What is the most active Forum? (Warning: no photos, can be boring)

[Olli]

I made some statistics about Forums. This might help you to decide in which Forums you should be posting, if you like to get most views or responses.

This is obviously nothing important and boring for most of you.

How did I analyze the data? Last Thursday evening I copied the information of the latest 500 new posts of the last 20 hours in to a spreadsheet.

I collected the numbers of Posts, Views, Replies and Threads in each Forum and made some cross-tables. Each different view to the Forum was ranked and score given.

There were posts in 26 Forums. I used the following criteria in each Forum (most active or highest scored Forum got 26 points, last 1):

1. Number of new Posts
2. Number of Views
3. Number of Replies
4. Percentage of Replies versus Views
5. Average number Replies per Thread
6. Average number of Views per Thread

So let's start with the summary. The most responsive Forums were People and Canon EF and EF-S lenses, the least Galleries and Small Compact Digitals by Canon.

http://www.laasanen.net/POTN/stats/final.jpg

Biggest number of new Threads were in Nature & Animals, Canon EF and EF-S lenses and People Forums, smalles in Other and Wedding Photography.

http://www.laasanen.net/POTN/stats/threads.jpg

Most Views were in Canon EF and EF-S lenses and General Chat Forums.

http://www.laasanen.net/POTN/stats/views.jpg

Most Views per Thread were in Glamour & Nude and General Chat Forums (so now you know where to post if you like to be seen). Galleries' Threads we least viewed.

http://www.laasanen.net/POTN/stats/viewsperthread.jpg

But what about to get comments to your posts? The winner in this scoring was ... Wedding Photography, and the good second was Macro Forum. Glamour & Nude was creating least number or Replies per View.

http://www.laasanen.net/POTN/stats/replies_vs_views.jpg

And finally, most Replies per Thread were given in General Chat and Forum Talk.

http://www.laasanen.net/POTN/stats/repliesperthread.jpg

[Scottes]

Pretty interesting.

I'd say that you have too much time on your hands, but I've done similar stuff before so that would be hypocritical.
:wink:

[Big_B]

What about endogeneity problems? Doesn't this tell you what type of photos you ought to be taking if you want to get responses, rather than where the photo should be posted? :)

[Olli]

Pretty interesting.

I'd say that you have too much time on your hands, but I've done similar stuff before so that would be hypocritical.
:wink:
Thanks Scottes for commenting. Some of us have weird hobbies, among photography :cool::rolleyes::D. Actually, copy-paste to excel, some sorting and using user-defined template for the graphs didn't take that much time. I used about the same time in selecting the background picture to the graphs (that yellow, massive sun-like metal ornament in the ceiling of the Peninsula Hotel in Manila, picture taken about year ago) than to the actual statistical play :lol::lol:.

[Olli]

What about endogeneity problems? Doesn't this tell you what type of photos you ought to be taking if you want to get responses, rather than where the photo should be posted? :)
Very valid point, Big_B. Endogeneity is a real concern in statistical analysis, and not thoroughly discussed in this study :p. Still I believe, both the Forums as well as photos are important if you are looking for maximum visibility. (But photos are the reason why we are here;).)

[defordphoto]

Just remember that you must post your post, photo or whatever in the proper forum and not just where you think it may get the most views.

[condyk]

I know you Finish guys only get maybe 10 minutes of daylight a year or something but haven't you ever heard of TV or Monopoly? ;)

[Big_B]

I know you Finish guys only get maybe 10 minutes of daylight a year or something but haven't you ever heard of TV or Monopoly? ;)

Rofl!!! :lol: :lol:

[Big_B]

Very valid point, Big_B. Endogeneity is a real concern in statistical analysis, and not thoroughly discussed in this study :p.

Lol, good reply Olli! :D

[Olli]

I know you Finish guys only get maybe 10 minutes of daylight a year or something but haven't you ever heard of TV or Monopoly? ;)
Just guess, what else that tv you can do when it's dark ... and that is not post processing :cool:. Btw, I envy you in UK, cause I like the rain (should post some pics from my hitchhiking in Isle of Sky some tens of years ago, if I just could find the slides, taken with FTb:rolleyes::D:D)

[Maureen Souza]

Very interesting........ thanks for the stats.

[Falkon]

That's a lot of work. Thanks for the graphs.

[JohnnyG]

I know you Finish guys only get maybe 10 minutes of daylight a year or something but haven't you ever heard of TV or Monopoly? ;)

And, it is dark at this time of the year too so I guess that explains everything!

Except endogeneity. All I could find was endogentic which are rocks forming beneath the surface or something like that?:confused::confused::confused:

[Big_B]

Endogenous just means created within. To give you an example, there is a high ratio of views to comments in Glamour and Nude. In this study you can't tell whether posting in glam&nude leads to high view:comment ratio, or whether there is a common characteristic of the photos posted in glam&nude that leads to them getting lots of views but few comments. (I'm sure we can all think of a few reasons why it is most likely to be the second of these. ;))

The key issue is that the endogeneity problem means that you can't tell what would happen if you posted the same photo in a different forum.

[JohnnyG]

Endogenous just means created within. To give you an example, there is a high ratio of views to comments in Glamour and Nude. In this study you can't tell whether posting in glam&nude leads to high view:comment ratio, or whether there is a common characteristic of the photos posted in glam&nude that leads to them getting lots of views but few comments. (I'm sure we can all think of a few reasons why it is most likely to be the second of these. ;))

The key issue is that the endogeneity problem means that you can't tell what would happen if you posted the same photo in a different forum.


Thanks for the explanation of endogeneous. Makes sense.
And we both know why the ratio for Glamour and Nude is what it is!;););)

[Olli]

Thanks Sam, I was about to post this link (http://faculty.smu.edu/tfomby/eco5350/powerpt/ch15.ppt), but your concrete explanation is closer to Forums. :lol:

Next I probably need to analyze how the length of the title impacts the view/reply activity.:eyes

Regarding glam&nude, here is one more example from today's stats.

http://www.laasanen.net/POTN/stats/final051121.jpg

[Big_B]

Ahhh.... another economist :) Are you studying or working?

Another interesting topic to analyse would be the effect of putting your title in all upper case. Some people seem to do it in order to gain attention for their thread and I've often wondered how much difference it actually makes.

[Nidz]

Interesting stats..

What about endogeneity problems? Doesn't this tell you what type of photos you ought to be taking if you want to get responses, rather than where the photo should be posted? :)

Quite the contrary. I feel that the neglected forum areas should be hyped up in some way so people will feel more encouraged to add more items to these areas instead of everyone taking pictures of the same type of theme eg/ macros of flowers. I guess there's going to be less people who are going to experiment with black and white photos etc but not a lot you can do about that.

I hope i'm not missing a point here.. lol. I know I try to cover all the areas and post comments on almost any pictures.

You forgot a graph showing the most active members in the foum.. That would be intersting.


Remember, 5/4 people are not too good with graphs.

[kram]

Do you also have views split by members and non-members? Would be interesting to see which forums are viewed most by non-members.

I love such analysis. I would never have guessed 'people' to be at the top. Had always thought 'lenses' was... Great work Olli :)

[Big_B]

Quite the contrary. I feel that the neglected forum areas should be hyped up in some way so people will feel more encouraged to add more items to these areas instead of everyone taking pictures of the same type of theme eg/ macros of flowers. I guess there's going to be less people who are going to experiment with black and white photos etc but not a lot you can do about that.



I don't disagree, I was just talking about what the stats show, rather than what we should be doing. By the way have you looked in on the Other forum recently? We have a few really interesting stickies going.

[Olli]

Ahhh.... another economist :) Are you studying or working?

Another interesting topic to analyse would be the effect of putting your title in all upper case. Some people seem to do it in order to gain attention for their thread and I've often wondered how much difference it actually makes.
Hehheh ... years fly, I'm an old f*rt and studies are somewhere in the shadows of the history ...

A quick and dirty answer to Sam's hypothesis: correlation coefficient between uppercase letters in the title and number of views in yesterday's sample of 500 data pairs was -0.02 (for the length of the title that was 0.01), so *shouting* seems not to help to gain attention :lol::lol: but I need to have a second look on the raw data, i'm afraid there are data points which might create a biased result.

[Olli]

Do you also have views split by members and non-members? Would be interesting to see which forums are viewed most by non-members.

I love such analysis. I would never have guessed 'people' to be at the top. Had always thought 'lenses' was... Great work Olli :)
Thanks Kram. I don't have access to this split data (or do I?), but certainly an interesting topic as well.

Btw, yesterday the top 5 forums, based on the same calculation method as last week were

Canon EF and EF-S lenses
General chat
Canon EOS digital cameras
Transportation
Macro

People Forum was 7th. obviously this sample of 500 is not representative enough. :rolleyes: This means more work to me ... huh ...

[Pekka]

Thanks Kram. I don't have access to this split data (or do I?), but certainly an interesting topic as well.

Well, I could give you e.g. one year of thread view/title/forum data, all I need is to write a script for it. But my time is now limited so if you can do php and mysql grab a vbulletin and go for it :)

Of course all reports would have to pass though the office so they would always be positive :)

[Big_B]

A quick and dirty answer to Sam's hypothesis: correlation coefficient between uppercase letters in the title and number of views in yesterday's sample of 500 data pairs was -0.02 (for the length of the title that was 0.01), so *shouting* seems not to help to gain attention :lol::lol: but I need to have a second look on the raw data, i'm afraid there are data points which might create a biased result.

That is a facinating result. I'd be interested to hear an update when you've had chance to go through it in more detail.

[Olli]

Well, I could give you e.g. one year of thread view/title/forum data, all I need is to write a script for it. But my time is now limited so if you can do php and mysql grab a vbulletin and go for it :)

Of course all reports would have to pass though the office so they would always be positive :)
Aaah, this is getting serious. :confused: Pekka's proposal is tempting, as I'm planning to take my first infantile :o steps towards the unknown world of mySQL. But as the bread and butter is coming from other sources, I also need to put a limit somewhere.

Still, as long as the office allows, I might post some more complitely insignificant views to the behaviour of Forum members. Soon I have to take a short break, but before that a quick view to the lengths of the threads.

Top three in average length of the title of the threads (last Monday) were

* Exhibit Engine discussion (need to be long as topic is so difficult?)
* Canon G-series digital cameras (hardware guys like to be exact?)
* Forum Talk (talk, talk, talk?)

Shortest titles were in

* Critique Corner (hit me?)
* Glamour & Nude (less words, more pics?)
* Other (?)

Questions: Is there a deeper meening in this? Will the mankind get better after knowing this? Which was first, chicken or egg?

Here is a graph (edit: in this updated graph special marks have been counted separately).

http://www.laasanen.net/POTN/length.jpg

[Scottes]

I'd think you guys were all nuts, except that I once wrote a Perl script to access all the forums and get post/view/reply counts... Olli, can you do Perl? The scripts would need some tweaking.

[ericgtr]

Olli, your work is admirable considering you did all that manually. A few basic sql queries will parse all of this out instantly though, hopefully pekka writes it up for you to save some time on all of that. Your graphs are very cool :)

[Olli]

I'd think you guys were all nuts Can not agree more :shock:

Olli, can you do Perl? Would love but, sorry, can not :(

[Olli]

Olli, your work is admirable considering you did all that manually. A few basic sql queries will parse all of this out instantly though, hopefully pekka writes it up for you to save some time on all of that. Your graphs are very cool :)Thanks Eric for the compliment. If Pekka just will have time to write the scrip, and I can get the data, I might promise to have a new wierd look for the stats :p(published only after The Office' formal approval :D).

Btw, there are more moderators' than members' comments (12 vs. 7, I did not count myself) on this thread [shivering].

[ericgtr]

If it's one thing they have around here it's a great bunch of mods who are hands on and helpful all over the place.

[ericgtr]

I'd think you guys were all nuts, except that I once wrote a Perl script to access all the forums and get post/view/reply counts... Olli, can you do Perl? The scripts would need some tweaking.
I used to dabble in perl a bit but that was long ago, still your script must contain the queries to the mysql db right? I would think that perl only acts as a handler for connectivity then runs the queries, no? If you want go ahead and send it to me Scottes and I will see what I can do to dial it in with vb on my end for you guys.

[Olli]

If it's one thing they have around here it's a great bunch of mods who are hands on and helpful all over the place. ... and I truly hope they will continue to do so. This is one of the big things why I hang around here.

[Scottes]

I used to dabble in perl a bit but that was long ago, still your script must contain the queries to the mysql db right? I would think that perl only acts as a handler for connectivity then runs the queries, no? If you want go ahead and send it to me Scottes and I will see what I can do to dial it in with vb on my end for you guys.
Actually the scripts do a bunch of HTTP GETs and then parse the source for each page. Simple stuff - I don't do very well with SQL.

[ericgtr]

Scottes, got your scripts and it looks like you've done a good job of grabbing stuff from the outside. Here's the thing though, I also run a vb 3.5 baord at one of my sites and looking in the admin control panel these stats are mostly all built in already from what I can tell. No point in reinventing the wheel if you know what I mean :) Maybe Pekka will parse some of it out (sorry dude, didn't mean to throw you under the bus ;) ) when he gets some time and post them up for Ollie pick through.

[Scottes]

...looking in the admin control panel these stats are mostly all built in already from what I can tell. No point in reinventing the wheel if you know what I mean :) Maybe Pekka will parse some of it out...
And there's the rub - Pekka's time. It took me less time to write those scripts than to compose a begging email to Pekka. :-)

[ericgtr]

And there's the rub - Pekka's time. It took me less time to write those scripts than to compose a begging email to Pekka. :-)
Not to mention incentive to even do it, I get requests like this and usually ignore them myself lol. :p

[Olli]

Scottes, Eric, looks like you are both doing a great work, creating exiting expectations and putting some pressure to find our something new. Let's see how extreme we go ... :lol:

What else should/could we find out from the material? Do the profile of members who are posting mainly pics differ those who are just commenting? Which country is most active (population weighted) in posting/commenting/forum? At what hour-of-the-day/day-of-the-week should one post to get best response? When ... how ... why ... ? :rolleyes::eyes:rolleyes:

[elTwitcho]

I think it's always interesting and a somewhat depressing statement on some people's priorities that there are more threads in the "equipment talk" section than there are in the "share photos" section. But, you can't tell people where to focus their interests so... :neutral:

[Belmondo]

I think it's always interesting and a somewhat depressing statement on some people's priorities that there are more threads in the "equipment talk" section than there are in the "share photos" section. But, you can't tell people where to focus their interests so... :neutral:

The fact that it depresses you really speaks more of your own priorities than the majority of other members. And you're absolutely right: We can't tell people what their priorities should be.

Historically, this has always been more of a 'gear' forum. At least that's the impression I got when I first joined 2 1/2 years ago. Nonetheless, the 'share' forums have really grown, both in numbers and in participation, especially in the last year, and we're developing more 'niche' users (i.e. wildlife, macro, cars, etc.).

I think these are very positive signs, and are signs that the forum is an ongoing work in progress. Chances are, it will look even more different a year or two from now.

[elTwitcho]

The fact that it depresses you really speaks more of your own priorities than the majority of other members. And you're absolutely right: We can't tell people what their priorities should be.

Historically, this has always been more of a 'gear' forum. At least that's the impression I got when I first joined 2 1/2 years ago. Nonetheless, the 'share' forums have really grown, both in numbers and in participation, especially in the last year, and we're developing more 'niche' users (i.e. wildlife, macro, cars, etc.).

I think these are very positive signs, and are signs that the forum is an ongoing work in progress. Chances are, it will look even more different a year or two from now.

Which I think is interesting actually, because I joined simply because it was the best photography forum I could find, Canon specificness and gear discussion wasn't at all an issue. When I went looking for a forum three or so months ago (or maybe longer, hell if I can remember when I joined) this forum had the best photographic discussion I could find anywhere, so here I stayed. As someone who wasn't around for the early days when it was primarily gear focussed, I guess I have a whole different perspective than some of the older members might have.

Kind of neat really.

[Big_B]

When I went looking for a forum three or so months ago (or maybe longer, hell if I can remember when I joined)

There's a small clue below your avatar ;)

[kram]

I landed up in this site when I started doing some research into what to buy for my DSLR setup - and my interest is still more on the gear forums than the picture forums.

To be honest, there are tons of photograph sharing websites around but none that has such a helpful set of really experienced folks to discuss gear decisions specifically to suit your needs.

[Michaelmjc]

Where the heck is a wedding forum? I dont see it anywhere!

[Pekka]

Where the heck is a wedding forum? I dont see it anywhere!

It's a subforum to pro forum: http://photography-on-the.net/forum/forumdisplay.php?f=78

[Olli]

Where the heck is a wedding forum? I dont see it anywhere!
You might be also confused because Wedding forum was visible in the first analysis I made but not in the last. The reason for this was the data mining technique: the first set included all 500 newest posts (clicking 'New Posts' in the top bar), there you had new wedding forum postings visible, and the last was collected from the 'main view (http://photography-on-the.net/forum/index.php)', and there you can't see that as Pekka explaned.

- olli

[Wazza]

And here I was manually counting the stats... one hundred and forty six thousand two hundren and errr ummmm ohh dearrr... damnit, lost count
One, two, three

[Olli]

Wazza, that was your two thousands' post - congratulations! :D:D

43961

But who would be the happy one to post the first millionth? Somebody keeps track on that?

43962

[Olli]

But who would be the happy one to post the first millionth? Somebody keeps track on that?

43962

Today 14 Dec 05 (one day later) about 2500 more posts! One million is going to be broken in couple of days!

44173

[Olli]

Now, 15 Dec 2005 at GMT 09:00 there are c. 997,950 posts.

Tomorrow we will break 1,000,000! :D:D