"Members: 19,139" is misleading/false.

Post Reply New Topic

PS. for phpBB3, the default session length, after which it will expire, is one hour. However, it also maintains a separate concept of "Who is online", where the default is five minutes.

The "unusual profile stats" which appeared on profile pages for 48 hours in April, ( and were noticed by quite a few people who posted on the stickied thread "Updates" in this forum), accurately recorded, or at least displayed, how long we had been on during the current visit, and how long the previous visit had been, ( I think) aswell as a number for hits and visits on a certain day or since a certain time. ( I can't now remember exactly what they indicated ).

What were those figures if not some sort of useful stats about member activity/site use?

Why did they show up all of a sudden, for 48 hours, only to be hidden again by Alex because as he said we weren't supposed to be able to see them?

What is their purpose, if they are meaningless, bearing no relation to actual use/activity?

The clock time never matched, I presumed because of my being in France/a non-standard time zone, but the duration did. It seemed to have no trouble knowing that I was still around, when I was around, and how long I had been around the last time aswell. I am referring to the data which was only visible for about two days back in April.

I did notice that some people did appear as "logged on" for weeks at a stretch despite not having posted for ages, I supposed as a result of not logging off when finished session.

Why would the data about my activity on my page be/have been more accurate than that about you appearing on your profile page?

I just explained that.


I think you just answered your own question?


You read the threads - some people felt it was intrusive.


They were not meaningless. I explained them.


WP knows ONLY when you make a request on it. This is what "state-less" implies. It can only know what you are doing at the instant you request a page. It guesses at some sort of continuity. This is configurable. I explained this.


How did you determine this? I don't recall the extended stats in detail. If anything suggested a session was being maintained for a long while, I would suspect that the user was operating a robot (literally), in order to continuously make requests, for some purpose. I think Alex tolerates such things, provided the user does not impose too great a load on his server. As I explained, I have never, to my recall "logged off" from this machine/browser combination, therefore this particular "login" should read as about four months long, I think.


I have no idea. Why do you think it is (was) more accurate?

The reason I asked all those questions is because your previous posts seemed to be saying that the figures had nothing much to do with indicating/registering real activity, that they were unreliable/erroneous/unconnected to actual use, etc. I couldn't understand it. That's all.

So presumably it would be possible for Alex to show figures somewhere of actual site-use/activity.

So, I thought I explained it. Several times. I'll say it again.

All WP ever knows about is those instants when you make a request for a page. It has no idea whatever what's going on in between. This is the same for every HTTP "connection" in the world - it is not a connection at all, just a series of page requests, separated by minutes or hours.

WP (and every HTTP site is the world) makes some attempt to string those requests together into a "session". It is entirely heuristic.

Given that, what do you mean by "actual site-use/activity"?

Ok, I got it this time. Thank you for the repeated exposition!

I didn't know that about http connections. That is interesting.

About "actual site use" I meant stats showing simply "how long people spend on wp". Though I now understand what you mean and see why would never be exactly exact. Sorry . Computers are like foreign language, one with a different alphabet and completely different sentence structure, to me. Thanks for explaining.

OK. I know about these things... too well, at times. It means I (nowadays) realise that my "explanations" aren't very explanatory.

I was blocked from WP for a while (and I didn't know I was) and tried to create a couple of accounts (I had left a week and a half before saying some suicidal stuff, and I wanted to say that I was okay)
I never got banned for it luckily (I don't even remember the account names)!
I wish you could delete accounts after a years worth of not logging in (if you wanted to start again and your account was deleted I think it would be good to get another account or just find another website)

I thought I'd bring up a interesting statistic Alex has restricted access to, of the 19,189 accounts, of which there really are 19,189 accounts as I write this, there are 8,606 without a SINGLE post, EVER.

Can one say dummy accounts? Spam bot created accounts? How about wasted database and overhead that could easily be deleted without affecting anyone, maybe just, say, dump all of the 0-post accounts that are over 1 month old and haven't been accessed? the function is built into the server, and it'd reduce the load greatly.

A lot of people like to read the site but don't have much they want to say. I guess that explains why some people don't post.

Makes better sense to me, alex.

It's not restricted access. All members can find that out on using the "Browse Users" function on their account page, choosing listing by Total Posts, and scrolling to the page in the list of members where see 0 posts marked against users names and seeing what number in the list is. It's easy, if a little laborious to page to there.


You need to be a member to access two forums on the site, Adult and Members Only. If people who currently don't post had to post at least once a month to keep that access, the load on the site would be greater than now.

The useful figure would be the "approximate" time spent on site/logged in, or http page-requests. That would show whether were inactive or not. But it's kind of personal that, when shown individually.

I'm also a not-too-atypical case.

You can easily see that pretty much 50% of people do indeed join, but not post at all to start with.

In my case, I joined on June 18th, 2006, but my first post was Feb 7th, 2007. I.e. it took me nearly eight months to work up the effrontery to contribute to the site.

I have no idea, during that eight month period, exactly how often I even looked at the site. It was basically my "self-diagnosed" period. I would have been extremely unlikely to post at all, if my membership had been arbitrarily removed. (Maybe you think that would have been a plus? )

Another member, a while back, admitted(?) an even longer gap between initial registration and first post.

=============

I can see no point in any discussion of "how accurate" WP's member count happens to be. It is no different from the member counts of any other organisation - even non-free ones.

I read this thread

Then the guys username

THEN I LOL'D

From what I've seen on other forums I'm on, this is very typical.
People join a site, sit there for months or a year, and then post.

Me too. I'm still loling.