You are viewing a single comment's thread from:

RE: Daily Steem Stats Report - August 17, 2017

in #steemit7 years ago

I'm quite impressed with the unique style with which you present steem statistics, @penguinpablo. My first concern is in the area of the ratio of active users versus total number of accounts (23,740/315,955 = 7.51%). I see a problem with only 7.5% active users. Secondly, average comments/post of 3/1 means that humans are not being engaged. From my personal experience, over the last week or two, we now have spam and bots infested blogs. Sadly too from my experience, when these spam comments are flagged, they owners initiate a flag war. This statistics of yours is revealing a lot of things. Meanwhile, I would like to thank you for creating steemnow.com app, which I am now addicted to using.

Sort:  

I don't think the daily active users to registered accounts metric is particularly important, for a few reasons:

  1. Registered accounts is itself highly inflated by botters looking to a) profit with free Steem and/or b) just want multiple accounts for various reasons including sockpuppetry
  2. Daily Active Users is not a commonly used metric for competing social networks, where monthly active users is the preferred metric
  3. We measure daily active users much more tightly than competitors. Reddit for example has 250 million monthly active users at last report, but only 36 million accounts at last report. Our active users are only included if they actually have an account and post things, not just if they land on the site.

I agree with you that the spam commenting trend lately is a problem, our current methods for fighting spam are proving insufficient.

Thanks, @demotruk, for the detailed explanation about daily active users and registered accounts. So, what I should focus on is monthly active users. Well, from @penguinpablo's statistics and @arcange's that I've been following, I think the monthly average of active users is still very low. I'm glad that you've noticed the terrible trend of spam commenting lately, like they are desperadoes. The recent challenge that I had with flag war from mini accounts, led me to know @spaminator. @spaminator uses @sadkitten to deal with spam posts, comments, etc. They came to my rescue this morning and I am glad they did.

You have to have something to compare it to, otherwise how can we say that we're doing well or poorly?

Obviously we are really tiny compared to our incumbent competitors and are going to have much lower active users no matter how it's measured. We can measure it as a ratio of registered accounts, but I just wouldn't say that is particularly meaningful when registered accounts is so inflated, and the metric we are using is not comparable to other platforms where merely landing on the website typically makes you an 'active user'. On Steemit a user could land on the site, have an account and log in with it, and they still might not get counted as an active user because they did not create a transaction which made its way to the blockchain.

I have been downvoting spammers myself, especially self-voting spammers, but I don't believe that current methods are sufficient. I hope to see spammers constrained in their network bandwidth.

I have a better understanding of what you mean now, @demotruk. ''On Steemit a user could land on the site, have an account and log in with it, and they still might not get counted as an active user because they did not create a transaction which made its way to the blockchain.'' This is very true. Same here, @demotruk. I hope to see spammers constrained. Let's I forget, how are you faring with your steemit car? I just like that car. Greetings to your sister, whom I saw standing beside the car that day you put up a post about it. My kind regards.

Great post. I would love to see monthly active users as a result, vs daily but i do think the daily stat is useful as well. But we cannot expect every user to be active every day, so i think the daily 20k users is pretty good already.

Most importantly is growth and that users come and use steem regularly and get value from it.

Daily active users is certainly useful for comparison to ourselves over time (and by that it has been clearly improving).

Bots have value. They are part of every network

Those 7.5% active users - 20% Bots. So, yes, Steem is a dying network since May/June this year.