[Bitcoin-development] Proposed alternatives to the 20MB step function

Mike Hearn mike at plan99.net
Thu May 28 17:34:42 UTC 2015

> Twenty is scary.

To whom? The only justification for the max size is DoS attacks, right?
Back when Bitcoin had an average block size of 10kb, the max block size was
100x the average. Things worked fine, nobody was scared.

The max block size is really a limit set by hardware capability, which is
something that's difficult to measure in software. I think I preferred your
original formula that guesstimated based on previous trends to one that
just tries to follow some average.

As noted, many miners just accept the defaults. With your proposed change
their target would effectively *drop* from 1mb to 800kb today, which seems
crazy. That's the exact opposite of what is needed right now.

I am very skeptical about this idea.

> I don't think us developers should be deciding things like whether or not
> fees are too high, too low,

Miners can already attempt to apply fee pressure by just not mining
transactions that they feel don't pay enough. Some sort of auto-cartel that
attempts to restrict supply based on everyone looking at everyone else
feels overly complex and prone to strange situations: it looks a lot like
some kind of Mexican standoff to me.

Additionally, the justification for the block size limit was DoS by someone
mining "troll blocks". It was never meant to be about fee pressure.
Resource management inside Bitcoin Core is certainly something to be
handled by developers.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150528/3359ad4a/attachment.html>

More information about the bitcoin-dev mailing list