How many Snap! users are there?
Can you put other statistics about the site too - developers?
Yes, we'll do that as part of the planned big site redesign, whenever that happens.
Oooh, Sounds interesting!
Bump?
Bumping won't help; we have to get out from under school contracts.
Hi,
I'm working on a keynote on Snap! for the University (31 of march and 7th of april) and I'd like to access Snap!'s statistics, as we have for Scratch here :
Would it be possible to have some reports about users number, projects number while waiting for the new redesigned website ?
Thank you.
Hi, Nathalie. As of today we have a little more than 511.000 user accounts and 4.525.000 projects, out of which roughly 37% are shared. Does that help you? Do you need anything else?
Hi Jens,
That's pretty good. Thank you !
We should list active users (past month) and projects owned by active users. Let's not lie with statistics even if Scratch does. 
why do you say that? We need to report these very numbers every month. Yes, they are accumulated over time, just like everybody else's. But they are the hard truth, it's the state of our data base. Trying to determine and quantify "active" users is much, much hairier, and - frankly - exactly the playground for "lying with statistics".
One more request :
could we have an idea of a percentage breakdown of users by country ?
wait you can see unshared projects?
There's no difficulty in quantifying the number of users who've logged in in the past 30 days. In fact I believe we already do keep track of that number. I don't see how it would be lying to report that.
I suppose people on staff, who are concerned with how much we pay for file storage space, care about numbers since the beginning of time. But for most people, the question they're trying to answer is "How popular is Snap! lately?" And the answer to that isn't a cumulative total; it's how many people are actually using Snap!. When we finally get student accounts implemented, we should report active student accounts and active personal accounts separately.
I guess what you're saying, Jens, is that we can give a more precise answer to the question "How many Snap! accounts are there?" But imprecise results aren't lying unless you misstate their precision. What's lying is answering the wrong question, like the classic example of the anti-drug programs who try to convince kids that marijuana is a "gateway drug" because 87% (or whatever the number is) of heroin users previously smoked weed. That's the wrong question because, after all, 100% of heroin users previously breathed oxygen. Is oxygen a gateway drug? The right question would be, what percentage of pot smokers go on to use heroin?
What makes something the right question isn't that it can be answered easily, or that it can be answered with exact precision. It's that the question truly helps understanding.
So, for example, you might argue that the past-30-days question will give a misleading answer during the summer, and so we should instead report the past-365-days number. That's an arguable position. I think that a better way to deal with the summer problem is to report a graph of past-30-day numbers per month, over a year. But we could try both and see which feels more insightful to us.
If SAP wants to know the cumulative user number, (a) they're stupid and (b) fine, tell them that. But it's not what anyone else wants to know.
Ach, Brian. There must always be fighting about everything. What's wrong with answering Nathalie's question in all truth, and - I believe - she was asking for the total numbers, so she can state them in her talk, same as I have stated them in my talks - and others of us, including Michael - since forever? And, same as every other speaker at every conference is reporting the total numbers. If we did have other numbers (we don't, why do you claim this? What do you know that I don't? We might be able to write a program that derives them from other data, but we don't store the number of active user logins in our data base), we could throw them in, but if you just want to give your audience an overview over how big Snap is, those numbers aren't lying at all. It's the normal thing to say and the normal thing to put on a slide, no stupidity involved, and nothing wrong with it. Geez, what is the problem here, and how is it that there is constant quarrel about the most insignificant issues? This is so getting on my nerves. I'm just going to retreat from answering questions for good.
Sorry, @nathalierun, we do not keep track of geographic information, we only store a username and an email address so can reset your password. That's it, no further personal information.
Geez, is any disagreement with you a fight?
So I've just looked, and here's what Scratch reports:
So at the top of https://scratch.mit.edu/statistics/ they report both cumulative and current numbers, but then the rest of the page is full of
- Monthly Activity Trends
- Monthly Active Users
- Monthly Project Shares
- Monthly Comment Activity
graphs, all with various view options. (And all growing fast!) No cumulative graphs. (They have two more graphs, showing the age and geographical distribution of users, for which the issue of current vs. cumulative doesn't arise.) I don't know why they don't include the new users and new projects counts in the top right "past month" box; they clearly know them, because they're in those graphs.
I think it's fine to report cumulative numbers along with current numbers. But the point of giving numbers in a talk is to convince people how popular we are, and cumulative numbers just don't do that.
(Even current numbers aren't ideal, because, for example, some accounts are created because someone forgot their password and don't know how to fix that. In a perfect world, we'd try to count those. For example, we could notice two accounts with similar names made from the same IP address, one of which doesn't have any activity.)
I mean, thought experiment, say you want to know which is more popular, Scratch or Snap!. They're, what, five years older than we are? So our cumulative numbers are never going to catch up, even if 20 years from now we have more active users than they do. Under those circumstances, if they kept reporting cumulative numbers, and comparing theirs to ours, you'd complain.
uh, yeah... they're stored on snaps servers, which I'm pretty sure jens has access to. I doubt he looks at unshared projects, just shared projects.
you guys can just say
total users
total active users
total shared projects
total unshared projects
total collections
total active collections
total forums
total forum posts
total forum edits
and other stuff
so every Snap! Developer can see unshared projects