Forum Moderators: not2easy
Page Speed is a tool we've been using internally to improve the performance of our web pages -- it's a Firefox Add-on integrated with Firebug. When you run Page Speed, you get immediate suggestions on how you can change your web pages to improve their speed. .... It also identifies issues such as JavaScript and CSS loaded by your page that wasn't actually used to display the page, which can help reduce time your users spend waiting for the page to download and display.
the more tools the better..
loading it on to another machine ..and then we'll watch packet traffic .
<added>..interesting it wants to do a direct install from the page ..no download ,inspect and install later ..
so ..given that in the docs there is nothing to say how to uninstall it ..it's going on a machine that can be formatted and re 'dozed etc if it turns out to have chrome type habits ..I dont trust the plex .chrome is spy/malware ..
and yes I know that g said page speed is "open source" ..however I dont trust "open source" that tries to auto install direct from download ..from g pages
..once bitten etc
[edited by: Leosghost at 10:54 pm (utc) on June 4, 2009]
that bit I don't get unless the Add On itself doesn't get the correlation between CSS Specificity and HTML/CSS File sizes?
example warning
- Tag key with 2 descendant selectors and Class overly qualified with tag#content div.section p
the tag with 2 descendants is very likely specificity.. maybe they don't want to target every <p> in the page? - and as for a class being qualified by a tag, isn't that what classes are for .. so they can be re-used? If you can't re-use them with different tags (I do frequently) how do you qualify the difference?
It's a good job it's only a medium priority warning as I would hate to see the size of the HTML/stylesheets that add unique Classes and ID's to absolutely everything.. kind of defeats the purpose of keeping all that design and presentation separate. Classes do not have to be unique as long as they can be uniquely targeted via Descendant Selectors ..hmmm
.. the rest of the extension is, so far, looking good - a useful addition to Firebug, what do you think?
edit reason: can't spell descendant
[edited by: SuzyUK at 12:49 pm (utc) on June 5, 2009]
They e.g. promote classitis, unbelievable ... similarly asking to use javascript over :hover, ...
It does appear to be a tradeoff with what browsers can cope with and with no real proof yet either .. Steve Souders [stevesouders.com] has recently written of this and run some tests.. the results are interesting in that even he says it will only make a very small performance difference and that there will be no discernable difference for 70% of sites
His last message (June 1st) suggests they're running more complex tests to prove or disprove the theory
Interestingly Chrome 1 is the big outlier in the performance stakes between baseline and descendant but Chrome 2 is fine so perhaps they learned something along the way too and are trying to make life easier for their browser. They say in the original release post that they have been using this tool themselves internally for years, I presume Chrome 1 was their basis.. (though I know one should never presume ;))
I'm also thinking that their recommendations would make it easier for SE's to parse the CSS for their own reasons as well as being solely a browser performance helper.
In all this would mean adding loads of IDs (why bother with classes if everything is better unique :o) at coding time so think I'm going to watch for the tests
>>Page Speed Activity
me either, I'm going to install it to a separate FF instance as the documentation [code.google.com] says you are best to shut down FF and restart with just the single page you are testing
