The last difficulty change took about 20% longer than expected.  How large does the time between difficulty changes have to get for us to make changes?  In other words, if, at some point, block confirmation times are averaging, say, hours or days, will we hardfork to speed things up?

One option is NO.  When enough economic interests align to amass the computing power to get important bitcoin transactions into a block, then they will work out a way to get that block confirmed.  This allows other cryptocurrencies and technologies like LN to fill in.

There may be a group that will fork the code in order to adjust the difficulty more rapidly, and bitcoin holders will put a value on bitcoin-FDA ("Faster-Difficuly-Adjustment"), which is fine with me.  We can learn how to fork peacefully from what we learned when BCH was born, and what we learned when it split.

I think some insight into how core developers will handle increasing demands to use faster difficulty adjustments (if they respond at all) will be helpful, and this is why I'm asking.

Dave Scotese