This can be handled by third-party front ends by buffering votes until 15 minutes for comments. I know certain front ends already buffer comments for 3 seconds for next block.
Might be done as a browser-plugin replacing the votes button. Should be no big deal. But is it worth? When voting I just have the author in mind who is not affected.
I don't think this is correct. $0.75 still goes to the author regardless of when you cast it, it's just a question of where the curation rewards go.
That is not how I have been seeing it. I just posted this comment and immediately upvoted it 100%. (~$1 vote). I had a post the other day after the fork that had an auto vote, and it showed the full vote value, then not long later the vote changed from ~$10 to $0.04. (Will see how this comment pays out). I can't explain why it shows $10 initially, then an hour later it was $0.04 but that would fit the scenario I explained.
Very possible, might have been something with 0.20.2 (which is what we were on at the time), but I did see it first hand do exactly as I describe, but very strangely it initially showed the full vote value.
That sounds really strange! Could you link the corresponding comments here?
from $10 to $0.04 is a huge step, but certain fluctuations on the pending payout value are certainly expected from the general lack of voting stake in the last couple of days.
When I looked from my main account, probably around 1 hour later, it showed it only had $0.04 and that vote was on the post. I removed the vote and tried again, and it showed properly. But now I am unable to mimic the behavior and think it might have been related to v0.20.2 as it was on the 25th when hard fork just happened to land.
I have a post above I am using a test to see how it pays out. But I may have been wrong, this may have been an issue with an older build and has been corrected.
Argh, due to the unvote/revote the original rshares aren't stored anymore, so it's almost impossible to know what happened there. Let's see how your experiment above goes. From what I understand from the steem sourcecode, I agree with @tcpolymath.
Yes, that's how I understand it as well. It is the curation reward that will turn into dust. The author will now always get 75% of the reward. So early voting will no longer bump the reward to the author. For the curator there is no change compared to HF19, except for the threshold decrease to 15min.
That's how I understood it should be, but I was not seeing that right after the hard fork landed on the 25th. I ended up removing the vote and re-executing it like an hour after the post and it registered as a full vote. I'm starting to suspect something was fixed between v0.20.2 and where we are now.
In fact, now that I think of it, I think the vote was right before HF20 was launched. It was after HF20 when I noticed the vote didn't count but was on the post.
It seems we all complain a lot about hf20 these days.
The only thing, next to showing up all errors and miscalculations in HF20, is to remove our witness votes from the top20 witnesses, who all agreed to it and found nothing wrong with it.
Not one opposed it.
All went immediately to the next version.
It looks as if T20 witnesses are super fast at code review and testing.
But on the other hand..
You already vote for good witnesses and rarely any t20
No need for me to preach here.. Sorry:)
I don't mind. I figured out a while ago that my vote is meaningful to witnesses outside the top 20 - it increases their chances of getting a block - and almost meaningless to those in the top group.
Hi! The behavior you have described does not seem to be intended. Only the curation rewards (25% to 0% scaling linearly for first 15 mins) is returned, so the author should still get 75%.
It says the "early voting penalty" is returned, which is the 25% at 0 mins, so the author should still be getting 75% (at all times instead of the old 100% to 75% scale) and curators have the same scale as before (0% to 25%).
Hi Mark. Waited for the 15 minute mark and it is a little more hands on now. I never really cared about when I voted but making more of an effort to actually think about it now.
Good point. Was wandering why some peoples comments I voted on were different with curation so must remember to wait on comments as well. Makes sense now. Have already voted on a comment straight away recently now will remember to wait. Thanks you have been a big help.
It's really nice explanation there Mark, although I think that the votes (either it is for the comments or the post itself) are not affected at all.
What IS affected though, is the curation rewards for each vote. Meaning that your 1$ vote on a comment will still be 1$ even under the first minute but the curation rewards for that comment if it is voted within the first minute, drastically reduces.
WHAT!? I really don't like the whole voting deal. I love to give my followers large comment votes to encourage more comments and conversation. I understand how this can be helpful in a post scenario. But bad for comments. I wish there was a difference.
Nothing has changed @themarkymark. I have an account who votes for me at zero minutes. He has been doing this since way before HF20, possibly as far back as HF18 he started.
He is actually a friend who doesn't bother with Steemit anymore, and I've told him to change the time of his vote several times. As he get's zero rewards by voting for me on zero minutes.
The only change is who his reward goes to, before it went to me, now it goes back to the pool.
Yup.
I can see myself opening tabs up, writing replies ready to submit at around 15 mins with a vote, and forgetting.
That's for the replies I would consider voting for. The rest will either be ignored or replied to right away.
I agree, As a forgetful person myself. I could see this regularily happening.
Automation could come in handy here. But just sad it has to be that way...
This can be handled by third-party front ends by buffering votes until 15 minutes for comments. I know certain front ends already buffer comments for 3 seconds for next block.
What about notification tools optionally delaying their notifications by 0, 7.5 or 15 minutes?
Or perhaps they could have an option to notify twice. Once for the initial appearance and a second time when votes are optimal.
That's a very good point. I guess we may see front ends offering delayed votes for self voters too.
Thank you for the heads up about the timing. I did that just before reading your post. That's why is good to read first.
Might be done as a browser-plugin replacing the votes button. Should be no big deal. But is it worth? When voting I just have the author in mind who is not affected.
I don't think this is correct. $0.75 still goes to the author regardless of when you cast it, it's just a question of where the curation rewards go.
It is nice to see someone else complaining about this, though, as I've been kind of getting tired of it. :)
That is not how I have been seeing it. I just posted this comment and immediately upvoted it 100%. (~$1 vote). I had a post the other day after the fork that had an auto vote, and it showed the full vote value, then not long later the vote changed from ~$10 to $0.04. (Will see how this comment pays out). I can't explain why it shows $10 initially, then an hour later it was $0.04 but that would fit the scenario I explained.
Test
If so this is a bug. It's not intended behavior.
Very possible, might have been something with 0.20.2 (which is what we were on at the time), but I did see it first hand do exactly as I describe, but very strangely it initially showed the full vote value.
That sounds like a weird condenser bug. It shouldn't show up on the display at all.
I'm starting to think it might have been. I had to remove a $10 vote and re-execute it because it was pretty much zero'd as it was done with the post.
Interesting experiment. Let's see the result. I'll keep an eye on it too. Thanks
interesting, lets see what happens if you upvote this comment 100% ;)
That sounds really strange! Could you link the corresponding comments here?
from $10 to $0.04 is a huge step, but certain fluctuations on the pending payout value are certainly expected from the general lack of voting stake in the last couple of days.
It was this post, a vote was made immediately and it showed up in the condenser as it normally would.
https://steemit.com/curation/@buildawhale/buildawhale-curation-digest-09-25-18
When I looked from my main account, probably around 1 hour later, it showed it only had $0.04 and that vote was on the post. I removed the vote and tried again, and it showed properly. But now I am unable to mimic the behavior and think it might have been related to v0.20.2 as it was on the 25th when hard fork just happened to land.
I have a post above I am using a test to see how it pays out. But I may have been wrong, this may have been an issue with an older build and has been corrected.
Argh, due to the unvote/revote the original rshares aren't stored anymore, so it's almost impossible to know what happened there. Let's see how your experiment above goes. From what I understand from the steem sourcecode, I agree with @tcpolymath.
Was this your first vote after the fork? Those were exceptionally weird.
The vote was before the fork but I noticed it after.
Yes, that's how I understand it as well. It is the curation reward that will turn into dust. The author will now always get 75% of the reward. So early voting will no longer bump the reward to the author. For the curator there is no change compared to HF19, except for the threshold decrease to 15min.
That's how I understood it should be, but I was not seeing that right after the hard fork landed on the 25th. I ended up removing the vote and re-executing it like an hour after the post and it registered as a full vote. I'm starting to suspect something was fixed between v0.20.2 and where we are now.
In fact, now that I think of it, I think the vote was right before HF20 was launched. It was after HF20 when I noticed the vote didn't count but was on the post.
It seems we all complain a lot about hf20 these days.
The only thing, next to showing up all errors and miscalculations in HF20, is to remove our witness votes from the top20 witnesses, who all agreed to it and found nothing wrong with it.
Not one opposed it.
All went immediately to the next version.
It looks as if T20 witnesses are super fast at code review and testing.
But on the other hand..
You already vote for good witnesses and rarely any t20
No need for me to preach here.. Sorry:)
I don't mind. I figured out a while ago that my vote is meaningful to witnesses outside the top 20 - it increases their chances of getting a block - and almost meaningless to those in the top group.
Hi! The behavior you have described does not seem to be intended. Only the curation rewards (25% to 0% scaling linearly for first 15 mins) is returned, so the author should still get 75%.
It says the "early voting penalty" is returned, which is the 25% at 0 mins, so the author should still be getting 75% (at all times instead of the old 100% to 75% scale) and curators have the same scale as before (0% to 25%).
I hope this is correct and makes sense!
Hi Mark. Waited for the 15 minute mark and it is a little more hands on now. I never really cared about when I voted but making more of an effort to actually think about it now.
and I want to vote your comment, but it's 2 minutes old. My $1 vote is now worth 13 cents, and I get $0 in curation rewards.
A good example here. I'll likely forget about this by the time it gets closer to 15 minutes, especially since I'm going to bed.
Good point. Was wandering why some peoples comments I voted on were different with curation so must remember to wait on comments as well. Makes sense now. Have already voted on a comment straight away recently now will remember to wait. Thanks you have been a big help.
It's really nice explanation there Mark, although I think that the votes (either it is for the comments or the post itself) are not affected at all.
What IS affected though, is the curation rewards for each vote. Meaning that your 1$ vote on a comment will still be 1$ even under the first minute but the curation rewards for that comment if it is voted within the first minute, drastically reduces.
If I am wrong plz someone correct me here
WHAT!? I really don't like the whole voting deal. I love to give my followers large comment votes to encourage more comments and conversation. I understand how this can be helpful in a post scenario. But bad for comments. I wish there was a difference.
I am someone who likes to respond to comments and I upvote the ones I like, so the 15min rule for comments def bites me in the ass.
!pancakes
Thanks for contacting Pancakes Express!
Would you like to order pancakes?
Please upvote this comment with your order:
If you would like to order your own pancakes, just make a comment
!pancakesanywhere on the blockchain and we will send you a menu.Darn, I need to wait for my pancakes longer as well now!
Excellent illustration
Nothing has changed @themarkymark. I have an account who votes for me at zero minutes. He has been doing this since way before HF20, possibly as far back as HF18 he started.
He is actually a friend who doesn't bother with Steemit anymore, and I've told him to change the time of his vote several times. As he get's zero rewards by voting for me on zero minutes.
The only change is who his reward goes to, before it went to me, now it goes back to the pool.
(voting up for visibility).
Cg
This is certainly going to hurt user experience. We will eventually adjust, but it will take a lot of time and things may not be the same