Thanks @roadscape, great to hear! Yes, if I remember correctly, both the before_date and start_permlink parameters did never really work. I don't need them currently but would be nice to get the start_permlink param to work in future.

Just dropping in on the most recent transaction I could find of yours @steemchiller, to show my support.

In your next update, would really like to hear what you think about all that is happening with Steemit Inc, the proposal from BlockTrades (is this Dan's company?), etc.

I'm pretty new "in here," but for the first time, I have some real concerns. And I don't know the history like I suspect you do. I certainly don't have your technical background, but highly value your opinion.

Appreciate all you do! 👍

P.S. Hope I'm not breaking any protocols by replying here in the middle of another exchange ...

Ok, I've implemented get_discussions_by_author_before_date in hivemind; and while it ignores before_date (which does seem to be completely broken in steemd anyway), it will correctly obey start_permlink. This call is very similar to get_discussions_by_blog except it ignores reblogs. And as broken as it is, it turns out it's quite popular. Can't wait for a shiny/clean new set of APIs 😝

Perfect! Yes, a clean new set of APIs would be much appreciated. One thing I noticed is that the returned items now contain a post_id, without hivemind it was just id. That's no problem for me, but maybe it would be better to use the old identifier. Just in case someone is using it already...


  1. The field last_update seems to be missing in the posts result. When using get_state( [tag] / @[author] / [permlink] ) the field is there in the result.content[] items but it is always returned as empty string.

  2. The field curator_payout_value is always returned as '0.000 SBD'. Without hivemind it contains the correct amount as soon as a post has been paid out.

  3. The method get_follow_count returns incorrect numbers in follower_count and following_count for my account.

Hast du Lust, dich mit mir zu unterhalten? Ich bin @afrog.

Moin @afrog, ich bin gerade sehr beschäftigt und mein Weekly-Post ist auch schon wieder längst überfällig, aber ich werde mich in den kommenden Tagen nochmal bei dir melden. Deinen SteemWiki-Post werde ich natürlich gleich mal resteemen ;)

Moin @steemchiller. Du bist sogar beschäftigt, wenn sie dich in Gips einschlagen, Danke für die Meldung. Auf Dich habe ich gehofft. Danke für den Resteem. Das brauchen Neue auf dem Steem ganz besonders.

Thanks for the update, could you please file the first two issues at The third one is already at

Dear Roadscape, this is how major stakeholders treat nowadays active developers who are working on improving the situation on Steem?

Posted using Steeve, an AI-powered Steem interface

I left the smallest possible downvote to collapse a completely off-topic comment in a discussion thread which I am monitoring for critical technical issues

I get it and I don't mind the small downvote, I just expected you would suggest a different communication channel, as it's hard for us to reach you.

I know you are very busy, but I genuinely thought you would be interested in our project, as it tries to tackle an important problem and aims to improve Steem as a whole.

Posted using Steeve, an AI-powered Steem interface

Hi, Roadscape. Sorry for posting an off-topic comment, but I couldn't reach you elsewhere. I saw you are supporting algorithmic curation via delegation to Steem-UA.

Have you heard about Steeve, an AI-powered Steem interface? We are combining algorithmic curation with manual curation by human readers. Steeve offers real users recommendations of potentially interesting posts outside their normal feed, which they then manually check and upvote if they like them.

The most popular posts are then rewarded with our @steevebot. This incentivizes

  1. creation of good content
  2. better curation

And of course this way we can cover much much more posts than when people upvote only what they find on their feed. You can read more about it here. I would be interested in hearing your thoughts on this approach.

Posted using Steeve, an AI-powered Steem interface