Showing a read count on an article, good or bad?
Just our of curiosity, How do the forum folks feel about sites that publicly display the amount of times their content has been accessed?
I'm torn on the issue.
If I see a site with large numbers I tend to think they are bragging but, If I see a site with low numbers I tend to move as they must not get much traffic.
I have such things turned off on my site. It makes it harder for me to track what articles people are reading but it also avoids the whole "refresh to check my page hits" thing.
Comments
Look at the websites of the world. You don't see "real" websites putting up hit counters. Google, Yahoo!, Amazon, Engadget, nobody does that kind of stuff. It is entirely in the MySpace/Geocities realm of websites. If you don't mind being a crummy personal home page with hit counters, animated skulls and a tiled background, then go for it. If you want to be professional, do like the professionals do.
I don't really understand how it is that you can't check the hits on parts of your own site without a public counter though?
In the process of doing this GET or POST, the web server gets lots of information. Some of the information is given by the web browser, some is on the TCP/IP layer and some the server asks for. The server takes this information and writes it down. Here is an example line from an Apache access log.
www.frontrowcrew.com 5.5.5.5 - - [13/Nov/2006:10:57:45 -0500]
"GET/forum/images/avatars/icon_65.jpg HTTP/1.1" 200 1179
"http://www.frontrowcrew.com/forum/comments.php?DiscussionID=1140&page;=1"
"Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1) Gecko/20061010 Firefox/2.0" VLOG=-
This would actually all be one line in the access log, but I broke it up to make it easier to read. The first line contains the vhost, the ip address of the requester and the date and time of the request. Normally the vhost is not listed in the access log, but I customized the format of the access log in the apache configure to make it show up there. You can customize the log in any way you want.
The second line has the GET request. Someone requested a forum avatar jpg via HTTP 1.1. Notice the 200. That's the code for success. If they requested something that was not there, it would be 404! If it was something forbidden, it would be 403. OMG, you learned something about HTTP!
The third line is the referral page. They requested that jpg as part of forum discussion 1140 page 1. The last line is the User Agent. This is where the browser tells the server what browser, operating system and rendering engine it is running. This person visited the forum with Firefox 2 on Windows XP. If you want, you can make your web page pay attention to this information.
Anyway. If you want to track stats you just keep the access logs around. Do you read them directly? Heck no. You get a program like analog or awstats or some other program to read the logs and generate a document full of statistics for you to read.
The other way to get stats is to use something like Google Analytics. What you do there is you put a Javascript on your web page. It doesn't change the way your page looks, but when people visit your page, they will end up talking to Google while the page loads. Google asks them for information, and hopefully gets it. Google takes that information and uses it to generate statistics about your visitors. You go to the Google Analytics web site and login to read it.
Those are the two major ways of getting web stats. It should be obvious why they are all flawed, and there are no better ways.
Thanks, Apreche! Sweet sweet knowledge...