RE: Hardfork 20: What to Expect Tomorrow
The level of uncertainty expressed in this post is worrisome, taking into account previous hard forks and large-scale user behavioral changes in the past - specifically following last year's hard forks.
I think that the RC system upcoming as described is a set of interesting choices. Not good choices, but interesting choices.
I have commonly been saying "you get what you reward" when it comes to the mechanical underpinnings of the steem blockchain because it's absolutely true. What you reward is what you get more of. Contrariwise, what you punish you get less of.
Given that straightforward pair of axioms, and ignoring what Steemit Inc. is saying and looking at what they're doing, what do they want to incentivize and get more of?
Votes, which are effective leveraging of SP to change the allocation of the reward pool – not surprisingly scaled to the SP of the voter. Token transactions, which are simple financial exchanges and don't represent the addition of any content to the social network side of the service (which is also true of votes, in a real sense).
What are they punishing? Comments, interestingly enough, which represent one of the two ways with which agents on the blockchain can actually create and engage with content and creators. Given that were talking about RC's as proxies and replacements for bandwidth, if the cost of commenting is going up then likewise the cost of posting has to be going up, and I think it's interesting that wasn't mentioned. After all, for all that comments are the vast majority of the time purely text blobs, but tend to a much smaller size, posts have to be costed at a much higher rate even than comments because from the long-term view, far more people are going to see them than any given comment and we've already been told that things with long-term payload transmission costs should be costed higher.
So what do they want more of? Things that don't actually interact with the social side of the steem blockchain. Token transfers. Votes. What do they want less of? Posts and comments, and anything that actually has a long term side effect for storage or presentation.
What happens when you provide downward pressure on the number of posts and upward pressure on the number of votes? Fewer posts get more of the larger number of votes, pushing the distribution of the reward pool to a smaller number of people.
This isn't hard thinking, because it's basic game logic. This is me giving it five minutes of thought. It's not a surprise.
Also, the stated assumption is that witnesses want to increase the amount of traffic on the steem blockchain along with its value. That's patently not true. In fact, if we look at the cost/benefits for witnesses, because they are essentially just database nodes, what they really want is for the amount of traffic to decrease while the value of STEEM increases, reducing the amount that they have to pay in order to remain witnesses in terms of hardware, bandwidth, and maintenance costs to earn the same or more amount of STEEM.
Again, five minutes of thinking. This is not hard. The pressure on all participants is to do less work for more pay; that's just good sense. That's what the witnesses are incentivized to do.
Now, I don't know if this particular post is a product of PR spin first by someone who truly doesn't know any better or someone who is just lying to our faces. Not only do I not know, there is no way for me to know so I can't know.
But what I do know is that there are some obvious impacts which will, if these changes are of the magnitude it's implied they are, provide the pressures as I've just laid them out. With those pressures manifest, it's not hard to predict what the results will be.
It was never hard.
I also find it very interesting to note that despite having access to all of the data on the whole blockchain for the entire history of its existence, nobody sat down during the planning for hardfork 20 and wrote up "this is how the average user experience will change given these changes," and then put that bit of text into this posting. They literally know every activity of every account that ever happened and know what an average account set of operations are, but can't tell us whether or not the average user engaging with the blockchain as they do right now will have to change their behavior going forward, despite knowing what they are costing every transaction except for account creation at in RC's.
That's a real problem for me. That should have been one of the first things that they looked at when thinking about how want to change the resource allocation mechanics. Will the average user have to change their behavior in the new regime? If so, how? These are core and essential questions without some concept of which you can't make the decisions that we've been told are made.
Somebodies full of crap about something. Today, it's not me.
If I’m not misunderstanding the implications of the change, they’ve indeed shot themselves in the foot if they’ve decided to charge to the poster all future bandwidth and storage costs for every reader they project to read a post. There are other choices they could make. It’s inane to make the poster responsible for the costs due to readers without providing the poster a means to monetize their content. That’s the job of the blogging (or more generally social media) aggregation platform to holistically monetize publishing.
I’m roughly calculating the current design is awarding 1 comment per day for every 20 SP (40,000 Vests). At 1% APR (compounded daily) opportunity cost that is $0.00054 fee per comment (opportunity cost charged even if not posted).
Additionally it’s well known that micropayments don’t work when the cognitive load involves the Steemians analyzing the cost of every action they do.
Also the way this HF was handled is an epic PR fuck-up. Displaying a “you need to buy more Voting Power in order to continue posting” to 99% of users is a great way to give the impression of being a scam and/or totally incompetent.
Pack your bags folks, Steem is likely going away soon into the dustbin where lies other fuck-ups such Murdoch’s destruction of MySpace and Marissa Mayer’s aimless shuffling of the deck chairs at Yahoo… the writing of (not just mediocrity but self-immolating) incompetence may be on the wall although I will hold out final judgement in hopes I might be mistaken.
Voting power percentages seem to have recovered about halfway for me, so those are coming back up – but I still have absolutely no sense for how many Comments I could post and still have RC to use on the blockchain. In this is a real problem.
If your numbers are right, and I'm holding about 600 SP, that means that I could write 30 comments a day before I ran out of RC just through the process of commenting. Not posting, not doing any other interaction on the blockchain, just commenting. While I'm not doing anything near that kind of density right now, there have definitely been moments in the past where I have made 30 comments in a day on an active, back and forth conversation.
At least from the discussion regarding Comments (and the active, aggressive avoidance of discussing any RC costs for Posting, which should follow the same line of reasoning), it would appear that they are attempting to adjust those costs to cover future bandwidth and storage costs for activities – which is an interesting idea, because a cynical game designer would point out that you should charge people with more SP a higher RC value because those people are more likely to have posts and comments which drive more aggressive traffic and thus rack up more network and storage costs in the long run. I mean, if we are calculating future costs based on activity, then it would seem that those which are most likely to cause activity bear the brunt of that cost, right?
(In actual practice this would be a truly suicidal design since it deliberately punishes those who are successful, but since so many members of the steem blockchain seem to be into that sort of thing, I would be tempted to give it to them, and give it to them good and hard.)
Though I think you might be missing something by not noting that posters do have the ability to monetize their content – and, in fact, that's pretty much the sole and solid underpinning of the steem blockchain living underneath this relatively ad hoc social network living on top of it. The monetization architecture is the only thing that's people seem to talk about. Technically, even we are talking about it right now.
Now, the fact that votes within the 15 minute window burn their portion that would otherwise have gone to the author during the previous hardfork, that's ugly. At least if voters started hitting a new piece of content within the open window in order to jockey against one another for the higher curation awards, that gave the original author a reward for being the kind of content they think they can turn a curation dime on. Now? Those funds get burned, which in theory increases the rewards for everyone – but in practice, since rewards are scaled by SP, it is literally taking money out of the hands of the creators and putting it into the hands of high SP curators, and it falls hardest on those who actually want/need to take their money out of the blockchain and spend it on goods and services.
Additionally, I'm not down with the assessment that "micro-payments don't work", because we have some pretty good examples of micro-payment systems that do work – they just have real trouble getting traction because for one to be particularly useful, they have to see widespread adoption and the way of managing the funds needs to be relatively transparent. Personally, I'm really fond of Flattr because it only asks how much money I want to put in my tip out jar once a month, then counts all the tips that I give during that month by number of times I've interacted with pieces of media put up by people who use Flattr, and splits up the amount I have already decided to distribute across all of the tip shares I've handed out, restocks the jar, and the next cycle begins.
If anything, I wish the voting system on the steem blockchain worked as transparently. Rather than worrying about voting windows and voting order, if accounts simply had an amount of STU determined by their SP for the week, they go out and vote as they see fit as much as they like or as little, on content dropped immediately or a year old, and at the end of the week their STU is divided between the shares they've handed out – it would be a much better, more effective, more useful system.
Which is why I never expect it to happen.
I don't expect that steem is likely to go away soon simply because it does have an active (if ridiculously messy and largely content-corrupt) social network attached to it. It is, in some sense, doing actual work, which puts it somewhat ahead of quite a lot of other cryptocoins, but it's certainly possible that at some point we may turn around and look at HF 20 as "the beginning of the end" if some of these other issues aren't tackled.
The nature of a blockchain is inherently problematic for storing content like that of a social network. You need the whole thing any time you want to add a new node, it never gets any lighter, there is no way to create adjuncts which easily create a proper database for using a digital application making it much harder to do anything serious – really it's just a mess, and the wrong tool for the job.
But it's the one that's active right now. We'll see if it makes a difference.
All I can say is when CRED?