Aaron Heckmann

Javascript RegExp

2010-10-12T17:02:00.001-04:00

What are Programs?

2010-10-08T21:15:00.006-04:00

Some programs are community centers, and you want to entice people to
come inside.

Some programs are factories, and you want to make sure the workers
don't hurt themselves.

Some programs are shops, and you want people to visit, but only a few
go in the back.

And then some programs are art, and while you might hope that some
people find them beautiful or perhaps even useful, and contributions
may be accepted and feedback always welcome, the real purpose is just
to hang it in your house and let it bring you joy.

There is not just one kind of thing in this world.

--i

Micro-optimization: Javascript Recursion vs. Loops

2010-01-23T15:05:00.009-05:00

I was just reading Juriy Zaytsev's excellent post about the delete operator and got to this statement:

"Even if function is calling itself recursively, a new execution context is being entered with every invocation."

This surprises me. I figured that javascript engines were able to detect recursion and reuse the same execution context somehow. So is recursion in javascript even more expensive than I thought? I knew that calling a function on each iteration of a loop was slow (compared to straight up loops) but how does recursion compare? To find out I set up a quick experiment to benchmark three different options:

Recursion

A function called on each iteration of a loop (function loop)

A vanilla loop (no function calls)

Each benchmark will run a maximum of 490 iterations per test with each test running 400 times per page load. The test page will load 50 times in each browser. A maximum of 490 iterations is used because Safari 3.2 begins throwing "too much recursion" exceptions at 500 and there are a few overhead calls for each test. Here are is the code I used:

Recursion

function () {
    var i = 0,
        rec = function () {
            i++;
            if ( i < 490 ) rec();
        }
    ;      
    rec();
}

Function loop

function () {
    var i = 0,
        fn = function () {
            i++;                
        }
    ;

    for (; i < 490; ) fn();
}

Vanilla loop

function () {
    var i = 0;
    for (; i < 490; ) {
        i++;
    }
}

What were the results? Here's the breakdown by browser type. Green background means it's the fastest loop method when functions are involved. Bold text means it's the fasted loop method overall.

Browser	recursion avg	fn loop avg	loop avg
Firefox 3.5.7 Mac OSX 10.5.8	92.440	200.400	1.240
Safari 3.2 Mac OSX	70.620	39.980	7.660
Safari 4 Mac OSX 10.5.8	6.500	4.420	0.600
Firefox 3.5.7 WinXP	76.680	179.940	2.580
Firefox 3.0.7 WinXP	63.180	43.020	6.460
Firefox 2 WinXP	861.340	775.600	14.300
Chrome 4.0.288.1 WinXP	1.720	1.720	0.140
Safari 4 WinXP	4.780	3.780	1.540
Internet Explorer8 WinXP	137.880	66.260	3.800
Internet Explorer7 WinXP	418.980	388.180	18.380
Internet Explorer6 WinXP	512.260	479.860	16.640
Opera 10.10 WinXP	22.860	23.400	8.460
Safari 3.2.3 WinXP	93.020	52.220	12.900
Internet Explorer8 Vista	192.760	124.700	9.360
Total	2555.020	2383.480	104.060

Function loops were faster overall when nothing was happening on each iteration. Let's test again with a little more going on; hopefully somewhat closer to a real world application. The code:

Recursion

function () {
    var i = 0,
        rec = function () {
            var c = i + "a" + (Math.random() / i);
            i++;
            if ( i < 490 ) rec();
        }
    ;      
    rec();
}

Function loop

function () {
    var i = 0,
        fn = function () {
            var c = i + "a" + (Math.random() / i);
            i++;                
        }
    ;

    for (; i < 490; ) fn();
}

Vanilla loop

function () {
    var i = 0;
    for (; i < 490; ) {
        var c = i + "a" + (Math.random() / i);
        i++;
    }
}

Browser	recursion avg	fn loop avg	loop avg
Firefox 3.5.7 Mac OSX 10.5.8	1030.120	1542.200	1174.200
Safari 3.2 Mac OSX	854.480	764.040	705.580
Safari 4 Mac OSX 10.5.8	523.380	519.420	534.000
Firefox 3.5.7 WinXP	750.020	1125.260	712.780
Firefox 3.0.7 WinXP	776.860	787.260	1270.620
Firefox 2 WinXP	2458.480	2726.200	1506.620
Chrome 4.0.288.1 WinXP	436.360	448.200	451.060
Safari 4 WinXP	370.180	367.380	365.240
Internet Explorer8 WinXP	691.160	583.160	450.840
Internet Explorer7 WinXP	1753.160	1711.140	1153.140
Internet Explorer6 WinXP	1357.240	1289.140	841.480
Opera 10.10 WinXP	682.420	656.040	626.000
Safari 3.2.3 WinXP	897.320	846.180	752.440
Internet Explorer8 Vista	723.300	618.580	481.060
Total	13304.48	13984.199	11025.06

Keep in mind these tests did not test looping speeds only, they were effectively testing each browsers Math.random(), string concatenation speeds, etc.

So what does this tell us?

The results aren't very clear in terms of execution times. Strict tests with nothing going on in each iteration tended to lean toward function loops being faster than recursion. Tests with more going on generally leaned toward recursion being faster in browsers that have tracing techniques enabled. Firefox optimized more for recursion than in other browsers. In Internet Explorer and Safari function loops were always fastest. Opera and Chrome were pretty much a wash.

If we think of "fastest" not in terms of overall execution time but also in terms of how many actual users would benefit from either technique, then function looping is the winner since IE still has the most market share.

Conclusion

The take away on this is that it doesn't matter if you use recursion or function loops, if you really want to be fast, inline your code and refrain from any function execution at all.

Writing a jQuery plugin manager - Part 3 - Fully resolving URIs cross browser

2010-01-15T18:13:00.005-05:00

I recently started thinking about a plugin manager for jQuery, something that could integrate with a server side file rollup service like the YUI Configurator. I looked around a bit and didn't see anything quite like I wanted so I decided to roll my own.

Surprises

There were a few surprises along the way that I wanted to document, because I don't want to forget them and hopefully you'll find them helpful too.

Detecting when stylesheets are loaded

Detecting when downloaded javascript was parsed by the browser

Fully resolving URIs cross-browser

This post will talk about fully resolving URIs cross-browser.

Overview

If you've ever spent time reading URIs from anchors, scripts, or links, you probably know what this post is about. Reading these attributes cross-browser isn't straight-forward. For example, let's start with the following script tag:

<script src="/myscript.js" id="myscript"></script>

Now let's read the src attribute, simple right?

var script = document.getElementById("myscript");
alert(script.src);

In Firefox, Safari, Chrome, and Opera we see http://absolute/path/to/myscript.js. But in Internet Explorer we see myscript.js. So how do we go about absolutely resolving URIs cross-browser?

The Solution

It turns out that if we first create an anchor element using .innerHTML and subsequently read it's href attribute, it will be absolutely resolved.

var resolve = (function () {
    var div = document.createElement("div");
    return function (URI) {
        div.innerHTML = "<a href='" + URI + "'></a>";
        return div.firstChild.href;
    }
})();

Now let's adjust our original example to use our new resolve function and we'll be all set.

var script = document.getElementById("myscript");
alert( resolve(script.src) );

Conclusion

Hopefully this was somewhat helpful, and if not, thanks for reading anyway.

If you're interested in using the jQuery plugin manager in your own project or are just curious, there is a demo available here and the source code is up on Github.

Writing a jQuery plugin manager - Part 2 - Detecting when downloaded javascript was parsed by the browser

2010-01-09T13:35:00.007-05:00

Surprises

There were a few surprises along the way that I wanted to document, because I don't want to forget them and hopefully you'll find them helpful too.

Detecting when stylesheets are loaded

Detecting when downloaded javascript was parsed by the browser

Fully resolving URIs cross-browser

This post will talk about detecting when downloaded javascript was parsed by the browser.

Getting Started

You may be wondering why I bother writing a post about detecting when downloaded javascript has been parsed by the browser. Well, the reason is that in my testing I found that Internet Explorer rarely threw exceptions when using globals that should have been available when my success callback fired. So I came up with a work around that continually tests whether all of the properties specified are available in the DOM.


var _providesAvailable = function (provides) {
    var available = true,
        i = 0, 
        len = provides.length,
        names,
        nLen,
        ns,
        j
    ;

    while ( available && i < len ) {
        names = provides[i].split('.');
        nLen = names.length;
        ns = window;
        j = 0;

        for (; j < nLen; j++) { 
            if ( !(names[j] in ns) ) {
                available = false;
                break;
            }
            ns = ns[names[j]];
        } 
        i++;
    }
    return available;
};

var arrayOfExpectedProps =  [
        'foo.bar', 
        'something.completely.different'
    ],
    cb = function () {
        if (!_providesAvailable(arrayOfExpectedProps)) {
            return setTimeout(cb, 30);
        }
        console.log("done");
    }
;

cb();

While not perfect (if I pass in anything other than an array to _providesAvailable it will fail) it's fine for what I'm doing. I've since switched to using jQuery's ajax method and haven't had any additional problems with this bug but I still left the _providesAvailable method enabled for now just to be safe.

Plugin Registration

With a way to now detect when global properties are in the DOM, we can specify which properties that must be available before our success callback is fired for any of our registered plugins:


jQuery.use.add({
    name: 'jquery-myplugin',
    file: ['someFile.js', 'somestyles.css'],
    provides: 'jQuery.myPlugin.sweetness'
});

jQuery.use('jquery-myplugin', function ($) {
    // window.jQuery.myPlugin.sweetness is now available
});

In the above example we use the jQuery.use.add method to register our plugin. name determines which unique key you'll use to pull in all of the assets you specify, file determines which file[s] to request, and provides determines the global properties that need to be available in the DOM before firing your callback. For more information about what options are available, see the project on Github.

Conclusion

We've seen why we need a method to detect when dynamically created scripts are appended to the DOM and how to go about detecting when they're truly ready for use. Hopefully this was somewhat helpful, and if not, thanks for reading anyway.

If you're interested in using the jQuery plugin manager in your own project or are just curious, there is a demo available here and the source code is up on Github. Have fun!

What's Next

In my next post I'll cover fully resolving URIs cross-browser.

Writing a jQuery plugin manager - Part 1 - Detecting when stylesheets are loaded

2010-01-06T09:45:00.023-05:00

Surprises

There were a few surprises along the way that I wanted to document, because I don't want to forget them and hopefully you'll find them helpful too.

Detecting when stylesheets are loaded

Detecting when downloaded javascript was parsed by the browser

Fully resolving URIs cross-browser

This post will talk about detecting when stylesheets are loaded. See a demo or fork the Gist on Github.

Getting Started

In order to get things moving we'll need a link tag pointed at some css:


var css = document.createElement('link');
css.type = 'text/css';
css.rel = 'stylesheet';
css.href = "http://groups.google.com/groups/style.css?ig=1&stock=1&av=4&hl=en&v=639";

// I chose a different domain here because I want to 
// support loading stylesheets from any environment.

Implementation

Now on to the good stuff. Surprise! In Internet Explorer and Opera life is easy. You can detect when the stylesheet loads by attaching an event listener directly to the element and listening for it's load event. This even works cross-domain.


jQuery(css).bind('load', function () {
    // this stylesheet loaded!
});

But this doesn't cut it for other browsers since they don't implement the load event on stylesheets. Really, they don't. So I took a trip down a path of possible workarounds.

In Safari and Chrome we can detect when the stylesheet is loaded as soon as it's sheet.cssRules properties are available.


(function sheetLoaded () {
    try {
        css.sheet.cssRules;
   
    } catch (e) {
        setTimeout(sheetLoaded, 50);
        return;
   
    }
    // if we got here the stylesheet is loaded
})();

So far so good. This even works on stylesheets loaded from other domains. But what about Firefox? Unfortunately, I could only get Firefox to cooperate with this technique when loading stylesheets from the same domain. If anyone knows a work around feel free to share it in the comments below or grab the Gist and fork away. For now, I've left detecting when stylesheets are loaded out of my plugin manager since I want to provide consistent behavior across all supported browsers.

Update

Julian Aubourg has been working on rewriting the jQuery ajax logic and discovered a solution to detecting when stylesheets load in Firefox.


(function sheetLoaded () {
    try {
        css.sheet.cssRules;
   
    } catch (e) {

        if ( "NS_ERROR_DOM_SECURITY_ERR" != e.name ) { 
            setTimeout(sheetLoaded, 50);
            return;
      
        } 
    }

    // if we got here the stylesheet is loaded

})();

Here's how it works. Gecko (Firefox's layout engine) throws an NS_ERROR_DOM_INVALID_ACCESS_ERR exception if you try to access the cssRules property while the link element is loading. Once the stylesheet loads it either stops throwing exceptions (in the case of same domain requests) or throws a different NS_ERROR_DOM_SECURITY_ERR exception. By watching for the security exception we can safely determine when the stylesheet is loaded.

I updated the Gist and plan on implementing his great technique soon.

Tested Browsers

Safari 4 Mac

Firefox 3.5.7 Mac

Chrome 3.0.195.33 WinXP

Firefox 3.5.1 WinXP

Opera 9.64 WinXP

Opera 10.10 WinXP

IE 6-8 WinXP

Safari 4 Vista

Firefox 3.5.2 Vista

IE 7, 8 Vista

What's Next

In my next post I'll cover detecting when downloaded javascript is parsed by the browser.

Auto-silence Mobile Devices?

2010-01-05T11:35:00.000-05:00

Maybe this is a bit too Big Brother but I think that places like libraries and movie theaters should broadcast a signal, that when detected by a mobile device, the device switches automatically to silent mode.

This could work for airlines too. An airline could broadcast a signal that forces devices to either shut down or switch to an acceptable mode during critical times like take-off and landing.

Too big brother? Maybe. But providing greater context-sensitivity to our mobile world has huge potential.

Save MySQL

2009-12-13T07:05:00.001-05:00

http://monty-says.blogspot.com/2009/12/help-saving-mysql.html

Legislate the poor into prosperity

2009-07-24T07:25:00.004-04:00

"You cannot legislate the poor into prosperity by legislating the wealthy out of prosperity. What one person receives without working for, another person must work for without receiving. The government cannot give to anybody anything that the government does not first take from somebody else. When half of the people get the idea that they do not have to work because the other half is going to take care of them, and when the other half gets the idea that it does no good to work because somebody else is going to get what they work for, that my dear friend, is the beginning of the end of any nation. You cannot multiply wealth by dividing it."

Adrian Rogers, 1931

Internet Explorer 8 hashchange slowness

2009-05-04T20:43:00.006-04:00

Is it just me or does IE8 perform terribly when trying to set window.location.hash to anything?

I need to do some more testing but so far it varies between 500 - 3500ms just to update it. What's up with that? I actually disabled my history management completely in IE8.

window.location.hash = '#somethingElse';

I am using http://www.asual.com/jquery/address/.

I'll post again if I find any work arounds. If you find anything please post them here too.

Update:

This seems to have been caused by some plugin or something on my windows box. I have not been able to recreate this issue testing on others.

Javascript Performance

2008-09-02T20:46:00.005-04:00

If you're interested in web browsers and javascript you're probably aware of both Google Chrome and Firefox 3 beta 1 (tracing enabled). I won't go into the details about them except for what results I found using Dromaeo by John Resig , an excellent tool that performs javascript performance testing.

The results are truly amazing, Chrome absolutely stomps on Firefox 3 beta 1 in just about every catagory in the test...

FF3beta1(tracing) VS Chrome:
http://dromaeo.com/?id=25058,25118
Note: Chrome shows up as Safari/525.13 in dromaeo

VS todays latest browsers:
http://dromaeo.com/?id=237,24882,25206,232,25058,25118
Note: Chrome shows up as Safair/525.13 - the rightmost column

Note: tests perfomed on Windows XP

IE8 b2 - CSS Expressions

2008-08-27T16:51:00.002-04:00

"CSS Expressions are no longer supported in Internet Explorer 8 Standards Mode

Also known as 'Dynamic Properties', CSS expressions are a proprietary extension to CSS with a high performance cost. As of Internet Explorer 8 Beta 2, CSS expressions are not supported in IE8 standards mode. They are still supported in IE7 Strict mode and Quirks mode for backward compatibility."

http://www.microsoft.com/windows/internet-explorer/beta/default.aspx

Cross Browser Testing

2008-08-27T09:35:00.003-04:00

I came across a site the other day called CrossBrowserTesting.com. It provides free access (limited to five minute intervals) to several different operating systems and browser configurations. I kind of like it. It's like a virtual running in your browser. Once your OS launches, you just open the browser you want to test in and type in your public url. That's it! You even get to test out your page functionality since it really is a running instance of the browser and not just a screen shot.

Here are their operating systems:

Windows 98
Windows XP
Windows Vista
Ubuntu Linux
Apple OS X

and the browsers:

Firefox
Internet Explorer
Mozilla
Netscape
Opera
Safari
AOL

Paid priority access to the virtuals is also available. Check out their demo.

Fire Eagle Walk-Through Trouble

2008-08-13T13:15:00.004-04:00

So today I was going through the walk-through for Yahoo's Fire Eagle. If you don't know, it's a way to share your location with others people / apps. It seemed pretty cool so I thought I'd try it out. If you're not on Windows XP and running Apache then you'll probably not be interested in this post. If you are and intend on going through the PHP Fire Eagle walk-through then this may help you get through it.

The walk-through states that you will need a copy of "curl-ca-bundle.crt" and add it into fireeagle.php. However, the link provided in the walk through did not provide the .crt file required by Fire Eagle to do any SSL curl calls. Apparently it used to. So that left me searching on Yahoo for any other posts from anyone in the same situation. I couldn't find the file anywhere so I ended up adding this to fireeagle.php:


   curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, FALSE);

I added it after this line:


   curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);

See here for more details on this option. I didn't have any trouble with the rest of the walk-through. Hopefully neither will you.

Overall Fire Eagle is a pretty cool idea, I encourage you to check it out.

Nintendo Wii Points Order Form Nightmare

2008-08-07T20:59:00.006-04:00

So tonight I wanted to buy a new Nintendo Wii Ware game but I needed to buy some Wii Points first (in case you don't know it's fake money that you pay for which can only be used to buy Wii stuff). So I got online with my Wii, got into the place to buy points, filled in my credit card type, card number, experation date, billing address, and number of points - only to receive an error message informing me that my address could not be validated (invalid zip). So I clicked OK, fully expecting to return to the form with all of my data pre-filled, so I could simply adjust the zip code. But guess what I got instead? I was kicked all the way out and had to start over from the beginning - I even had to enter the number of points I wanted to purchase again! I couldn't believe it.

I guess I gave Nintendo to much credit. I like the interface of the Wii in general. I like the interface on nintendo.com. They just totally failed at their ordering experience. It's as if a completely seperate group of evil ninjas broke into Nintendo and implemented it in the middle of the night. Oh well. Nobody's perfect, we all make mistakes. Just remind me next time I want to buy some Wii points to go to the store instead.

Cost of Living

2008-08-07T15:33:00.003-04:00

I recently turned down a dream job offer from Fox Interactive Media in Los Angeles that would have paid 120K a year. Some of you may be scratching your head thinking "Idiot! What were you thinking?!". Well let me tell you that it wasn't an easy decision and there were a few not so obvious factors involved.

#1) Cost-of-living
A quick search for "cost of living" will return a few online calculators that you can use to compare Charlotte, NC with LA. You will quickly see the huge difference, most noticeably housing at 220.8% higher. After further research it seemed to me that to live in a relatively safe area and have a house similar to my current one would cost me about $900,000! Obviously that's just ridiculous! I have a family and really did not want to alter our lifestyle at all - which brings me to my next point...

#2) Family
I am unwilling to change my family's lifestyle and quality of life just to go work my dream job somewhere, anywhere for that matter. I can provide much, much more for my children here in Charlotte than I ever would in LA: a big house with a big yard in a quiet neighborhood with low crime. Also, my wife would need to stop working too which would have thrown in another financial curve.

120K in LA doesn't get you very far compared to Charlotte. In fact, after my research I'm quite surprised not to hear of more people bailing on California in general. And I didn't even mention earthquakes, fires, or traffic yet! These days after I drive my 10 minutes home from work I have plenty of time to hang out with my family, every now and then stopping to think... I've got it made.

Well, almost! I'd still love to be working a fulltime hardcore object oriented javascript/web 2.0 type job but maybe one will turn up around here sometime soon.

The Web Design Survey

2008-08-05T11:19:00.000-04:00

Tricky Code

2008-08-04T08:45:00.003-04:00

So here's what I think is some interesting but tricky javascript:


   window.prop || (window.prop = 'value');

If window.prop does not yet exist it evaluates to false and the interpreter continues on to the right side of the OR operator, which then assigns a "value" to it. A very terse way of ensuring a variable has a value assigned to it.

I often use a similar pattern when assigning default values during the initialization phase of object instantiation:


   var MyObj = function (property) {
       this.property = property || 'defaultValue';
   }

Here I think this code is obvious but I'm not convinced that the tricky code in the first example is obvious enough to use on it's own. It's really only slightly more concise than using the ternary operator:


   window.prop = window.prop ? window.prop : 'value';

I find this much easier to understand and should be much more familiar to non-javascripters due to it's availability in many other languages as well.

So while I think it's cool code, I don't think I 'll use it.

New Look

2008-08-04T08:43:00.003-04:00

I decided that it was easier to read black text on a white background than vice versa - so here's the new look. Hope you like it.

js Observer updated

2008-07-21T12:54:00.010-04:00

I've been using the Observer class quite a bit and come to realize that I really wanted the publish() method to return boolean. That way if any subscriber returns false I'll know about it. Example:

var widget = new Widget();
widget.onSomethingHappened = new Observer();
widget.onBeforeSomethingHappened = new Observer();


widget.onBeforeSomethingHappened.subscribe(function(){
return false; // new
});

widget.onSomethingHappened.subscribe(function(){
    alert('Something just happened!');
});

if ( ! widget.onBeforeSomethingHappened.publish() ) {
    // the following will not published b/c
    // onBeforeSomethingHappened returned false
    widget.onSomethingHappened.publish();
}

Works for me. Thoughts?

js Observer

2008-07-15T15:41:00.013-04:00

I don't think this is anything new but I thought I'd share it with you. While working on some client-side widgets I realized the Observer pattern would come in handy. So I put a little script together for it that you can use. Example:


var widget = new Widget();
widget.onSomethingHappened = new Observer();

function alertTheWorld () {
      alert('Hello World!');
}

widget.onSomethingHappened.subscribe(alertTheWorld);

widget.onSomethingHappened.publish(); ==> 'Hello World!'

Besides passing in a function to the subscribe method, you may also pass in a context object as the second argument: e.g. an object in which whose context the function will be executed (think function.call(context) ). Example:


var myContext = { details: "Hello, I'm the details!" };

var alertMyDetails = function () {
alert(this.details);
};

widget.onSomethingHappened.subscribe(alertMyDetails , myContext);
widget.onSomethingHappened.publish() ==> "Hello, I'm the details!"

You may also pass in an event object to the publish method which will be passed to each subscribed function. Example:

var alertObjectDetails = function (obj) {
alert(obj.details);
};

widget.onSomethingHappened.subscribe(alertObjectDetails);
widget.onSomethingHappened.publish({ details: 'These are the details!'}) ==> "These are the details!"

Yes, you can also unsubscribe: widget.onSomethingHappened.unsubscribe(alertObjectDetails, context). Just make sure if you subscribed with a certain context that you pass that same context into the unsubscribe method too.

That's about it. Download and play around with it and let me know if it's missing anything or can be improved.

UPDATE: This is kind of like a simple version of YUI's custom events.

New ASP.NET Ajax Roadmap ... yawn

2008-07-12T09:14:00.004-04:00

The ASP.NET Ajax guys recently published their roadmap. As someone who's used ASP.NET I must say it peaked my interest. But, after reading it, I realized that I just don't care anymore. I've completely given up on waiting for Microsoft to get Ajax right. Sure there are some good ideas in the roadmap but I just can't shake the feeling that they're wasting their time on the client-side library. Most of the features they're talking about adding are good but commonplace on the web today in other libraries, the libraries that we client-side developers already committed to using. Are we going to switch back for the same feature set? I won't.

I don't want to rag on MS like so many others do, there are a lot of smart people working for them, people much smarter than myself. But when it comes to web development, I'm always feeling confused with the role MS plays. I see MS as desktop and server side development only. Even if their new Ajax features came out tomorrow and were the most amazing, simple, powerful, wonderful tools for web development ever made, I would hesitate to use them. I guess it's their reputation.

This is why I'm having a hard time:

The ASP.NET framework was not really designed to be client-side friendly. Getting client-side element ids is annoying and the first Ajax library they released was very weak. While the MS team didn't do much of anything (that I know of - feel free to correct me), other popular client-side libraries continued to innovate, pushed forward with new features, and built up great communities. On top of Microsoft's historical lack of commitment to client-side web development tools, Internet Explorer has been public enemy number one for a long time.

Yes, I know, things have been turning around this year. We've got IE8 looking pretty good now and this Ajax Roadmap is a big step too. But I want MS to stop playing catch-up. If they want my respect, they need to get on the forefront and stay there for a long time, publicly pushing, encouraging, and educating developers in client-side web development. Then I'll start to come around.

jQuery 1.2.6

2008-07-10T22:05:00.003-04:00

I love jQuery. I was just checking out the latest version and noticed this small trick:

function now(){
return +new Date;
}

A funny little short cut for new Date().getTime().

Nice.

My blog: it's about time

2008-07-10T21:29:00.001-04:00

So I finally got my blog set up. Hope you enjoy it. Now I just have to think of something to post about...

Aaron Heckmann

Javascript RegExp

What are Programs?

Micro-optimization: Javascript Recursion vs. Loops

Recursion

Function loop

Vanilla loop

Recursion

Function loop

Vanilla loop

So what does this tell us?

Conclusion

Writing a jQuery plugin manager - Part 3 - Fully resolving URIs cross browser

Surprises

Overview

The Solution

Conclusion

Writing a jQuery plugin manager - Part 2 - Detecting when downloaded javascript was parsed by the browser

Surprises

Getting Started

Plugin Registration

Conclusion

What's Next

Writing a jQuery plugin manager - Part 1 - Detecting when stylesheets are loaded

Surprises

Getting Started

Implementation

Update

Tested Browsers

What's Next

Auto-silence Mobile Devices?

Save MySQL

Legislate the poor into prosperity

Internet Explorer 8 hashchange slowness

Javascript Performance

IE8 b2 - CSS Expressions

Cross Browser Testing

Fire Eagle Walk-Through Trouble

Nintendo Wii Points Order Form Nightmare

Cost of Living

The Web Design Survey

Tricky Code

New Look

js Observer updated

js Observer

New ASP.NET Ajax Roadmap ... **yawn**

jQuery 1.2.6

My blog: it's about time

New ASP.NET Ajax Roadmap ... yawn