What about a whitelist for specific accounts? Opt-in lists are flexible but it would be a big effort to maintain reasonable lists across the 1000's of blogs. And frontends still have the challenge of selecting which one(s) should be default -- since the default view (say, for a non-logged-in user) should not highlight toxic comments to a reasonable extent.
In addition to a long tail, it seems like some users have a tendency to harass specific authors. It may only be an issue on certain blogs and not globally. Giving authors some power to moderate their blog greatly distributes the responsibility.
Yes this is much better
yeah
Would seem to be addressed by:
Given that consensus downvoted comments are hidden or at least end up at the bottom of the stream.
Downvoting to hide comments only works if you are a relatively high SP user. Also, reward disagreement and content visibility are separate concerns which are bundled in the current system.
It works even if you have low SP, if others agree with you and concur with your downvote by also downvoting. We see that in practice quite a bit when it is clear that content is abusive. Maybe we'll see even more of it after HF21.
I believe you are sightly misunderstanding my point re. reward disagreement and downvotes. I'm not suggesting that the issue here is reward disagreement on the comment itself, but on the post, and the comments are a way to discuss it. If the poster has unilateral control (including ability to suppress opposing views) over the comments, then it obstructs the ability of everyone to assess reward-worthiness. This is precisely the issue with @cheetah and @steemcleaners but I don't think it stops there; any critical opinions are potentially important and allowing the poster to unilaterally hide them is dangerous.
The suggestion to downvote the comments is purely for visibility, but done in a way that requires stakeholder consensus for the comment to be hidden and stay hidden. IMO this is the most sensible approach when comments, particularly critical ones, may weigh on stakeholders' assessment of the post.
I suppose it is true that the two concerns are currently bundled, and they could conceivably be unbundled (allowing stakeholders to vote separately on visibility and rewards) but that would certainly make things a lot more complicated and probably confusing. As things stand now when you downvote content that isn't being rewarded, or minimally so, that is implicitly for visibility, and upvotes on currently-hidden comment is also mostly for visibility (to vote in favor of restoring it).
If it's plain to see who was muted, and if "Reveal all" was a sticky setting, isn't this a trivial change for anyone concerned about hidden opinions?
There is a difference between one person being concerned about hidden opinions and the systemic effect of posters being able to unilaterally hide opinions from most of the readers. Defaults are powerful and there is a good chance that if the default is not showing, the most people won't ever see. That has a systemic effect on the nature of consensus, and not just an effect on individuals.
That being said, of course, these options are on a continuum. It is clearly better to have an easy place to click to see hidden comments than to have to go and dig through a chain explorer, try to find criticism in different posts altogether, or use some alternate interface. I'm still uncomfortable with the unilateral nature of it. A blog that participates in a shared reward pool is not the same as a personally-owned blog that is standalone, and I'm not sure that the same model of unilateral control by the blogger over conversations should carry over.
BTW since you brought up the issue of downvoting for reward and downvoting for visiblity being bundled, why are you proposing to bundle muting? I don't think it really makes a lot of sense. What I don't want to see is distinct from what I might want to hide from readers of my blog. Bundling them would seem to degrade the value of both. I might decline to mute someone I would prefer to mute because I have no intent to hide their comments from others, or I might decline to mute for hiding because I personally want to see it even if I think it detracts from my blog if displayed there. These are really quite distinct.
I've taken a look at how other platforms handle muting and blocks/mutes are generally just applied to your own posts (well, also prevents mentions/DMs). I can block someone on Twitter and they can't respond to me but I can still see their comments elsewhere (as well as view their feed). This also happens to be more efficient for the backend. For the sake of simplicity and efficiency, my preference would be for a user's mutes to be applied only to the discussion threads a user starts, and no longer across the site. This gives the author more control over what kind of discussion they want to highlight, at the cost of having to see comments by the same user on other parts of the site. IMO it would be a net gain.
As for using comments to discuss reward disagreement, it's convenient but not a guaranteed ability. There's a
disable_commentsoption incomment_optionswhich doesn't allow anyone to comment. I don't agree this was a good idea, but it's one way ability to comment on a specific thread is already not a guarantee.It's hard to have any control over basic ordering in your own discussions as a normal-staked user. You can try to flag junk but there are a lot of users who can override your preferences by force, and this is often done not to bring light to valid criticisms, but rather to "mess" with the author.
On this continuum is also the option to whitelist certain accounts, like community-supported bots fighting plagiarism. Listing the muted comment authors in order of highest SP to lowest would be an interesting way to display the most interesting mutes first. Users would see patterns and if they were engaged in the site long enough I'm sure they would explore the muted side of discussions.
A large proportion of users come to the site just one time. Regardless, the impression anyone gets from the discussion may determine if they see any value in engaging further. An unordered mix of spam and harassment among the discussion may likely increase the echo chamber effect. It's just not welcoming.
Twitter has both mute and block, they are separate functions. As well as 'report' which really should drive home the inherent differences in a centralized vs. decentralized system.
As far as
disable_comments, I'd be in favor of removing that altogether (the easiest approach, and seems quite reasonable to me since no one uses it) or disallowing it unless rewards are disabled.Again, you seem stuck on the idea of unilateral control (over ordering here), vs. voting by the community, of which you as the author are still only a part, even if you did create the post at the top of the tree. My answer to that would be if you want unilateral control, don't do it on a blockchain with a global community-based rewarding scheme.
Of course, steemit.com is a centralized web site and can do whatever it wants. I hope this doesn't get implemented in the form described on the centralized web site which happens to be the most used window to the Steem blockchain though.