In the Steem 0.17 Change Proposal Introduction post, @steemitblog proposed several changes for HF 17. It looks like they are nearing their final set of changes, and a tentative date has been set.
Disclaimer: I do not work for Steemit. This is not an official Steemit post. Everything in this post is subject to change. I am just reporting what I see in the Steem GitHub repository.
Hardfork Date
The HF date is tentatively Set to Tue, 7 March 2017 16:00:00 UTC (11:00:00 EST)
Changes
Comment Reward Beneficiaries
As a comment option, authors should be able to specify a number of beneficiaries to receive a percentage of the rewards and the percent of the potential liquid rewards to receive as Steem Power. Unallocated rewards by default go to the author as they currently do. (GitHub Issue 773)
Aggregate Reward Payouts into Reward Balance separate from Checking
All users will have their rewards paid into a reward balance for STEEM, SP, and SBD. These balances can have higher precision than the main checking account to accumulate the rounding errors from payouts. Once per day all rewards greater than a minimum balance can be swept into their checking account through an explicit virtual operation signed by the witness.
This is a step toward keeping reward calculation as a separate / independent thread from the rest of the blockchain. (GitHub Issue 659)
Remove Posting Limit
Currently there is a 4-post per day soft limit on posts, after which author rewards are throttled. This limit will be removed, so that authors can create an unlimited number of posts without being penalized by the blockchain. (Bandwidth limits will still apply.) (GitHub Issue 732)
Remove Bandwidth Rate Limiting from Consensus
Currently there is a consensus based bandwidth limit, which prevents users from using an excessive amount of bandwidth. The amount of bandwidth users are allowed to use is based on the amount of SP they have.
This limit is being removed from consensus, although it can still be enforced outside of consensus. What this means is that it will be up to the witnesses to enforce this. If they decide to allow a transaction through despite a bandwidth violation, it will still be considered a valid transaction by the blockchain. (GitHub Issue 766)
Remove Comment Depth Limit
The limit on comment depth that prevents you from replying after the 6th comment will be removed. (GitHub Issue 767)
Pay Comments Independent of Discussion
Instead of the comment payout being tied to the parent post, each comment will have its own independent payment window which starts as soon as the comment is created. (GitHub Issue 768)
Change Payout Periods
The 30 day payout period will be removed. The 24 hour variable payout period will be changed to a fixed single week payout. The payout will always be 7 days from when it's posted, and will not be extended. (GitHub Issue 769)
Disable Proof of Work
Proof of work (mining) will be removed. In place of mining, a 20th "top witness" position will be added. (GitHub Issue 770)
Permanent Editing
Comments and posts can be edited forever. It looks like as an added security measure, users must allow this first using their active authority for a limited period of time in order to make changes past a certain point in time. This is to prevent a computerized account from having all its content deleted/changed. (GitHub Issue 772)
Comment Reward Pool
The current reward pool will be split between posts and comments. Posts will receive 62%, while comments will receive 38%. Voting on comments will not produce curation rewards. Voting on posts still will.
The equation for calculating the rewards for comments will be changed from n^2 to n^2 / (1 + n).
For each user, all voting (whether on comments or posts) will be made out of a single voting power pool, with the same maximum of 40 100% votes per day (across both posts and comments) without depleting it.
Update Trending to use same algorithm as Hot
The changes being made to the API are as follows.
get_discussion_by_trending will use the same algorithm as hot, but with parameters tuned for a single day turnover of posts rather than the one hour used by hot.
get_discussion_by_payout now actually returns posts by payout. This sorts both posts and comments.
get_discussions_by_trending30 will continue to work until the hardfork when the index becomes obsolete.
These changes are being made in such a way that existing applications should be able to continue using them as they currently are and the new sort orders will be returned automatically. (GitHub Issue 811)
Steem Power Delegation
An account can delegate their Steem Power to another account which will give that account extra bandwidth and extra voting power. When Steem Power is delegated curation rewards do not propagate back to the source, but belong to the account that received delegation.
This feature will enable people with multiple accounts to delegate all of their Steem Power to a single account and reduce unnecessary voting spam. It will also allow users to create accounts on the blockchain without needing to actually fund the account (the account creation fee can be paid by delegated SP instead).
Users with a high SP account balance have the ability to give other users with less stake a "boost" by delegating them additional voting power. (GitHub Issue 818)
Allow Replies on Paid Comments
Because comments can be edited indefinitely, discussions will be able to continue indefinitely. (GitHub Issue 869)
Other Misc Changes
- Add order ops to account transfer_history (GitHub Issue 157)
- steemd should not be an ntp client (GitHub Issue 727)
- Equalize Per Block rshare Value (GitHub Issue 771)
- Refactor reward computations out of database (GitHub Issue 782)
- Missing ampersand in price constructor parameter (GitHub Issue 789)
- Refactor adjust_rshares2 to be non-recursive (GitHub Issue 803 )
- Remove discussion rewards code (GitHub Issue 805)
- Remove unused code in steem_objects.cpp (GitHub Issue 837)
Image CC0 public domain from pixabay.com.
Remember to vote for witnesses!
If you aren't sure who to vote for, check out this Witness Voting Guide.
https://steemit.com/~witnesses
Excellent post on what is going on and I learned a lot on the comments given. I hope steemit doesn't forget about the minnows. It will be interesting to see how this all works out. Thank you to all the steemians who worked so hard cranking out these changes.
Welcome! Glad you found it useful :)
Big thumbs down on making folks wait a week to get paid.
If we get a better rewards curve it will make it some better, but who wants to wait a week?
What happens to the 30 day payout?
We will have to wait and see how the 7 day payout plays out.
The reward pool was currently being split between the 30 day and 24 hour payout periods. With only one payout period now, it will no longer be split.
Yeah, I'm not sold on the 7 day, but it looks like we will be giving it a try.
As long as the minnowmath gets changed to something more realistic, I'm pretty good.
Any chance that we simply raise the bottom to something relevant while leaving the exorbitant rewards for being a whale in place?
I like the format better for it's ability to aggregate content and facilitate comments.
Any word on when we will get notifications that somebody else has commented on a post I have commented on, but not necessarily in reply to me?
You know, like a forum?Maybe a button to access @jesta's db forum would be nice.
It sounds like after this HF if the
n^2 / (1 + n)formula works well for comments, they are planning to look into changing it for posts as well. It also sounds like they will be reevaluating the formula for curation rewards as well.I don't know the plans for integrating third party apps into the Steemit.com website. For now, users will have to find it through word of mouth / the "Steemit App Center".
No word yet.
Ok, so rewards for posting are gonna stay the same?
The minnows are irrelevant in the math, currently.
Why would anyone stay when they have to catch the eye of the sycophants of those at the top, who have clearly stated that no controversial personalities need apply?
I guess if that is not fixed, it can only mean that the minnows continue to not matter to those at the top.
Hardly a way to win over folks that will start at the bottom, for the most part.
I'm with you. We need to find a solution to that. Minnows will likely have more voting power over comments if the changes go through.
So, the minnow math isn't becoming relevant after this hf?
I don't fully understand the question. The only change to the minnow math that I am aware of is the change in rewards formula for comments.
Gonna have to do something, the bottom is falling out.
Welcome whether this helps the platform be better
Excellent post congratulationsThank you very much sir @timcliff for the valuable information, there are very important points
My opinions.
Payouts:
I think the 7-day payout is a good idea BUT I think the posts should remain open for payouts in perpetuity because Google does pick them up and some evergreen posts may not find traction until 30+ days. That being said, I've yet to see a post make a shit-ton of money beyond the initial 24 hours anyway but blocking potential earnings sucks.
And what's the point of permanent editing if the payout stops after 7 days? Just wondering about the mind-think on that one.
Comment payments:
I don't know that it's necessary to separate these. Sometimes comments are just as informative and thought-out as a post and develops good conversations. Now with expanding threads, this makes it a lot easier. That being said - and I mentioned it on @steemblog's post - I would change comments payouts to reward comments over 100 characters (or 80 characters + pic/gif). Comments under 100 characters would get nothing. "Nice Post" with bot upvotes get no reward. See where I'm coming from? With this in place you don't need to worry about beneficiaries. How's a newbie going to understand all this???
Remove Posting Limit:
Bad idea. Look - if we really want quality then it actually takes TIME to write and format a post. I forecast a shitload of meme's and super short posts clogging the feeds. I've personally removed a lot of people from SteemVoter because they posted more than 3 times a day, which was siphoning my vote power.
Oh, and by the way - Hi Tim! I've missed you. ;)
One of the problems with this, is the power down period needs to be longer than the potential payout window, otherwise users could keep powering down their SP, powering it up into a new account, and revoting the same content over and over to siphon rewards.
The first thing that came to mind when I read your response is: You have to think like a crook to beat the crook. LOL
Doing something like that would never have crossed my mind. I guess I need to hang out with you more my sullied friend. hahaha
LOL!
Hi @merej99 - missed you too! Thanks for sharing your thoughts :) I'll be addressing a lot of these points in my witness post today.
Thanks for sharing :-)
I know people will still have issues with this. A single payout is not ideal, but it doesn't seem practical to keep paying forever
A tip button would fix that problem. Then anyone who finds content at any time can reward the author. No complicated rules, no stupid timers, no voting bot manipulation, just a simple tip for good content.
The way it sits now your posts are locked forever and make you nothing after the voting payout. That's why people stop posting.
Great suggestion! There are plans to add this.
There are?
Sorry if I miscommunicated. Based on the conversation in the GitHub issue (and there was a link to a trello item) I kind of assumed this was a matter of "when" more than a matter of "if". Regardless though, I should not have worded it the way I did. It is not up to me to say what is/isn't planned -I just try to report on what I see. I apologize for that.
https://github.com/steemit/condenser/issues/401
I'd love to see a tip button. Just make it simple to give some extra. It should be simple as it's just a web site feature. I wonder if one of the alternative sites will do it first
Perfect suggestion.
We could have a one week with a 1 month payment... Just abandon 24h
I guess 7 days is enough to get 90%+ of what it's likely to get. There's a long tail, but some posts may get discovered via Google much later. I understand people will at least be able to comment then
@timcliff u do the witnesses update. Can you do an informal poll to see whom is going to go HF17?
I will be doing my witness report on Sunday. It is actually hard to get a response from a lot of the witnesses though. Some probably haven't made up their minds yet either. It will be up to each witness to communicate their position on the HF with the community.
I can't say I am in favor of most of these... I feel like if we fix the voting power issues it may solve some of the other problems with the site... That and we haven't really had a large enough user base to really find out if the system works as is or not.
Isn't the problem currently that steemit is losing members with same speed that it is gaining them? Most of these new account are probably bots, just to keep numbers pretty and the members leaving actual people with bad taste in their mouth. Steemit had its moment in the spotlight and back then you had tons of posts with this message "Steemit is not pyramid scheme" but if it looks like a duck and sounds like a duck, it probably is a duck. So not changing things probably isn't an option if Dan Larimer ever wants his name to be associated with something else than failure, though if just getting money was his goal he did succeed. So he probably doesn't care. Good for him.
This is a view widely shared in cryptospace and when steem price reaches that of bitshares I guess we can all start looking forward for his next project. Can't wait to see that.
Just joking Dan ;)
A lot of these are probably beneficial.
However, I still believe that the most needed updates are ones for the UI. I can't tell you how annoying it is that our posts cannot use CSS. Let us not forget that Steemit is not only a cryptocurrency platform, it is also a blogging platform, and is competing with all other blogging platforms out there (Wordpress, Blogger, etc.). Comparatively, we're still in the stone age. And if we want the same level of adoption as these other blogging platforms, then we need to catch up.
From a development perspective, it makes sense to build the "engine" first and then worry about the UI parts later. If there are plans to continue tweaking the engine, and each time a tweak is made they have to update the UI - it is a lot of double work.
And if Steemit were a car, I would agree with you completely.
But Steemit is actually a piece of software. Now, many software developers deceive themselves into believing that what customers really want is a fine-tuned engine under the hood. It is one of their most common misconceptions. If this were true, we would all still be using Word Perfect as our word processor.
History has proven, again and again, that the only thing users care about is the interface, because it is the only part that the users actually interact with. And Steemit's #1 problem, lightyears ahead of all other problems at this moment, is that it is having problems attracting users.
By insisting that the underlying engine is a more pressing concern, you have fallen into the same trap that so many engineers have succumbed to over the years. Your concern is no doubt well intentioned, but it is divorced from reality.
If we were getting ready to officially launch tomorrow, I would completely agree with you. The site that we have today is not even close to something that is ready for mainstream. We need a "sexy UI" upgrade, with a lot more bells and whistles and improvements on the user experience.
The order that things are done is important though. The dev team has indicated that they have a lot of things that they plan to implement at the blockchain level, and it sounds like many of them are going to have dramatic changes to the user experience (like communities).
I am not arguing in favor of keeping the crappy UI. I am just saying that as a community of "beta testers" (which is what we all are) - that we should accept a limited UI for some time while they work on more of the important back-end features.