Home > Computers, Work > Reducing My Bandwidth

Reducing My Bandwidth

I’m facing a rather practical problem at work, and it’s kind of highlighted to me one potential downside of AJAX based websites and Web 2.0.

I was recently shown a chart of bandwidth usage and number of web hits for people in my department at work. This made me immediately think that I should start to reduce my non-work web traffic to avoid showing up as a major blip on the radar.

One simple way I realised that I could do this was to reduce the amount of RSS feeds that I unnecessarily subscribed to, and to reduce the number of times I refreshed the feeds.

Then I realised, that every time I displayed a feed, I would cause my hits to go up. If looked at a feed in my RSS reader (the “Sage” plugin for Firefox,) I would cause an HTTP request for the page, and a request for every image. If I went away from the feed and then back to it, then I’d refresh everything again. If there are 100 RSS items in the feed, each with (say) 3 pictures in, that’s a lot of web requests. Couple to that, the “Read” status was based off of your browsing history, I have my suspicions that every time I marked a post read (even though I didn’t read it,) it caused a web request as well.

Since upgrading to Firefox 3.5, Sage has unfortunately stopped working, so instead I switched to Google Reader, which I was hoping should off load some of my hits to the Google servers and instead I would just see the changes.

Then I realised that all of these wonderful AJAX websites that I use are going to be doing the same damn thing. Every time an AJAX request is sent, that’s another hit. Thinking of GMail, that sends an awful lot of requests. Google Reader seems to be even worse – every time I scroll past a post I don’t want, it gets marked as read – that’s an AJAX request. It refreshes the posts list a lot too, that’s more requests. I’m now wondering too, whether the RSS feeds aren’t just parsed locally on your PC, and you’re still doing the same number of requests.

Unfortunately, I can’t find any technical reference for Google Reader to explain how many requests are made, and from which location (my PC or Google Server.) I guess Google aren’t really ready to give out the information to some of their architecture.

Even now, typing this blog entry, WordPress is continually using AJAX to save a draft version and to provide a word count.

It seems that the one downside to the Web 2.0 world is that if you’re being monitored in terms of bandwidth usage or number of web requests, AJAX (and the like) are most definitely not your friends.

For now, if anyone can point me in the direction of an RSS reader that definitely does cache images (and other embedded items,) and does not make web requests to mark something as read (and is also “trustworthy” in terms of security and preferably on some Free license) that would be helpful

Categories: Computers, Work Tags:
  1. July 8th, 2009 at 09:53 | #1

    Using FireBug you can see that Google Reader makes requests for all RSS feeds on the google.com domain. These are set to expire immediately (although you can go to off line with Google Gears enabled).

    Requests for images however are made directly to the image URL in the RSS feed. This is normally the same domain as the RSS feed, and will be cached if that web server is sending any max-age or Expires headers.

    I don’t know of an RSS reader that meets your needs. You could write one 😉

  1. No trackbacks yet.