TrainWeb Visit Counters
http://www.trainweb.com/visits.html

What do the TrainWeb Visit Counters found at the bottom of each page at TrainWeb mean? Which pages are counted and which pages are not counted? When are the counters incremented?

We are often asked the above and other similar questions. You will find three numbers at the bottom of almost every web page at TrainWeb:

• "Visits since Month Day, Year."
• "Railcam images viewed since Aug 01, 1998."
• "Other pages viewed at TrainWeb since Aug 01, 1998."

"Visits since Month Day, Year usually refers to just the page that you are currently viewing. There are a few minor exceptions to this rule. Sometimes we use one hit counter for all pages of the same type. For example, we use just one counter to keep track of how many photographs have been viewed at TrainWeb. We found this to be more useful than creating a separate counter for each photograph at TrainWeb.

There is a lot of scepticism among knowledgeable web designers regarding the legitimacy and usefullness of web page hit counters. There are three main reasons for this scepticism: (1) any web designer can preset, change or increment a page hit counter to any number that he would like, (2) due to the interaction between various page hit counter implementations and the way various versions of different web browsers cache graphics generated by CGI scripts, not every visit might be counted, and (3) page hit counters count the number of times a page is visited and not the number of unique individuals who have visited that page (that is, each time the same page is visited by the same person, or that person clicks on the "page reload" button, it is counted as a new page hit).

In answer to concern #1: At TrainWeb, it is critically important for us to know how much of the information that we post is interesting enough to entice visitors to keep returning to TrainWeb time and time again. TrainWeb is a full-time occupation for us and not a hobby. If we are not providing interesting and useful information to the rail community, then we might as well pack up our bags and go home! Thus, we do not monkey around with nor inflate our page hit counters! A page hit counter is set to ZERO on the day that page is created. From that moment on, the page hit counter is incremented by one each time our web server sends that page out to a visitor's web browser.

Does the page hit counter also count the visits to that web page by TrainWeb's own staff? Yes, it does. The number of hits generated by our own visits or proofing of the web page will often account for a dozen or so hits to each new page. Since each page at TrainWeb is viewed hundreds or thousands of times, we don't believe that our dozen or so visits to any one page has a significant impact on the interpretation of the number of visitors.

In answer to concern #2: We are using a method of incrementing the page hit counters at TrainWeb that insures that every visit will be counted as long as you arrived at a page at TrainWeb by a link or banner from another page at TrainWeb. The method used to cache the graphic display of the counter number by some versions of some web browsers or with certain user configurable options on almost any web browser will not defeat this method of counting the visits to the page. However, it is possible in some cases that you yourself will not see the most current count of visits immediately. The visit count might appear to be stuck, but that is only because your web browser has cached the graphic of the number and it is your web browser that is showing you the same number over and over again even though the actual count has incremented. Eventually the graphic will expire in your cache and a new graphic will load showing the correct count, which may make it appear that the count suddenly jumped up.

This may be a good time to bring up another odd observation that you might make once in a rare while. Sometimes it may appear that each time you visit a page, the counter appears to be going both up and down! There is more than one way to get to a web page. When you approach the same web page using two different methods (for example, getting to the web page by clicking on a link at TrainWeb as one way, and using your "BACK" function as another way), your web browser may pull a copy out of your local cache on one approach and pull a new copy of the page off the web server on another approach. The copy pulled out or your local cache will have an old copy of the visit count and the copy pulled off the web server will have the most current count. Thus, if you move from a current copy to a cached copy of the same page (using your "BACK" function is a common way of doing this), the old visit count will be displayed which is lower than the current actual visit count! Thus, the visit count can appear to go down when you are really just looking at a copy of the old visit count number.

Is every visit to every page always counted? No. As mentioned above, when you click on a link or banner at TrainWeb to get to a page, then the page hit counter will usually be incremented. However, if you arrive at a page by some other method, the counter often will NOT be incremented, especially if you have previously been to that same page recently. When you click on a TrainWeb link or banner, we can bypass the cache of your web browser and make sure that you get a fresh copy of the web page as well as make sure that the page hit counter is incremented. If you get to a web page by using a bookmark, your web browser "BACK" function, or click on a link outside of the TrainWeb website, we are not able to get around the caching logic of your web browser. In those cases, the web page hit counter will not always be incremented. Thus, the count on every page at TrainWeb and the count of the number of all pages viewed at TrainWeb actually underreports the total number of times pages have been viewed. This factor more than makes up for hits generated by TrainWeb staff members browsing our own web site. You should consider all the counts at TrainWeb as a measure of how many times each web page has been visited, but keep in mind that the real number of visits to each page is probably a good percentage higher than what is shown.

Although the web page hit counters underreport the total number of visits to each page, we feel that these web page hit counters are a quick and useful yardstick at measuring the relative popularity of each page and give a feel for at least the minimum number of times that the page has been visited.

In answer to concern #3: Many people don't consider the web page hit counters useful because they count the number of times the web server has sent the page to a web browser, even if the same visitor views the same page multiple times. Thus, the counter reflects the number of times the page is viewed and not the number of unique visitors that have seen the page.

To us and to advertisers, both numbers are actually valuable. The number of visits to each page can be obtained easily on the fly through these web page hit counters. The number of unique visitors to each page can be obtained, but requires post analysis of web server log files. This number is not as easy to generate and display on the fly. Rather than not display any information on each page, we have decided to display what information is easy to generate on the fly.

Look at these number from another perspective. When you are watching television, the program is considered to still be of value by both the broadcaster and the advertisers, even if the content is a repeat. Nobody makes the claim that viewers in the audience who have seen the program before don't count. If they have tuned in to watch the program again, they are included in the viewer count. Advertising on that program is still counted as exposure of those products and services even though the program is a repeat.

In the same manner, a repeat view of the same page by the same visitor does have value. Thus, counting such repeat visits in the web page counter does not make these numbers meaningless. This is especially true because the clickable banner ads change on each repeat visit to that same page when the counter increments. Another way of looking at each web page hit counter is that it is counting the number of times that each clickable banner ad is displayed! That is a very important number to us, our sponsors, and the other web sites that are hosted on the TrainWeb server that have banners displayed at TrainWeb.

You will notice that there are 2 clickable banner ads at the top of each TrainWeb page. The left ad is usually that of a TrainWeb Sponsor. The right ad is usually that of another rail web site of a non-profit organization or individual hosted at TrainWeb, or a promotion for another section or page at TrainWeb. Thus, each web page count indicates how many times these 2 banner ads have been displayed on that page. The "pages viewed at TrainWeb since Aug 01, 1998" counter indicates how many times these 2 banners have been displayed since August 1, 1998 throughout TrainWeb.

Are all web pages on the TrainWeb server counted in "pages viewed at TrainWeb since Aug 01, 1998"? Not by a long shot!

First, this counter only counts pages that have been created by the TrainWeb staff for the TrainWeb web site. There are over 500 other rail related web sites hosted on the TrainWeb.ORG web server. These are non-commercial rail related websites created by individuals or organizations for the purpose of entertaining or educating the public. More than 75,000 pages are visited at those websites every day! The visits to those web pages are NOT included in the count of pages viewed at TrainWeb.

Second, the number of views of the TrainWeb RAILcams are also not counted in the count of pages viewed at TrainWeb. TrainWeb RAILcam #1 might be the most popular page in all of TrainWeb. When someone is viewing RAILcam #1, the image automatically reloads every 10 seconds. A new clickable banner is displayed to the viewer each time the image is reloaded. However, we do not count these reloads in the count of the number of pages viewed at TrainWeb. There are so many visits to the RAILcams that including those counts would distort the counter for the rest of the website. Thus, visits to the RAILcams are not included in the total number of pages visted at TrainWeb.

When you visit a RAILcam, you will see the RAILcam counter update each time a new image is sent to your screen. If the RAILcam counter increments by more than one each time the image updates, that means that there are additional people viewing that RAILcam at the same time you are viewing it. This counter is not counting each image taken by the camera. If nobody is watching the camera, the RAILcam counter doesn't increment at all!

Third, the only pages that we count in the number of TrainWeb pages viewed since Aug 01, 1998 are those that have the TrainWeb logo at the top of the page. You will notice that most web pages at TrainWeb have a similar structure. You will usually find the TrainWeb logo, 2 small banner ads and one large banner ad at the top of each page and three counters at the bottom of the page. With one minor change, we were able to add the TrainWeb global counter to all of these pages on August 1, 1998. Those pages are included in the total count of pages viewed at TrainWeb.

There are also hundreds of old pages at TrainWeb, especially old travelogues and photos, that are not included in the count of the number of pages visited at TrainWeb. You can recognize these pages since they do not have the TrainWeb logo at the top of the page, usually have a light blue background, and if they have banner ads at all, those banner ads are not at the very top of the page. These old pages do not have the html code to increment the global counter and thus are not counted.

In mid-1998, a massive effort was made to modify all of the TrainWeb pages to conform to a standard format. In this process, a method was used to make future global changes able to be made very easily. We were not able to devote time to modify every one of the thousands of web pages at TrainWeb to conform to this new format. Thus, several hundreds of pages still remain in the old format and you will run into these pages from time to time as you dig into the lower levels of our page tree. The pages in the old format are not counted in the total count of pages visited at TrainWeb. Thus, the count of "pages viewed at TrainWeb since Aug 01, 1998" underreports the actual number of visits.

Why does TrainWeb get so many visitors? The authenticity of our counts has been questioned more than once. This doubt stems from looking around the web at dozens or even hundreds of other rail related websites. Most of these other web sites have counts of just a few hundred or just a few thousands of visits per year. This even includes the web sites of some fairly large multi-million dollar companies in the rail industry!

So, the question comes up: How can TrainWeb possibly get so many visits when most other rail web sites don't get a fraction of that many visits? First, let me assure you that our numbers are real, as explained above. There are multiple reasons as to why we get so many visits, which I will try to explain below.

Most important is that this is our business. Our main business is obtaining exposure for our rail related sponsors on the Internet World Wide Web. Our business is not selling rail books, videos, magazines, photos, tickets, T-shirts or other railfan, model railroading, rail travel or other rail related products or services. Our #1 focus is making sure that our sponsors' clickable banner ads get seen by as many people as possible that are interested in rail and are surfing the web.

We ask ourselves every day how we can get more people to find the TrainWeb website who are interested in rail and what we can include to encourage those visitors to explore the website and keep them returning to the website. We have used many techniques to make sure that our URL is known throughout the rail community, though we continue to find new sectors where we are not known and where we can continue our publicity efforts. We also use a number of online techniques to insure that TrainWeb will be found by anyone seeking rail information through any of the online search mechanisms. This is another area when no effort ever seems like enough. On top of all that, we are continually working out reciprocal arrangements with other organizations and individuals involved with rail to mutually share awareness and information.

The bottom line is that exposure of the TrainWeb website is a primary concern to us and we probably devote more time to this effort than most other organizations involved in rail. This is not to say that many organizations don't put far more effort and money into getting the word out about their organization, but those same organizations seldom put that amount of time and energy into getting knowledge of their web address out to the public. Since the web is our primary business, most of our energy is devoted to promoting awareness of the TrainWeb web address. Hence, the reason why there are so many more visits to TrainWeb than most other rail web sites.

For additional information on how TrainWeb promotes itself and why we get so many more visitors than any other rail related website, click here to review our "Advertise At TrainWeb.com" page.