Which brings me to IE7.  At some point, qrisper got totally broken on IE7.  Ended up using the * hack in my CSS files, which now won’t validate.  To that I say fuggem.  Rules are meant to be broken.

The development theme for the past week was spring cleaning.  I consolidated a number of pages, modularized my CSS (as much as I could), cleaned up and moved all my js files to the footer and tidied up all of my control/view files after the redesign.  qrisper just feels lighter to me now, although there wasn’t any noticeable improvement in performance.

A couple of new/updated pages:

• created a personalized 404 page…heard it’s a good thing to have
• redid the splash page screenshots after the layout redesign…but need to redo them yet again
• added Disqus comments to all answer pages.  I had comments in the pre-alpha version of qrisper but took them out because of poor site.  Still kinda slow but the ability to sign in via Facebook Connect was irresistable.  Too bad that ain’t working…can’t figure out what I’m doing wrong
• Tweaked my existing autocomplete functionality…basically made it look nicer.  Only thing missing now is to add keyboard navigation functionality.

Next up on my list is security and validation.  I’m already using sanitize classes to filter out all user input but I want to add additional layers of security.  Also want to incorporate jquery into the validation process so that users have a clear sense of what’s going on.

Removing the ads gave me some room to play around with.  So I changed the content and layout a bit, primarily by removing the leaderboard section.  I also gave search its own tab and removed the searchbar from the page.  And I moved the recent answers into the search tab and added categories to the left column.  After chewing on it for a bit, I think I’m going to bring back the search bar…just need to figure out where to put it.

Had a couple of duh moments.  I was maintaining duplicate files with slightly different content for logged in/not logged in users when I realized that a couple of PHP if statements could swap out the content using the session id.  Yeah, duh.  Doesn’t end there folks!

I was using Simplepie to pull the latest q&a’s from my database.  But then I realized, why the heck am I not pulling the data directly from my database instead of using my own RSS feed?  Umm, beats the heck out of me.  Anyway, changed that.  Progress!  Slow but tangible.

I didn’t forget about the non-existent logged in users out there!  The right column will become the exclusive domain for the logged in user.  Added some stats (questions answered, skipped, remaining, etc) so that users can keep track of their progress.  Gonna be adding a lot more exciting stuff so stay tuned!

Furthermore, the main questions page was modified to  focus is more on the users answers.  I came to the conclusion that I need build two very different qrispers – one for folks seeking answers, another for folks seeking questions.  And with that, qrisper just became a lot more complex.

I’ve been meaning to write a few thought pieces on where qrisper is headed.  Read a couple of interesting blog posts that I plan to respond to.  Hopefully some time next week.

Decided to take a breather on the php to focus on the interface.  Actually, the php issue was interface related so I guess what I really mean is focus on making qrisper a bit flashier looking.

The landing page was getting a little noisy so I decided to put the registration into a modal window. I used jquery and followed a tutorial from yensdesign.com.  Implementation was pretty straightforward. One thing I want to tinker with is the location of the window.  It looks a bit off-center…I noticed that on the tutorial as well.  If anyone can see the reason why, please holler (code below).

function centerPopup(){

        var windowWidth = document.documentElement.clientWidth;
	var windowHeight = document.documentElement.clientHeight;
	var popupHeight = $j("#popupSignup").height();
	var popupWidth = $j("#popupSignup").width();

	$j("#popupSignup").css({
		"position": "absolute",
		"top": windowHeight/2-popupHeight/2,
		"left": windowWidth/2-popupWidth/2
	});
}

I’m using jquery with prototype so I needed to make sure that I included jquery’s noconflict clause and change the $ variable for all jquery scripts (hence the $j).  There’s so much information out there about jquery that it’s hard not to use it.  But now that I have the power, I need to be responsible and practice restraint.  qrisper’s getting a little large around the waist from overindulging on all of those image files. Found this site on web optimization, telling me that qrisper needs to go on a  diet.

Also added a little feedback button…meh…looks alright.  The button made me realize that I’m digging myself into a trap.  I am maintaining slightly different versions of the same page.  Which version the users see depends on whether or not they’re logged in.  Decided that I could probably use a php if statement to swap out includes depending on whether or not a session id exists.

Wanting to trim the fat further, I went to clean up my css files and optimize them with a few new tricks I learned.  Big mistake that was.  Spent a few hours fixing all of the damage I caused.  I found the tips from David’s old blog.  Actually found a number of good resources…a few of which I used to fix a few issue that had been bugging me.

In the meantime, I’ve been taking care of a couple of other things:

• created the first drafts for the About Us page and the Sitemap (plain HTML, not XML)
• spiffed up the images in the What? tab…I might redo them still
• changed the content in the Skipped tab.  Instead of a form listing all of the questions that users skipped, it simply becomes another pool of questions that users can choose to answer or skip again.
• made enough changes to score a B on YSlow.  But feels like the sites actually slower…
• added the Digg this and Yahoo buzz icons to all results pages.  But I added them only to the non-logged in pages.  I’m sure this will need to change.

I’ve been playing with affiliate programs and adding a few ads here and there.  Figured out a strategy to diversify my ads portfolio.  Now all I need are users to view and click on those suckers.