Re: [ACFUG Discuss] (new topic) You CAN use CFINCLUDE to deliver css file, fewer http requests
H No. You're wrong. Seriously, not trying to tick you off, but you should read this: http://yuiblog.com/blog/2006/11/28/performance-research-part-1/ I fully understand and am aware of the separation of content and style, it's pro's and con's. How the browsers save things is not the issue, that's a minor point. The issue is with the http requests. It is not a direct 1 to 1 trade off of bandwidth vs. requests. To say that it is, is misleading. A 9k style sheet via an http request versus an additional 9k in your index file is not the same comparison. And since 50% or more of the US is on broadband anyway... it's a no brainer. The conclusion is the same: Reducing the number of HTTP requests has the biggest impact on reducing response time and is often the easiest performance improvement to make. In the next article we’ll look at the impact of caching, and some surprising real-world findings. _ Derrick Peavy Sales and Web Services CollegeClassifieds.com http://www.collegeclassifieds.com A Service of Universal Advertising, inc. ___ On Mar 17, 2007, at 8:07 AM, Howard Fore wrote: One of the nice things about using style sheets that are external to your HTML is that modern browsers won't download the CSS file again if the file hasn't changed. So you save n KB per HTTP request. Small, but it does add up, especially for a site with a lot of pages, like a shopping site. As far as solving the user's problem of the style sheet not coming along with a save, all the big broswer players have a save complete option that will grab the style sheet. On 3/16/07, Universal Advertising Derrick Peavy [EMAIL PROTECTED] wrote: Honestly, I didn't follow the cfinclude/css thread after the first post because it was not what I thought it was. However, if it is even remotely related, I thought the OP might want to know, you can use CFINCLUDE to deliver your CSS files in the head of your pages. I do this to reduce the number of http requests and also because it solves the problem of a user saving the page locally just as well as using a full qualified path in the head of the page would. -- Howard Fore, [EMAIL PROTECTED] In any moment of decision, the best thing you can do is the right thing, the next best thing is the wrong thing, and the worst thing you can do is nothing. - Theodore Roosevelt - To unsubscribe from this list, manage your profile @ http://www.acfug.org?fa=login.edituserform For more info, see http://www.acfug.org/mailinglists Archive @ http://www.mail-archive.com/discussion%40acfug.org/ List hosted by FusionLink -
Re: [ACFUG Discuss] (new topic) You CAN use CFINCLUDE to deliver css file, fewer http requests
Nope not ticked off. But first, how the browser saves files was your issue, not mine: it solves the problem of a user saving the page locally just as well as using a full qualified path in the head of the page would. The Yahoo study is a good read, though it seems a little duh to me. (No way! You mean a page that has less HTTP requests will be completed sooner?) Nah, I'm not wrong, we're talking about different issues, the Yahoo study is looking at the user side of the coin, in how long it takes the page to respond. In the page cited in the Yahoo study, only one of the 30 HTTP requests in that page view was a CSS file. The vast majority were images. And the second part of the Yahoo study showed that over a 2 week period, 20-40% of users would be using those out of a cache anyway. So I think more will be gained from simplifying page design than using inline CSS. The value of caching the CSS comes in when you pay the hosting service for bandwidth. Using inline CSS means that you will always transmit that CSS data, regardless of whether it could be used from the cache. Look at the data from a big name, high traffic CSS site: ESPN.com. ESPN's front page has two CSS files that are external to the HTML. Combined they are 33K. According to Mike Davidson, EPSN's associate art director at the time, they were getting 40 million page views a day. That means that by not including that 33K CSS as inline data, they have reduced their bandwidth by more than 1258 GB a day: 33K * 40 million pageviews. (The conversion to CSS from table based layout saved 2 terabytes a day, not that this is what we're talking about here!). Even though only 20-40% of users will have cached the CSS, that's still about 250-500 GB of bandwidth saved. By using inline CSS you would throw away 250-500 GB of bandwidth. Plus by using the inline CSS the only time you you are saving is the setup and tear-down time for that one HTTP connection, it will take the same amount of time to transmit the CSS data inline as in an external file. I don't disagree that the number of HTTP connections has an impact on the performance of a page (as does the type of connection the user has, their workstation, the quality of service used for that request, how much RAM the user has, the configuration of the web server, whether a clueless backhoe operator has severed the T3 connection to the Internet, and a variety of other things that we have no control over). I just don't think that the number of http connections in a page is the final measure of that page's performance. On 3/17/07, Universal Advertising Derrick Peavy [EMAIL PROTECTED] wrote: H No. You're wrong. Seriously, not trying to tick you off, but you should read this: http://yuiblog.com/blog/2006/11/28/performance-research-part-1/ I fully understand and am aware of the separation of content and style, it's pro's and con's. How the browsers save things is not the issue, that's a minor point. The issue is with the http requests. It is not a direct 1 to 1 trade off of bandwidth vs. requests. To say that it is, is misleading. A 9k style sheet via an http request versus an additional 9k in your index file is not the same comparison. And since 50% or more of the US is on broadband anyway... it's a no brainer. The conclusion is the same: Reducing the number of HTTP requests has the biggest impact on reducing response time and is often the easiest performance improvement to make. In the next article we'll look at the impact of caching, and some surprising real-world findings. _ Derrick Peavy Sales and Web Services CollegeClassifieds.com http://www.collegeclassifieds.com A Service of Universal Advertising, inc. ___ On Mar 17, 2007, at 8:07 AM, Howard Fore wrote: One of the nice things about using style sheets that are external to your HTML is that modern browsers won't download the CSS file again if the file hasn't changed. So you save n KB per HTTP request. Small, but it does add up, especially for a site with a lot of pages, like a shopping site. As far as solving the user's problem of the style sheet not coming along with a save, all the big broswer players have a save complete option that will grab the style sheet. On 3/16/07, Universal Advertising Derrick Peavy [EMAIL PROTECTED] wrote: Honestly, I didn't follow the cfinclude/css thread after the first post because it was not what I thought it was. However, if it is even remotely related, I thought the OP might want to know, you can use CFINCLUDE to deliver your CSS files in the head of your pages. I do this to reduce the number of http requests and also because it solves the problem of a user saving the page locally just as well as using a full qualified path in the head of the page would. -- Howard Fore, [EMAIL PROTECTED] In any moment of decision, the best thing you can do is the right thing, the next best thing is the wrong thing, and the worst thing you can do is nothing. -
Re: [ACFUG Discuss] (new topic) You CAN use CFINCLUDE to deliver css file, fewer http requests
On Mar 17, 2007, at 9:53 PM, Howard Fore wrote: I just don't think that the number of http connections in a page is the final measure of that page's performance. If it did, we'd never see anyone use AJAX. AJAX apps are noisy little things! Thanks for confirming my suspicions on this one Howard. -dhs Dean H. Saxe, CISSP, CEH [EMAIL PROTECTED] What is objectionable, what is dangerous about extremists is not that they are extreme, but that they are intolerant. -- Robert F. Kennedy, 1964 - To unsubscribe from this list, manage your profile @ http://www.acfug.org?fa=login.edituserform For more info, see http://www.acfug.org/mailinglists Archive @ http://www.mail-archive.com/discussion%40acfug.org/ List hosted by http://www.fusionlink.com -
Re: [ACFUG Discuss] (new topic) You CAN use CFINCLUDE to deliver css file, fewer http requests
Actually I think Ajax uses a different set of tubes. On 3/17/07, Dean H. Saxe [EMAIL PROTECTED] wrote: On Mar 17, 2007, at 9:53 PM, Howard Fore wrote: I just don't think that the number of http connections in a page is the final measure of that page's performance. If it did, we'd never see anyone use AJAX. AJAX apps are noisy little things! Thanks for confirming my suspicions on this one Howard. -dhs Dean H. Saxe, CISSP, CEH [EMAIL PROTECTED] What is objectionable, what is dangerous about extremists is not that they are extreme, but that they are intolerant. -- Robert F. Kennedy, 1964 - To unsubscribe from this list, manage your profile @ http://www.acfug.org?fa=login.edituserform For more info, see http://www.acfug.org/mailinglists Archive @ http://www.mail-archive.com/discussion%40acfug.org/ List hosted by http://www.fusionlink.com - -- Howard Fore, [EMAIL PROTECTED] In any moment of decision, the best thing you can do is the right thing, the next best thing is the wrong thing, and the worst thing you can do is nothing. - Theodore Roosevelt - To unsubscribe from this list, manage your profile @ http://www.acfug.org?fa=login.edituserform For more info, see http://www.acfug.org/mailinglists Archive @ http://www.mail-archive.com/discussion%40acfug.org/ List hosted by http://www.fusionlink.com -
