I don't see why this has to be "private". Discussing ideas is what the forum is for. Not "clutter" to me.
The basic system goes something like this:
One host computer is running, that takes most of the action.
The host computer may passively receive data from the datalogger, or actively request it periodically.
When the data is received, it is stored by the host computer.
The host also has a separate program running which reads the stored file, prepares an update, and sends that to a remote internet server.
I think that is done by FTP. It doesn't have to be complicated, but firewall on the host computer needs to be aware of what's going on.
The internet server's job is to host the data file and a script file that periodically reads the data file.
If the script file finds an update, it prepares a refreshed HTML page, and overwrites the old copy of the HTML file.
The HTML file will refer to any numbers or graphs that were prepared by the program(s) on the host computer, and included in the update message.
A visitor who is viewing the HTML file will see the last update.
Buried in the HTML code is a Java command that tells the visitor's web browser to check for updates periodically.
When the time elapses, that visitor's browser will refresh with the updated HTML file. Up pop the new graphs and statistics.
My weather station came with a suite of little programs that can do all this. I've experimented with them but I haven't loaded the scripts onto my website. I chose to use the other auto-update feature in my weather station's software instead, which doesn't require my home computer to be running all the time.
If I was to use it, it would look like this:
http://www.rigel.ca/Wx/Current_Vantage_Pro_Plus.htmThe page isn't complete or updated any more. It doesn't even show the time & date of the last update for some reason. Anyway it serves as an example of the simple HTML page that needs to be generated by the script, if you right-click to "view source".