Sign up FAST! Login

MIT boosts webpage loading speeds by 34% with new Polaris system

In the days of dial-up internet access, web pages did not load instantly — not even close. How quickly we forget. Now, most people will simply move on if a site doesn’t load in a few seconds. There are probably other pages that willload instantly, so why bother? Researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) are trying to solve the problem of slow load times with a new system that is 34% faster than traditional methods.

When your browser loads a page, it has to reach out to a server someplace in the world and download objects like JavaScript code and HTML files. Often, when that object is integrated into the page, it calls upon other objects to be downloaded. These “dependencies” can slow down the rendering process because the browser only finds all the files as it goes, and it’s best to minimize the number of trips across the internet tubes. On a complex page, this process can get very sluggish with thousands of dependencies.


Stashed in: MIT TR, Microentertainment, The Web

To save this post, select a stash from drop-down menu or type in a new one:

And will better efficiency mean that more work should get done?

Yes. Note that lots of people are trying to speed up the web these days, notably Google AMP:

I was thinking more in terms of what output will be expected from Cubicle Person.

Oh. Well it still takes the same amount of time to read.

Reducing the load times is small compared with the amount of time to read.

... and now cue in the uptick of Reading Retention programs.

So that people can improve the speed of their reading?

So that people can remember what they've read. (That's the hardest for me — half my focus goes towards "I must remember this", and not to what I'm actually reading.)

Reading is easy if you don't have to comprehend and retain.

Retaining is much much harder.

Thank you.

Happily, Marlene!

You May Also Like: