Programming Diary #31: Beyond Proof of Brain
5 comments
Summary
This post covers programming activities during the first half of December. These activities included a new toy browser extension that reports on blockchain activity levels, continued improvements to the Steem Conversation Accelerator (SCA) and the Steem Curation Extension, as well as some modifications to my personal autovoter.
As is customary, the post also looks at future plans for these tools and reflects on some aspects of the blockchain. Specifically, it revisits the topic of alignment between post value and post "attractiveness" and then discusses the new initiative of Steem as a platform for sponsors, instructors, and learners.
Background
In my previous programming diary post, here's what I anticipated for the coming two weeks
Short term, I expect to continue working on the tag-following capability in the SCA for (at least) the next one or two intervals.
I can already see that time is tight, so progress during the coming interval may be slow.
Basically, during the holiday time of the year, there's not a whole lot of spare time for hobby activities in my spare time. I expected things to be slow, and they were, but I do have some activities to write about. Basically, I guess I worked on four things after Programming Diary #30. They are:
- A new toy browser extension for graphical display of blockchain activity (SAM).
- Continuing progress towards tag following in the Steem Conversation Accelerator (SCA).
- Updating the autovoter
- Minor updates in the Steem Curation Extension
Let's move into the next section for some details about each.
Activity Descriptions
The SAM browser extension
I already wrote about SAM (the Steem Activity Monitor) here. Here's what it looks like.
I put that together as an AI prompting challenge for myself. I tried to see if I could do it in a single prompt one morning, but didn't manage to accomplish that. AI got me close enough that I could get it functioning fairly quickly, though.
I had not intended to work on it anymore after my previous post, but it bothered me that it was done in Manifest V2. I should have left well-enough alone, but I thought it would be a quick change to update to Manifest V3. Wrong. The problem now is that Manifest V3 doesn't like to let things run continuously when the browser goes idle, so now it's periodically shutting down and losing its data.
I ran into this same problem with the SCA and somehow worked around it, but I never completely understood why the work-around worked. So... in a strange sense, I guess it's good that I can continue to work through it with SAM.
Tag following in the SCA
As-of the previous post, I was collecting the posts with desired tags in a background script and passing them to the activity list window, but I had yet to display them.
After some modifications, I am now displaying the posts in the activity list window. Obviously, additional formatting is still needed.
As you can see, this example was built by following a large number of commonly used tags just to be sure that content exists for testing. Following these tags in the real world might be overwhelming.
Updating the autovoter
My auto-voter was already set-up so that it could follow votes from a designated account, but until now that was only for upvotes. Recently, as alluded to here, some abuse fighting has been taking place. So, I decided to lend my two downvotes per day to the initiative.
This meant I had to update the auto-voter to follow an account's downvotes and also to cap the downvotes at two per day and to account for regeneration after a downvote.
I have made the updates and followed a trusted account for downvoting. I have also tested that it's following and capping properly. I still need to verify that the regeneration happens as expected.
I should mention here that my opinions have evolved quite a bit on downvotes over the years. Back in 2017, I posted the articles, A game theory requirement for steem downvotes has not been shown and Down-votes: Steemit's Achilles' Heel?. During the following couple of years, we witnessed a series of "downvote wars" and other forms of abuse and I didn't warm up to them at all. However, in recent years, I have come to believe that some downvoting is necessary with the existing reward algorithms. As they say, When the cat's away, the mouse will play. IMO, we're presently seeing that old cliché in action.
I still wish for a downvote free rewards algorithm where alignment of incentives is enough to prevent abuse, but I have concluded that the system today is not that system. So now I think the goal for the current system should be to find the right balance between the potential for downvote abuse and the potential for upvote abuse.
Steem curation extension
As with the previous post, I made some minor updates to the changes that @cmp2020 is working on. I'll let him write about the new features when he finds time, but I've been using it for a few weeks now, and the adjustments feel very useful to me.
Looking Ahead / Next Steps
Steem Conversation Accelerator (SCA)
My first priority now is to finish tag following in the SCA. This means the following:
- Format the HTML output in a more attractive way.
- Check comments/replies to see if they're under a top-level post with the followed tags.
Additionally, I'd like to add the ability to enable/disable polling and display of activity by followed accounts and followed tags. This will, hopefully, be done in a way that is extensible when I add additional features for conversations where the observer has participated and for subscribed communities. Eventually, my goal is to be able to independently activate/deactivate each form of notifications, polling, and displays.
SAM
I don't know if I should spend any time on this, but it feels like an unsolved puzzle now, so I probably will. I need to figure out how to get it to poll continuously under manifest v3 and as long as I'm monkeying around with it, also add an option to pause polling.
Steem Curation Extension
I need to work with @cmp2020 to finalize his updates, merge into the master branch, and publish a new release.
In the spirit of "visibility as a service", I'm also thinking of possibly implementing something like the "vanity message" capability that I had implemented in the Steemometer. In that implementation, I had set it up so that someone could burn STEEM/SBD and have their comment memo displayed in the tool.
In order to prevent this from becoming a spam vector, though the vanity message display will default to being disabled, and browser operators will be able to select a minimum threshold for visibility (i.e. don't show me vanity message unless the burn amount exceeds 10 STEEM.)
Short term
For the short term, my goal is to have tag following implemented in the SCA by mid-January. I am doubtful that I'll be able to finish it by the end of this month.
Reflections
Proof of brain and misaligned valuation of content
So... reflections. It seems that the blockchain's collective interest has shifted to overvaluation of posts recently, and my attention has shifted with it. If the theory of the Steem blockchain is that the token's value is backed by Proof of brain (i.e. human attention), then the content here needs to be ranked according to the attention that it receives. Maybe I'm wrong, but it seems to me that we're very far from accomplishing that.
How do we get closer?
The most obvious and least controversial way to get closer is by elevating the rank of undervalued content. Therefore, I was happy to see that @etainclub is pursuing a step in this direction with EverSteem, which will provide the capability to provide additional rewards to an author after a post's payout time. This uses a similar mechanism as SteemForever once did. The technical capability is one thing, but adoption is critical. And adoption depends on investors and curators, so I hope that investors and curators become aware of this initiative.
Right now, it seems that something like 37% of all Steem Power is delegated to paid voting services. This is a big obstacle that Proof of Brain somehow needs to overcome if Steem is to maximize its value. The paid voting services aren't a problem by themselves, but it seems to me that the way they're being used is a substantial problem.
The naïve answer to this challenge is that "nice users" should downvote more, especially tier-1 stakeholders. And maybe that's part of the solution, but maybe there are also better ways to improve the situation. Maybe it starts with persuasion.
We discussed this in comments after a previous post, and @danmaruschak had this to say:
I think we should be doing more to discuss the philosophy of how things are valued, e.g. seeing "Garbage post is worth X Steem" is a signal that 1 Steem is worth what that post is really worth divided by X. That DeFi that provides bank-like "interest" returns without doing bank-like economic activity on the other side (e.g. making money from loaning out the deposits) is a signal that however much Steem is earned in interest is worth 0 real value.
So, who makes the case to investors that their ROI will be better if they contribute votes to Proof of Brain instead of devaluing their investment by propping up the value of 0-attention posts? And how?
Solving this is something like an XPrize Challenge for Steem. There are no easy answers, but if we can solve it I think that the whole ecosystem will benefit. So, I keep thinking about it.
Maybe Steem is more than Proof of Brain
I think that this initiative by @alejos7ven and @steemcurator01 is one of the best types of content that I've ever seen on Steem. With this post, Steemit has launched an initiative to train people on Steem development. IMO, this has the beginnings of a virtuous cycle that could be harnessed by almost any business.
Why? Because it goes beyond the Proof of Brain concept. I have been puzzled for years about why marketing professionals haven't figured this out. Imagine that I'm an NFL marketing executive and I want to generate buzz about the Super Bowl? How do I do it?
Well, the tried and true method is that I can spend millions of dollars advertising on various forms of media. It certainly works, but it's expensive. At the end of the campaign, I'm millions in the hole.
What if, instead, I invested $100k in a Steem campaign where I upvote people who blog about the NFL? Then, at the end of the campaign I still have my $100k + curation rewards and +/- any change in STEEM's value, and I have also rewarded some of my best fans. And, of course those fans all shared their content on other platforms because they want to build their audience and increase their rewards.
To be fair, we have already seen a very basic variant on this with the Tron Fan Club, but that's only scratching the surface of what might be possible.
Now, let's go one step further. I'm not the NFL anymore. Now I'm Comcast, and I want to teach people how to get the most out of my xfinity cable TV and Internet package. Again, I could spend lots of money to create a course and publish it on my web site and hope that people take it. Alternatively, I could invest that same money in SP and use my upvotes to reward trainers and trainees. Then, people who participate in the training will have an incentive to learn as much as they can and to share the information. At the end of that campaign, I still have every token that I started with - and I've gained curation rewards, too.
It's so simple, but almost no one has figured it out. I don't get it.
But, this is what Steemit is building to with the initiative to teach people how to develop for Steem. At the end, they still have every resource that they invested in the initiative, plus they have curation rewards, plus they have learners who have improved their knowledge about Steem development. And, they have a proof of concept that they can point to that will show other companies how to do the same.
Conclusion
In this article, I detailed the progress made on several Steem-related tools and projects, including the continued development of the Steem Conversation Accelerator (SCA), the Steem Curation Extension, the SAM browser extension, and updates to my personal auto-voter.
Beyond the technical updates, the article delves into my reflections on the role of downvotes and the challenges surrounding the accurate valuation of content on the Steem blockchain. I acknowledge that there is a need for balance to avoid the extremes of both upvote and downvote abuse, but I have conceded to the viewpoint that some downvotes are necessary under the current reward structure.
The article also describes plans to further enhance the SCA with tag following capabilities and potentially implement a "vanity message" feature in the Steem Curation Extension.
Finally, the article suggests the potential for Steem to be leveraged beyond the "Proof of Brain" concept by providing an ecosystem where sponsors, trainers, and learners can all derive benefits in a wholly new way.
Thank you for your time and attention.
As a general rule, I up-vote comments that demonstrate "proof of reading".
Steve Palmer is an IT professional with three decades of professional experience in data communications and information systems. He holds a bachelor's degree in mathematics, a master's degree in computer science, and a master's degree in information systems and technology management. He has been awarded 3 US patents.
Pixabay license, source
Reminder
Visit the /promoted page and #burnsteem25 to support the inflation-fighters who are helping to enable decentralized regulation of Steem token supply growth.
Comments