Quote from: countryfree on September 20, 2016, 07:53:25 PM

I don't like topics about block size. It's smarter to talk about free space within each block. On an average computer, most people would say you need to keep at least 15% free space on your hard disk so that you work on files without hiccup. It's the same thing with BTC. If we're close to max capacity, things will be less smooth. It may even stall at some point and we don't want to see that.

In most blocks, less than 0.2% free space is left. I completely agree, and I've been saying it for a while: this limits Bitcoin growth. It also limits how many people can use Bitcoin, as any new user takes away a transaction that an existing user can no longer make. Lightning Networks may work in the future, but we need a solution now. And with "now" I mean about a year ago.

Quote from: zimmah on September 20, 2016, 08:58:17 PM

And people worrying about bitcoin getting more centralized if we get bigger blocks are not considering the fact that we have been sitting on 1MB for years now, and since that time internet has gotten cheaper and disk space has gotten cheaper as well. So we can easily increase the block size and we can do so again when we need to (and by that time, internet and disk space have become cheaper again).

Miners spend millions on hashrates, and barely anything on disk space. By changing that ratio a small fraction, all miners can afford enough disk space for much bigger blocks.
The problem is that miners now only have a financial short term incentive to increase their hash rates, while they have no direct gain from larger blocks.

Quote from: illyiller on September 20, 2016, 09:01:04 PM

If we perpetually increase capacity ahead of demand, fees will never rise. Transactions will always be free or nearly free for users. This will not end well in a future where block subsidy ends and fees alone must support the security of the network (by incentivizing miners). If we did this, we are basically depending on mass adoption and skyrocketing price being guaranteed. That's probably not a good engineering decision.

The demand is there already.
By not increasing capacity, mass adoption becomes impossible. It's already not possible for just 1 million people to make a few transactions per day, let alone much more people.