W3C home > Mailing lists > Public > www-html@w3.org > June 2006

Re: Identifying end tags

From: Laurens Holst <lholst@students.cs.uu.nl>
Date: Thu, 29 Jun 2006 03:05:32 +0200
Message-ID: <44A3275C.8060204@students.cs.uu.nl>
To: Shlomi Asaf <neoswf@gmail.com>
Cc: www-html@w3.org
Shlomi Asaf schreef:
> i shrank the page from 210Kb to 52Kb
> i don't use one single HTTP-Request. each HTTP-Request cost 0.2 
> seconds. i saved 7 HTTP-Requests, so i saved 1.4 seconds.

This is exactly my point. What kind of excessive cruft in class names 
and presentational elements were on that page that you could reduce the 
page to a quarter (!!!) of its original size? That means the styling 
hooks markup outweighed the content at least four-to-one, and probably more!

If instead sensible markup would have been used, the page weight could 
probably have been reduced even more. After all, the ‘class=""’ 
attribute already adds 8 bytes, and all the <div>s present only for 
presentational purposes even more.

So, do not use <div class="tree red font11px"><div 
class="something"><div class="something something"><span 
class="asdlj"><a>The text of the link</a></span></div></div><span 
class="qq"></span></div> but instead just <ul class="tree 
navigation"><li><a>The text of the link</a></li></ul>.

As you can see in that last example, changing the ‘navigation’ classname 
to ‘nav’ would have virtually no effect on the total page size. While in 
the first example, short class names do help a little, but making the 
markup more sensible would do so even more.

Rest assured that I know all about running sites in ‘real’ environments 
:). I do care about page weight, but shortening class names to the point 
that they become unreadable is clearly not the best means to achieve 
that goal.


~Grauw

-- 
Ushiko-san! Kimi wa doushite, Ushiko-san!!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Laurens Holst, student, university of Utrecht, the Netherlands.
Website: www.grauw.nl. Backbase employee; www.backbase.com.



Received on Thursday, 29 June 2006 01:05:44 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:06 GMT