I think it will be helpful to clarify where Krugman is right about bitcoin and one subtle but very important point he and his uber-nerd friend might be missing.
(1) First, let's accept that Krugman forgets more about economics while having his morning coffee than most of the rest of us will ever know in our whole lives. He fully deserves to be taken seriously on economic issues.
(2) I think most of his reaction to bitcoin is based on the fact that there is a vocal subset of the bitcoin community that says things like: (a) bitcoin will replace fiat currency or (b) bitcoin is better than the dollar because there is no 'money printing' or (c) bitcoin will replace the dollar some day.
This is all patently nonsense. Krugman is right to slap it down and he will be particularly sensitive to it because he has spent his whole life fighting austerians/gold standardians and so on.
So far, so good.
(3) In the last 24 hours however, Krugman has:
(a) Accepted that bitcoin is a technological breakthrough but also made a price / valuation call (aka it is a 'bubble' aka it is overvalued). My guess is that he would have made the same valuation call if / when bitcoin was valued at $10 (aka, it has no intrinsic value, it can't be a store of value and the price will fall) so it is not like he is saying it should be $100/coin vs $700/coin. He is effectively saying that it is intrinsically worth very little.
(b) He then got a lot of flack for his 1998 prediction that the internet would have no more impact than the fax machine, something that he is downplaying today with the fair justification that he is not a technologist, he is an economist.
So, in aggregate, he is saying:
(c) Bitcoin is, in fact, a nifty technology
(d) But it is overvalued (in a bubble)
Unfortunately, there is no way you can make the case that it is overvalued or undervalued or correctly valued without having an opinion about what its ultimate adoption rate would be.
This is the part he is missing: Since there is no way to use bitcoin-the-technology without buying bitcoins-the-coins, so long as there is demand for bitcoin-the-technology, people who want to use it will need to buy bitcoins from people who already have bitcoins and those people will demand some non-zero price for them.
This is a bit counter-intuitive, but it is counter-intuitive in more or less the exact same way that it was counter-intuitive that the internet network would have a lot of value in 1998, which is why it keeps getting brought up as a possible 'blind spot' in his thinking. Even hyper-smart, Nobel prize winners can't be expected to keep the whole set of human knowledge in their brains at once!
In summary, you can't figure out what a bitcoin is worth without knowing (a) the future adoption rate and demand for bitcoins, (b) discounted to the present. Given this is an almost impossible thing to predict, this is not much different than Krugman calling Instagram overvalued at a $1B valuation. And if he is making those types of calls, I am not sure how much value his macro expertise adds - he is just another technology prognosticator at that point.
(4) So, to conclude:
(a) Bitcoin would be a terrible national currency. Krugman is correct.
(b) Bitcoin has not invented anything new in monetary economics. Krugman is correct.
(c) Bitcoin is a significant technical innovation. Krugman is correct.
(d) Bitcoin will have a non-zero value because of the value of its network and Krugman seems to be missing this part altogether. If he does accept this and is simply making a valuation call on bitcoin, well, he is a better man than me because I don't know how one could do that at this stage.