Livejournal population statistics, round 2.

[pyrop]

It seems like every time LJ does something stupid I get inspired to write another population post. I tried tackling LJ's gender balance back in November in this post in lj_research, but the raw data makes no sense and I sort of gave up. Oh well.

Today I'm going to update the population post I made in August of last year. I'm only going to show data from Jan 1

Comments

[insomnia]

It's a shame you did this now, and not a few months from now, as it would be more revealing

[oedipamaas49]

"How many posts people are seeing in their friends list over a fixed period of time is how active they perceive the site to be"

Yes. But many people would perceive this regardless of whether LJ is actually declining, because:
1) People make a lot of LJ-friends soon after joining the site, and then pick them up more gradually over the rest of their time here.
2) People mostly friend accounts which are active at that time
3) Accounts go through periods of activity and inactivity (maybe most commonly, a period of activity which trails off to nothing)

[you could probably pull together some stats for all these, if you tried]

So anybody who isn't constantly making new lots of new friends will likely see their friends-list activity drop off over time. And they may well wrongly attribute it to the overall decline of LJ.

[insomnia]

I agree, but it's still a problem, in that there is an increasing gap between "generations" of LJ users that is actually getting wider, which effects the overall appeal of the site.

Part of this is straightforward demographics... but part of it is site design. There's not enough being done to encourage new users to integrate more completely into the pre-existing community.

As a result, if you were to map LJ as connections, both the older and the newer generation of LJ are increasingly less connected.

[pyrop]

Yeah, those friends-only people. What a bunch of burdensome jerks.

Thank you for this post

[speck]

I wonder how much of the "accounts" are syndicated feeds? I know of a few people who are leaving LJ for IJ or other places, and they or others are creating "LJ feeds" to make it easy for those staying behind to keep up with them.

For example, someone very recently created a "feed" for the announcements account from insanejournal to keep up with announcements from there - there may be others, official or not.

Re: Thank you for this post

[matgb]

Feeds aren't counted in these stats, these are just LJ users actually posting. If feeds were counted, the numbers would be insanely huge, and the number of feeds created by decampers wouldn't scratch the stats.

Re: Thank you for this post

[ext_72796]

How about openid? It seems like logging in with OpenID causes a fake user to be created. Could it explain the peak?

Re: Thank you for this post

[matgb]

I don't think so, pretty sure OpenIDs are treated just as feeds-I know that earlier when I was logged in using one of mine several links were disabled and others gave "sorry, you need a real account to do this" error messages.

It definitely wouldn't explain a peak like that, unless something very weird happened to impel a lot of people to log in using OpenID in a very short time, OpenID use is fairly light on the site from what I can see.

My best bet is as said above about Greatest Journal-a lot of people came back when that site started dying, and others came here from there and created new accounts.

[nasick]

Regarding Big spike on Jan 28 in new accounts - it seems to me that it was a day of non-human registrations (bot or smth), and I feel like it were only Basic accounts (~33,5K), and the Q is how did they manage to pass through protected process? what was the bag they used?

[redhotlips]

Is it possible that after the purchase of LJ, that SUP merged the journals of a Russian provider into LJ?

I don't know enough about the sale, and the mergers, but it would make sense to me that the spike in users would be a combination of issues, that could include a merger.

[gossymer]

Your stats posts made for incredibly interesting reading and I was wondering what general changes you've seen with activity in the recent few months since SUP got control. Current active ljs in 30 days seem to be over 1 mill, so I was wondering if things had stabilized or if it was something else entirely...