1. 30
  1.  

  2. 4

    This is still relevant though, my school’s wifi is rather… mediocre at best and it’s hard to access much of the internet while in class, but lightweight pages are much easier to access.

    Edit: And I’m in Canada

    1. 3

      [2012]

      1. 2

        Coincidentally, thinking of this article, I went to look at some page weights yesterday. Almost everything I thought to look at (Twitter, YouTube, I forget) was about 2MB for the initial load. Outliers on the low end were HN (7k front page, 37k a comment page) and Wikipedia (half a meg for Life). On the high end I had Amazon (5MB for a page about a deep-cycle lead-acid battery) and either Twitter or YouTube once they were in use: about 12MB a minute for high-res YouTube video and maybe 2MB a minute for Twitter timeline updates.

        How much good could we do if we were using that bandwidth to eagerly replicate valuable data that needs to be archived, in order to prevent links from breaking?

        1. 1

          I can see an argument that weight doesn’t matter that much - a lot of the highest-traffic sites are bad offenders when it comes to keeping pages light.

          Reading the article it IS interesting how these optimizations can broaden your reach, but I get the feeling that bandwidth worldwide is generally increasing - even over cellular networks - so that aspect may become less relevant. And then again, as bandwidth increases it may be that we continue to increase bloat as well. I wonder if there’s any good indicator of average site weight over time?

          This reminds me of a debate I had with a colleague about whether or not to use jquery. My take was that it’s a premature optimization to start a green field project eschewing tools that will enable you to be more productive, even if it adds some bloat. Their take was that the user experience trumps any development cost incurred. In reality I think it’s a balancing act, and it’s hard to judge where the cost/benefit breakeven point is.

          Regardless of approach, it’s likely worth profiling sites you work on to see what low-hanging fruit is available. Browser developer tools have gotten REALLY good at profiling webpages now, so it’s really easy to take a look at. We can probably all do a little better.