Comment by lapcat | Hacker News Reader

Comment by lapcat | Hacker News Reader

Parent Full thread

> What do you think '~' stands for?

I think it stands for, you made it up. You still haven't given any citation whatever for that number.

> You gave a figure that is collected directly from the 'servers' page that literally says "Monthly Active Users".

It also literally says right below that, "Data collected by crawling all accessible Mastodon servers on Apr 15, 2023."

Look at the data. It has daily counts. For example, here's the April 15 data that matches with what the page says:

{"period":"2023-04-15","server_count":"9509","user_count":"6641082","active_user_count":"1201004"}

I think it's actually the "Monthly" label there that seems somewhat misleading, unless perhaps it means they update the page once a month.

In any case, as another commenter mentioned, DAU and MAU tend to be relatively close. You don't go from 1.2M monthly users all the way down to 50K daily active users, that's nonsense. https://news.ycombinator.com/item?id=35600846

Do you think Twitter has 6 billion MAU?

> I think it stands for, you made it up. You still haven't given any citation whatever for that number.

We both know there are NO sources. You gave a source for the MAUs. Thus, we can only give approximations and estimations of its DAUs.

> I think it's actually the "Monthly" label there that seems somewhat misleading, unless perhaps it means they update the page once a month.

Now you think it's the label's fault, after 7 years? Oh no! "It's misleading"!

Something that has been representative of Mastodon's monthly usage for 7 years is now not believable and is 'misleading'! /s Oh dear.

> In any case, as another commenter mentioned, DAU and MAU tend to be relatively close. You don't go from 1.2M monthly users all the way down to 50K daily active users, that's nonsense. https://news.ycombinator.com/item?id=35600846

Not for federated networks, yet here we are comparing centralized platforms like Facebook, Twitter and counting the numbers as if they are the same which we both know they aren't.

Henceforth, give a proper source that is verifiable that shows Mastodons DAUs and NOT its MAUs next time since you're still struggling to do so.

You're being deliberately obtuse. I said the raw data for 2023-04-15 on https://api.joinmastodon.org/statistics matches exactly the numbers on https://joinmastodon.org/servers for "Data collected by crawling all accessible Mastodon servers on Apr 15, 2023." Everyone can see that, including you if you choose. I didn't say the numbers were "not believable".

The question is what "Monthly" is supposed to mean exactly, given that the raw numbers change every day, as everyone can see from the above link. And as I said, one possible interpretation is "unless perhaps it means they update the page once a month".

> Not for federated networks

Says you, with no evidence, again. What exactly was your methodology for producing the estimate "50,000"?

> You're being deliberately obtuse. I said the raw data for 2023-04-15 on https://api.joinmastodon.org/statistics matches exactly the numbers on https://joinmastodon.org/servers for "Data collected by crawling all accessible Mastodon servers on Apr 15, 2023." Everyone can see that, including you if you choose. I didn't say the numbers were "not believable".

And that represents the 'monthly' usage as I repeatedly said. That only shows MAUs, not daily. Never was a problem for 7 years.

> The question is what "Monthly" is supposed to mean exactly, given that the raw numbers change every day, as everyone can see from the above link. And as I said, one possible interpreation is "unless perhaps it means they update the page once a month".

It is clear that it is representative of the MAU of every "active" Mastodon instance and corroborates with that fedi-list link I gave which also clearly shows "MAU" and matches with that.

There is no room for excuses or blaming labels after 7 years of Mastodon's MAUs count.

> Says you, with no evidence, again. What exactly was your methodology for producing the estimate "50,000"?

Yet it seems that you haven't even tried refuting it with any verifiable source and you are continuously passing the MAU metric as the DAU which is clearly incorrect, which I can dismiss your so-called 'DAU' source for Mastodon. Until then:

"Give a proper source that is verifiable that shows Mastodons DAUs and NOT its MAUs next time since you're still struggling to do so."

Me: What exactly was your methodology for producing the estimate "50,000"?

You: Yadda yadda yadda... [no answer]

I'm actually fine with taking 1.2 million as the assumption for the monthly active user count, but it's still absurd and inexplicable to go from that to a 50K daily active user estimate.

Me: Do you have a verifiable source for the DAUs?

You: This... https://api.joinmastodon.org/statistics

Me: That is MAUs, Not DAUs. Read this: https://joinmastodon.org/servers "Monthly Active Users"

Do you have a verifiable source for the DAUs? Not MAUs?

You: Errr......

So once again, you come back and have zero verifiable source(s) for the DAUs count for Mastodon.

> I'm actually fine with taking 1.2 million as the assumption for the monthly active user count, but it's still absurd and inexplicable to go from that to a 50K daily active user estimate.

It has always been that source for Mastodon's 'Monthly Active Users' for years and you knowingly used it to prove the numbers for the 'Daily Active Users', then when questioned you blamed the 'label' as 'misleading'.

The "Monthly Active Users" label could not have been more clearer and has been for years. You are free to dismiss my estimate, but once again you have admitted that you don't have any sources to even show me the DAUs.

So again:

"Give a proper source that is verifiable that shows Mastodons DAUs and NOT its MAUs"

I never had a strong opinion about how exactly Mastodon measures "active_user_count". I did find it curious from https://api.joinmastodon.org/statistics that they seem to be scraping the servers every day, and this suggested to me that they might be counting daily active users. I could be wrong though. I don't know the technical details of how they calculate it. And I already said earlier: "In any case, as another commenter mentioned, DAU and MAU tend to be relatively close. You don't go from 1.2M monthly users all the way down to 50K daily active users, that's nonsense." https://news.ycombinator.com/item?id=35602391 So I was already granting that 1.2M could be the monthly user account, even while considering the possibility that it might be the daily count. I don't really care either way. I don't think it helps your argument if it's the monthly user account. So let 1.2 million be the monthly user count. Now will you explain your estimate methodology?

At this point I doubt that you will. You seem to be very committed to avoiding the question, even though I've granted the assumption that you wanted.

Reply on news.ycombinator.com