Slashdot Mirror
Testing a Pre-Release, Parallel Firefox
Firefox, in its official version, still lacks support for multi-threading (running on different processors), though Chrome and Internet Explorer 8 both have this feature. A Firefox project called Electrolysis is underway to close this gap. A blog author tested a pre-release version of Firefox that loads different tabs in parallel, and he chronicles his findings, including a huge speedup in Javascript vs. Firefox version 3.5 (though the pre-release still lags Chrome in many of the tests).
For the hungry CompSci master's thesis and for the fidgeting moron who can't wait an additional 100ms for the page load.
Ahh, progress.
suck a dick you cock-loving fucker
NIGGERS. Galactus is the ultimate Space Nigger.
He makes the ultimate white man, the Silver Surfer, his bitch.
You are making the point that being able to run Javascript more efficiently is good. I would argue that any external program being able to run another program (i.e. executing Javascript) on ones computer is fundamentally evil and bad. It is MY computer! No external sources of programs should be able to use it for computing unless I have explicitly approved them. To do otherwise is "theft" (of my CPU resources). Failure to recognize this on the part of browser manufacturers/distributors contributes to the problems in which we currently find ourselves enmeshed. I don't want to run Javascript except under very explicit circumstances. I did not startup my browser to run other people's programs. Every single browser should on every single page that requests the ability to run Javascript -- i.e. say "May I?" -- until that point is reached any advancements in Javascript speed are completely useless to me -- because I view it running Javascript as anti-green, anti-security and anti-moving the world in a forward direction. HTML was laid out as a language to display information -- it was not laid out as a language (once enabled by Javascript) to allow advertisers (or others) to manipulate ones personal computer. One of the biggest mis-steps in the history of HTML/web development was Netscape's development of the ability of the server's to control the browser's through Javascript. That destroyed the fundamental principles of the web (your machine is your machine and I am simply providing you with information, which BTW works as a model for Google (with whom I have no connection with) but simply points out how they got it right the first time around, and we are going to have to work very very hard to recover those principles). Otherwise we are going to have to put up with a web where suppliers are always sticking their noses into our business in order to sell us something. Which if they did it with a passive browser would be fine (IMO), but once they do it with an active browser (i.e. Javascript enabled) they are a friggen pain in the rear end.
So Firefox, get the message. There are lots of us who DO NOT WANT friggen faster Javascript, in fact we would prefer, if like chromium, there were a startup option to completely disable it! (Though I can now effectively start Firefox w/o Javascript by "fudging" the preferences file. In which case the battle is forwarded to the vendors that insist that one "must" have Javascript for their sites to work -- go shove one's Javascript in a dark place, if you have designed your web site so that it only works with Javascript enabled, which was NOT the way the web was envisioned, and being drawn and quartered would probably be too good for you, I'm debating in my mind the difference between being keel-hauled and impaled...).
The bottom line is that browser vendors focused on Javascript performance are next-to-useless. The real test IMO is vendors who replace the default gtk poll function with something more intelligent that reduces power consumption (more green) while maintaining performance (but both Firefox and Chromium have yet to do that -- in spite of the fact, that at bug reports have been filed, at least with Firefox)) -- so the user base awaits... [1]
1. It would appear that neither the Firefox nor the Chromium developers have bothered to strace their primary processes (the ones which accumulate the CPU time) and recognized that there are a significant number of poll() and/or gettimeofday() calls which seem to by and large DO NOTHING). Any programmer concerned with efficiency would suggest there is either a flaw in the process model or a flaw in the implementation. Sad, that even in this modern era (where if the ice caps melt and you are living in a coastal area you are going to need to relocate), consideration of CPU use under Linux is going so disregarded.