RE: Hive Proposal Costs per Active User

As @themarkymark explained, its not cut and dry.

For example, I have a tool not currently available to the public till I find the time to register a domain, set up a webserver for it and publish it but what it does is look at your followers and tell you if they are active within a filtered time period you can choose and active means posted or voted on something - anything that can be seen in a block on the chain as them basically pressing a button.

I had to add a disclaimer to it that reads as follows:
"Keep in mind not all users post or vote, so they may be "alive" but not "active" on Hive."

Evident by users like myself, who didn't post in over 2 years, but I have an account, and sometimes came and just read things without taking any action at all. And posting things since I came back, has prompted contacts from many such dormant users citing they too have been merely lurking but not participating for lengthy periods ranging from months to years themselves.

Using your own profile as an example. your profile header on PeakD cites that you have 6636 followers watching your blog. That's a social media win if they're "real" followers.... for sure!

BUT

If I put your name into my tool, it looks like this:

so first we see that my tool gets a different number by checking the followers on your account via a different RPC node than PeakD uses. Little differences like this can be confusing and it is not entirely clear why they exist anyway, other than things like node sync timing and the code they use to make their numbers available and how, but they do.

Second we see that the number of these followers who filter out as having posted within the last month is only 538 of the 6K+ total.

And if I change the filter to "voted" - on a post or comment (child post) then it changes even more to 889 active within a month. More people voted than posted, which probably makes sense. Some people never write, they only read and curate. And I don't account specifically for bots versus humans and voting trail accounts and stuff that do count as followers.

So not only is it an imprecise science, I can certainly assure you there are MANY users that could be considered 'valid' accounts that are just away, taking a break, or not posting or not voting or both. But they count as being part of the ecosystem. Such prominent names are even all sizes of wallets, ranging from someone like me who left for a while, and left an empty wallet but came back and got re-engaged, or big names like CrimsonClad who is clearly part of the system but didn't post for a year or TimCliff who is a stalwart OG but barely interacts with any frequency and so on. Or ranchorelaxo/haejin, a whale not seen in years but who suddenly surprised us by dropping big votes on one of Enginewitty's posts about our new witness coming online a couple days ago after being silent for a very long time. I'd have assumed him gone and many others as well, till they weren't...

So yeah, I tried to run @themarkymarks 14K followers but my site isn't setup for pagination on the results and it broke because the http payload of his followers data was too damn big :D I'll fix that later before I publish the tool for the public one day soon!

I hope this helps. Because my tool is just a reference, it cannot tell you the truth, because the chain can't handle the truth - in any succinct, certain infallible way anyhow. Too many variables in what defines a "user".

Defining an "active" user on Hive is akin to asking someone to answer "What is the number of stars in the sky". Do they count if we can't see them but they are there? I think they do.

Not really, we have over 2M accounts but ~8,000 actively posting. You could use posts, comments, votes, custom json, there are lots of things you can do, but it isn't an exact science. Top level posts is a good metric, but for example people like Smooth and Freedom don't post, Blocktrades and GTG rarely post but are very active.

Front ends can't really tell who is logged in as it is broadcasts.

In most ecosystems, those who participate are like 10% of the total users the rest are lurkers, but in our case I think that doesn't really hold true due to the rewards. There are certainly lurkers though, but it just can't be quantified. I think "votes" is less reliable as that is more likely going to be 2+ per real user and many voters are completely gone.

I think it is safe to say it is somewhere around 5,000-15,000, but no one really knows.