Thank you for your hard work.

I'm a little disappointed in myself that I didn't manage to get to reading this until now.

I'm sure that a lot of witnesses did what they could to try to ensure that the hard fork was a success. Unfortunately, it was not, no matter what Steemit's blog says. Now, after a few patches, it's not bad, and the changes can even be said to be good, perhaps with a few further tweaks.

I personally wonder if perhaps part of the issue is communication. From what I've heard, there may have been some signs that there might have been issues on the testnet...if perhaps these issues were better communicated to witnesses and others that could examine the code and determine if there was indeed an issue, then perhaps HF20 would have gone out without a hitch, being patched before release.

Breaking things into chunks that are more easily digestible will help alleviate that without rolling over witnesses so heavily into feeling they must go with the flow or be the one to hold up needed development.

Breaking up the code into much more manageable chunks likely would go a long way into preventing similar problems such as this in the future. I've heard other witnesses say as much as well. Hopefully after this disaster, Steemit will consider that perhaps their opinion has merit.

Something like HF20 would be an absolute disaster to an even larger degree with a larger user base. The number of users that are inactive likely saved a lot of hassle.

It's not that we shouldn't have large goals for improvements, but perhaps, as you said, certain features should float between hard forks.

This isn't even how coding works. Certain features are ready really early, and possibly can be tested and even pushed before the main fork. Certain other things need to be tested more, and perhaps even delayed. In applications development, a planned release might actually end up split up between multiple versions, with additional changes added along the way. Features that are ready early are occasionally back ported into the next small point release, rather than the major release, and then the major release might even eventually be delayed.

Basically, we should move to something more akin to a rolling release style, rather than all-encompassing hard forks.

In addition to community outreach and other forms of spreading our intentions, to make a post (or series of posts) that states clearly our agreement with or disagreement with the Hardfork, specific features, and our adoption intent.

I wish more witnesses did this. Maybe I just completely missed all the discussion on the blockchain about the hardfork...other than posts lauding it as so incredible it will toast bread as it slices it, and inject cheese at the same time, so every slice will be a cheese toasty without ever having to turn on the toaster. Maybe some people can't handle hearing about potential issues, but I want to see posts about it. Open source needs honesty.

I'm still really annoyed that Steemit seemed to just completely disregard reality with their announcement of success, even as almost the entire user base could do practically nothing.

To be part of code auditing, as well as following github changes and updating our communications and testing as required based on that.

HF20 kinda made me feel like perhaps there aren't enough witnesses in the top 20 that are capable of reviewing the proposed code changes. I kind of feel like it's partially the fault of certain witnesses who are more qualified in coding for not pushing this as one of their qualifications. More users are seriously aware of how huge of an issue it is that not enough people are reviewing the code put forth by Steemit. Unfortunately, they can't even do anything about this currently, as witnesses are primarily chosen by the biggest of whales. The only possible solution I could see to this is to make a tool that allows users to band together, so their total votes are equal to the votes of whales.

To encourage and seek out better communication, earlier, more frequently, and more clearly, between all witnesses and Steemit Inc.

Yes. I've already stated that I wonder if this perhaps could have been a major issue as to why there was the failures of HF20. Those who knew there may have been issues did not communicate with others, or did not communicate with the right people so people knew what major issues there were. Communication is a huge thing. As is documentation, which is a form of communication. Steemit needs to communicate with witnesses and users better, and witnesses need to communicate with users better, and people in general need to make users more aware of what they think about witnesses and who they possibly might consider voting for.

Individual users might not have much sway when voting, but together we can make a difference possibly.

Thank you for your promise to increase communication.

I'm unsure why witnesses have previously been so bad at communication. I would think that it would only help them to get more witness votes. Whatever the reason, it does need to change.

I thank you for your effort in trying to be more active in communication, and hope you are successful.