A couple of recent threads have piqued my curiosity about how bandwidth transfer occurs when visiting a web site. My question is: Does bandwidth transfer of some sort continue for the entire time you’re logged on to a particular website, or is there an initial transfer when you arrive at the site, with no other transfer occurring unless you click on a link or refresh the page?
This question is particularly important to me as I often log on to some site or other, and if I know I’ll be coming back to it later or that I will use it sporadically on and off during the day, I’ll just stay logged on. But I don’t want to do this if it’s costing the website for bandwidth usage while I’m not actively using it.
Generally, once you have hit a page and it’s loaded, thats it. Having the page open on your computer is not going to download any more. There are a few exceptions:
(1) pages that automatically refresh, such as a webcam, or a stock market report. If it does this, you’ll see it flicker as it reloads, and if you’re on a slow connection, you’ll see images loading bit by bit. You can simulate this by clicking the “refresh” button on your browser. Each time the page does a reload, it’s contacting the server, and getting new data, costing the site (and probably your ISP) bandwidth.
Web designers can set the frequency that auto-reloads happen, could be seconds or minutes or hours.
Most pages on the internet are pretty small (the phrase “sparrow fart in a thunderstorm” comes to mind), but there are exceptions.
(2) Reallllllllly big pages that have a lot of inline images - it might appear to be finished loading at the top, but it’s still downloading stuff further down the page (an example might be a discussion board thread where people post pictures of their pets, say). If you usually just check the top of a page then move on, the page might not have ever finished downloading.
That’s a bout it, other might think up some other situations, but frankly they are all pretty unlikely. If you’re really paranid about it, list the URL’s you’re talking about leaving open, and we can check 'em for you.
A web server is essentially just a file server. Your computer requests index.html which contains information on how to display the page and what other files you will need to request/transfer (pictures, etc.)
After the web page is transferred (except with streaming data or something similar) there’s no more communication until you make a new request.
I’ve now thought of another question, if anyone would care to answer it. Does the webmaster know about it if I sit there at the site for several hours (or all day), and is there any practical reason why he/she would care?
I can’t think of any reason they would care, unless you are a competitor. In that case they might take a look at your website to make sure you aren’t “borrowing” code from their site. They might also be concerned if you are re-loading their site every 10 seconds, 24 hours a day.
Just one more comment on the OP. If the site contains client-side gadgets like Java, Flash, etc. the site may be degrading your PC performance.
If the HTML sent to your browser doesn’t contain things that will phone home (meta refresh tags, Java applets, Javascript that might load another resource, etc.) then the server has no way of knowing whether you clicked off to some other site, closed the window, or left it sitting there on your desktop all day. HTTP (which is the protocol used to send/receive web pages) is connectionless so the browser makes a query, the server responds, and that’s the end of it.
One reason the webmaster might care is if you expect to be able to resume some session when you start back up on the site. Since HTTP is connectionless, we play tricks on the server to maintain sessions so the server can link multiple independent requests back to a single user. A good example of this is a shopping cart where the server needs some way to associate you with your cart every time you hit the site. This is often done with cookies, and the cookies may be set to expire after a certain time period. A timeout may be used for server performance if the server is maintaining a lot of in-memory data for open sessions or it may be a security thing if you want to protect against someone walking off and leaving a logged-in session. In any case, the site doesn’t know if you left a window open for hours on your browser, but if a session times out in the meantime and you start clicking again, the server may not properly identify you. The site designers care about this delay only in as much as they have to accommodate it so they don’t inconvenience their users.
This phrasing indicates you aren’t understanding this completely.
When you first access the page, the server downloads all the files (text, images, etc.)
that make up the webpage to your computer. Then the server disconnects from your computer, and waits for another request. All the time after that you are not sitting at the site, you are sitting viewing files that are locally in your machine. The website is not aware of what you are doing at your local machine, and doesn’t care.
You are thinking of the web like a phone system, with a continuous connection between you and that website. It’s more accurate to think of it as a fast mail system, where you are sending messages back and forth to each other – no continuous connection at all…
P.S. This is describing a basic webpage. Sites with streaming video, automatic refreshs, adware/spyware type webbugs, etc. work a bit differently. But still no continuous connection, just packets of data going back and forth.
micco, t-bonham, thank you. That answers what I wanted to know. You are right that I didn’t understand the nature of the connection. Your posts clear that up very well.