To check out the app you can just tweet:
State Bird @sndchck

Then log into sndchck.com and you’ll see a ‘State Bird’ playlist. Repeat with any artist you want.

If you want to share music with a friend you can use this format:
@rickyrobinett Mac Lethal @sndchck

Then when your friend logs in they’ll have whatever artist you recommended added to their playlist.

I came into the Hack Day with high hopes of creating an application I’ve been wanting to make for a long time. Much like life, things didn’t turn out how I had planned. After coding for about 7 hours on my idea I realized that the task I was taking on was too big for one programmer to do in 20 hours. After much internal debate I decided to abandon my idea and try for something new. The problem was I didn’t come with any other ideas beforehand, and so I spent another hour and a half flailing around.

The application I ended up building is what I’m hoping will morph into a ‘social search’ of sorts and I’m really happy with what I accomplished. The application takes a search (keyword and location) and uses the foursquare and Yelp APIs to give you weighted search results that are more relevant/contextual than what you would typically receive. The process right now is… crude, but I think the idea is promising.

I’m looking forward to continuing with this idea and hopefully launching some version of it in the near(-ish) future to see what people think.

Outside of what I built, the thing that I enjoyed most and was most impressed with during the Hack Day is the amazing talent in the tech world. I felt very fortunate to be surrounded by developers who have launched (and often times sold) their own businesses. I was also humbled by how willing these developers were to help anyone and by their lack of arrogance. It was an environment of absolute respect, and that was refreshing.

Right now, Facebook gets my vote for scariest company on the web.

The obvious power play in launching the Open Graph API is concerning. People are starting to build applications using a Facebook API that has zero oversight from the community. Facebook, like Apple, has shown that they make decisions based on arbitrary and emotional factors instead of facts. Look at what’s going on right now with Zynga. They’re at the forefront of social gaming and when they refused to go exclusive to Facebook, Facebook responded by restricting  Zynga app messaging. As a result Zynga is considering leaving Facebook completely. (source: http://techcrunch.com/2010/05/07/zynga-gunning-up-and-lawyering-up-for-war-against-facebook-with-zynga-live/).

Just as concerning is Facebook’s acquisition of  scraping company Octazen Solutions in February. Although this was mostly a talent acquisition we’re already beginning to see what they’re going to be using this acquisition for (http://www.reddit.com/r/AskReddit/comments/c2ko3/how_did_facebook_guess_my_email_addresses/). The worry is no longer “what information am I giving Facebook about myself?” but instead “what information is ANYONE giving Facebook about me?”. It’s a scary shift in the way information is used and gathered on the web.

There are some languages we date, some we fall in love with, and some with which we even have nasty break-ups.

The First Date

Maybe it’s just me, but first dates were always awkward. You never really knew what the boundaries were or how to act. Should I walk her to the door? Should we do dinner and a movie or just dinner… just a movie? As well, you spend a lot of time focused on how to make yourself look good instead of on what your date is doing. I think this the same thing for when you first start using a programming language. How does scoping work? Are the variables loosely typed? What IDE do I use? The whole thing is like a clumsy first date. On top of that you typically try to take on an overly complicated tasks so you can feel ‘impressive’. Sometimes things are so clumsy that you decide its not worth it to pursue this relationship and sometimes you decide things went well enough that you want to take it to the next level.

Dating

From the first date you have a basic understand of the person you’re seeing. You know a little bit about their family, where they’re from, what they do for a living. As you go on more dates you learn some of the more intricate details about a person. You learn their favorite foods, their favorite television shows, maybe even their favorite toothpaste. After the ‘first date’ with a programming language you have a basic understanding of how it works. You’ve become familiar with the syntax, you know how to run it on your machine, you’ve even built a ‘hello world’ app. As you use it for more and more projects you begin to see how it handles sessions, arrays and database connections. The more you use it the more you think, ‘this could be love!’.

“Love”

You’ve just fallen in love and your partner can do no wrong. Its OK they don’t do the dishes. It’s OK they tend to only shower once or twice a month. You LOVE them. All your friends may tell you that you’re dating a loser, but you don’t care what they have to say either. In fact, you spend a lot of your time defending your partner against these attacks. When we fall in love with a programming language we can experience the same reaction. Its OK that ColdFusion  can’t handle the loads that other languages can, I don’t care about high traffic applications anyway. I don’t care that everyone makes fun of me for using PHP, they just don’t realize the true beauty of it. Many of us fall in love with a language and never leave this phase. We go to forums and plead the case for why our language is the best. We never look anywhere else for our needs, because our language seems safe and it takes care of us.

True Love

Ok, so before I had “love” in quotes. In a relationship, we have those moments where we think this is the one and then we have the moments where we know this is the one. I’m going to be with this person for ever. A big part of this is coming to the realization that we all have faults and that when you love someone you both work together to become a better couple. Before, you thought its OK they don’t do the dishes or take showerss, but when you reach the point of true love you talk to them about it. They work to improve these things that bother you because they love you. With a programming language this is the phase when you go beyond making excuses for the language and actually start working to make it better. Your language doesn’t interface well with the newest web 2.0 API, so you contribute a library that does. You join mailing lists and help other people who may be having similar problems. You even start going to user groups and meet people like you.

The Break-up

Sometimes you never reach true love. You realize the things that you initially thought you could ignore your partner now drive you crazy. The longer the relationship has gone on, the harder it is to break-up. The longer a relationship goes on, the more you put of the break-up and convince yourself you can make it work. Your friends are their friends and one of you is going to have to move out of the apartment. When we break up with a programming language it can be the same way. All your friends are [your language here] programmers, you’ve been  going to the user groups ever week for years, you may even have to switch companies or departments to get a real shot with another language. The longer you’ve used it, the harder it is. The longer you’ve used it, the longer you put off an inevitable break-up.

So where are you with your programming language? Are you going through a break-up? Are you just falling in love? Or are you a player who’s dating a lot of languages (Remember, even Jay-Z eventually settled down)?

I’m a big proponent of the philosophy that the way to optimize code is by making it do less work. Whenever I encounter an issue related to performance I look to see “How can I make this code do less work?”. After some investigating I isolated the problem to the following chunk of code (this a very simplified version of what I saw):

for(i = 1;i lte ListLen(someList);i++){

    value = ListGetAt(someList,i);

    // do something with value here

}

Immediately I saw a way to make this code do less work. The ListLen function was being used in the conditional statement for the loop. This means every time the loop was evaluated it was running this function. I immediately made the following change expecting to see my performance improve.

// put length list in variable to avoid rerunning of function in conditional statement

lengthOfList = ListLen(someList);
for(i = 1;i lte lengthOfList;i++){

    value = ListGetAt(someList,i);

    // do something with value here

}

Upon rerunning the function I saw an improvement in performance, but not as much as I expected. After some more debugging I determined that there was a lot of time being spent on the ListGetAt function. I began to research the management of lists in ColdFusion and found something a bit shocking… Specifically, the fact that although ColdFusion provides list functions it still just sees a list as a string under the hood. There’s no intelligence to the indexing, so when you call the function ListGetAt(list,pos) it actually traverses the string by delimiter from the beginning until it reaches the position. Though this doesn’t sound super costly, think of doing ListGetAt() in the loop we’ve been looking at. Then consider the fact that I was working with a list that contained almost 200,000 values. The cost of doing this loop is O(n^2).

// convert list to array so this loop doesn't last as long as a James Cameron movie
aConverted = ListToArray(someList);
lengthOfArray = ArrayLen(someList);
for(i = 1;i lte lengthOfArray;i++){

    value = aConverted[i];

    // do something with value here

}

After making this change my performance issues were resolved. I saw pages taking as much as 5-10% the amount of time the were previously taking.

Moral of the story is this, lists can be a convenient data structure in ColdFusion but be careful how you use them. If you’re going to be tracking and traversing a large amount of data then an array is going to be more efficient.

If you look at the first blog post you can see the entire code block. In this post I’m going to focus on the functions I use and not rehash the entire page. I did clean up some of the functions since the initial post and hopefully will at some point update that code.

$(function() {
$("#items").sortable({
items: 'li',
cursor: 'move',
stop:     function (){
renumberItems();
}
});

});

This is the first function you’ll find and it’s your standard sortable creation. The only thing special about this is that I have a callback function for stop that calls the renumberItems function.

function moveUp(instanceID) {
if(instanceID != $('#items li:first').attr('id')) {
$('#'+instanceID).insertBefore($('#items li:first'));
renumberItems();
}
}

You’ll notice the moveUp function looks a bit different than the one I originally posted. After posting it I discovered a bug that resulted in the first item disappearing if you tried to move it up. I now check to make sure that the item that’s being moved isn’t the first item. After that I just the jQuery insert before function and place the desired item before the first item in the list. I then call the renumberItems function that we’ll look at shortly the update the text boxes.

function moveDown(instanceID) {
if(instanceID != $('#items li:last').attr('id')) {
$('#'+instanceID).insertAfter($('#items > li:last'));
renumberItems();
}
}

The moveDown function just does the opposite of the moveUp function. I take advantage of the insertAfter function to move the selected item after the last item in the list.

function removeItem(instanceID) {
$('#'+instanceID).remove();
renumberItems();
}

The moveItem function just calls the remove function on the selected element. We then call renumberItems to update the textboxes with the correct numbers.

function renumberItems(){
count = 1;
$('#items li').each(function (i) {
id = $(this).attr('id');
$('#'+id+'-order').attr('value',count);
count++;
});
}

The renumberItems function has be referenced a couple times and I’m sure you’ve been eager to finally take a look at it. It’s actually a pretty basic function, we select all the li items and loop over them using the each function. We then get the id and use that id to set the related order text box to the correct number using an incrementing count variable.

function reOrderItems(instanceID) {
var curPos = $('#items li').index($('#'+instanceID))+1;
if($('#'+instanceID+'-order').attr('value') >= $("#items li").size()){
$('#'+instanceID+'-order').attr('value',$("##items li").size());
if(curPos == $("#items li").size()) {
$('#'+instanceID+'-order').attr('value',$("#items li").size());
return;
}
}
if($('#'+instanceID+'-order').attr('value') < curPos){
$('#'+instanceID).insertBefore($('#items li')[$('#'+instanceID+'-order').attr('value')-1]);
} else {
$('#'+instanceID).insertAfter($('#items li')[$('#'+instanceID+'-order').attr('value')-1]);
}
renumberItems();
}

This function function ended up being more complicated than I anticipated. It is called when the user changes the number in the text box. The first thing the function does is gets the current location for the element. We then check that the user didn’t select a number larger than the element list. If they did, we reset the number to be the size of the list. We then double check that this element wasn’t already in the last position and if it is simply return. Next we look to see if they’re trying to move to a lower or higher position. We use this to determine if we should either place it before or after the element currently in the position. Finally, we renumber the items.

The problem with citizen journalism is that it can’t be trusted. Who’s to say that the person posting on twitter has any idea of what they’re actually talking about? Professional journalist have a certain level of accountability which keeps them in check. This accountability typically happens before a story reaches the world. The dark side of citizen journalism is that its accountability doesn’t happen until after the information has already been reported. I was reminded of this today with the death of Chris Henry. An elaborate twitter hoax reported Chris Henry’s death 12 hours before it happened. In a similar fashion to the Wikipedia hoax incident with the death of Chris Benoit’s wife, an internet troll took a bit of public information and used it to report a fabricated, and twisted, conclusion. In both situations, that conclusion ended up being the truth.

The real question is, how do we enjoy the benefits of these ever-growing networks of citizen journalists while avoiding the dark side of its users? Is this dark side unavoidable? Should we just take any report from these networks with a grain of salt?

SELINUX=enforcing

To

SELINUX=disabled

You’ll then need to reboot your server. Once I made this change my PHP application was able to connect to the database without any issues.