 skrev Luke Dashjr <luke@dashjr.org>:
On Monday 11 November 2019 16:08:43 Hampus Sjöberg via bitcoin-dev wrote:
> I am advocating to keep the blocksize low right now,

It ISN'T low right now...

> but I don't leave out
> in increasing it in the future when we have a need for it, preferably via
> an extension block (softfork).

Extension blocks are not softforks, and are unreasonably convoluted for no
real gain. When the time comes, the block size should be increased only using
a hardfork.

Luke