First you need to grab a copy of Git for Windows. Once you’ve downloaded the latest version, launch it and go through the license and location screens. Once you get to the “Select Components” screen, make sure both Context menu entries are selected. If you want, select the TrueType font option. That one shouldn’t really matter.

Jump ahead a few more frames and you’ll get to the SSH executable option. Choose “OpenSSH”, unless you have a compelling reason to use Plink.

Run through the rest of the frames, accepting the defaults. You now have Git installed on your machine, but we’re not quite done yet. GitHub uses public/private key encryption. This means you don’t commit with a username and password, like most SVN setups. Instead, you tell GitHub your public key, then use your private key to authenticate.

If you don’t already have a public/private key pair, you need to generate one. You can do so with PuTTYgen. If you launch PuTTYgen, you should see something like this:

Click “Generate” and move your mouse around for a bit. In a few seconds you’ll have a key. Go ahead and copy the public key (the big text box). We’ll be entering that into GitHub, but don’t close PuTTYgen just yet. Back at GitHub, if you go to your account settings, there’s a page for “SSH Public Keys”. There you can click “Add a public key”, or “Add another public key” (if you already have one). Paste in the public key, give it a descriptive title, and save.

Great. Now GitHub knows your public key. This means you can use your private key to authenticate. Back to PuTTYgen. If you want to make your private key more secure, enter a passphrase. This makes it so, even if somebody gets your private key, they can’t use it to access GitHub unless they also know your passphrase. Once you’ve entered your passphrase, go to Conversions -> Export OpenSSH key.

You need to save your key in a very specific location. Go to your home directory (C:\Users\<username>\). Create a folder called “.ssh”. Now save the OpenSSH key to “id_rsa” inside that folder. So for example, if your username is “banana”, you should now have a file C:\Users\banana\.ssh\id_rsa that contains your private key. It’s also a good idea to save your public key, in case you need to use it for any other servers at some point. Click the “Save public key” button and save your public key in the same folder. Call it “id_rsa.pub”.

You’re just about there. Go to your GitHub repository and select and copy the text in the textbox near the top of the page. Make sure SSH is selected.

Open up Git Bash. You should be able to find it in your start menu. “cd” to the directory where you plan on checking out your code. For example, if you want your code to be located in C:\programming\project then you want to “cd” to C:\programming.

Type out “git clone “, but don’t hit enter yet. Go to the top left corner of the Git Bash window and click. In the menu that pops up, select Edit -> Paste to paste the URL of your repository, but still don’t hit enter. Finally, type the name you want to call the project folder. In our example, you’d want something like this:

git clone git@github.com:itsananderson/wp-sendgrid.git project

Now you can hit enter

If you entered a passphrase for your private key, it will ask you for it, then if all goes well, your repository should be downloaded. This isn’t really a tutorial on Git commands, but here are a few to get you started:

See what files have changed:

git status

Add a file (new or modified, it doesn’t matter. You need to add it to commit it)

git add file.php

Add several files

git add *.php

Commit your changes locally (doesn’t put them on GitHub, just stores an update locally)

git commit -m "Your commit message here"

Push your changes up to GitHub

git push -u origin master

Pull down updates from GitHub

git pull

That’s about it. There’s more to Git (which I’m still learning about), but this should hopefully be enough to help you get started.

This plugin extends WordPress’ wp_mail() function to use SendGrid’s API for all emails. This was useful to me because I’ve never been able to figure out how to get PHP to send emails in Windows. In the past, when I’ve worked on projects that send emails, I always had to upload it to a Linux server to test the email functionality. If anything didn’t work, I had to edit locally, then upload again to test it. After a few iterations, that gets really old. With WP SendGrid, I can send emails from my local machine, so debugging issues is much easier.

Fast-forward a few months. This week I decided to spend a little more time polishing the plugin. I’d only spent about 45 minutes on it at the HackDay, which shows how simple it is to use SendGrid’s API. What I put together was enough to meet my needs, but I was using hardcoded values, and wasn’t considering every edge-case. After a few more hours of work, I had a complete plugin with a nice configuration UI.

If you develop for WordPress on Windows, or you have trouble sending email from PHP on another platform, you’ll want to check this plugin out. Visit the plugin page.

This is a new plugin, so YMMV. If you run into any issues, let me know by leaving a comment here, or submitting a ticket on the GitHub project.

This is also a special plugin, because it’s the first one I’ve simultaneously published on GitHub and my site. I plan on uploading my other plugins to GitHub at some point (in fact, I uploaded Short Comment Filter, another popular plugin, a few days ago), but Minimum Password Strength is the first one that’s there from the start.

Anyway, despite all this pain, I decided to start using GitHub to host my open source WordPress plugins. This gives me some more experience with Git, plus it provides public source control, which is better for open source projects than the private SVN repository I was already using.

Long story short, if you want to see what I’ve uploaded so far, check out my GitHub page.

Anyway, I wrote this counter a while ago, but I had some weird bugs that I never fixed, so I never released it. When WordPress 3.3 came out, I decided to revive it. Here it is!

In the case of the Rose-Hulman lookup page, there are actually two parameters that it would be nice to query on. The first is the username of the person you’re looking up, and the second is which quarter (Fall, Winter, Spring, Summer) to show. With Chrome’s custom search engines, you can’t specify multiple parameters, so you have to hardcode the quarter parameter (which is hardly convenient). To resolve this problem, I wrote a simple PHP script that takes in the search a search query (like “anderswc 1″ or “jetheis 201130″) and translates it to the correct schedule lookup page query (my fall schedule, or Jimmy’s spring schedule last year). Here’s the script:

<?php

function get_academic_year() {
	if ( intval( date('n') ) >= 8 ) {
		return date( 'Y' ) + 1;
	} else {
		return intval( date( 'Y' ) );
	}
}

function get_quarter() {
	$month = intval( date( 'n' ) );
	$day = intval( date( 'j' ) );

	if ( 6 == $month || 7 == $month ) {
		return 40;
	} elseif ( ( 8 <= $month && 10 >= $month ) || ( 11 == $month && 20 >= $day ) ) {
		return 10;
	} elseif ( 11 <= $month || 2 >= $month ) {
		return 20;
	} else {
		return 30;
	}
}

$search = $_GET['search'];

$parts = explode( ' ', $search );

if ( 1 == count( $parts ) ) {
	$term = get_academic_year() . get_quarter();
} else {
	$term = $parts[1];
	if ( strlen( $term ) == 1 ) {
		$term = get_academic_year() . $term . '0';
	} elseif ( strlen( $term ) == 2 ) {
		$term = get_academic_year() . $term;
	}
	$term = intval( $term );
}
$url = 'https://prodweb.rose-hulman.edu/regweb-cgi/reg-sched.pl?termcode=' . $term . '&view=tgrid&id1=' . $parts[0] . '&bt1=ID%2Username';

header( 'Location: ' . $url );

Now you can create a custom search engine to point to this script (I’ve got a copy running at http://itsananderson.com/schedule.php), and specify “?search=%s”. For example, this is the custom search engine definition I have:

http://itsananderson.com/schedule.php?search=%s

Search syntax is pretty simple.

• “<username>” to look up a user’s current schedule.
• “<username> <quarter>” to look up a user’s schedule for a specific quarter this year. Specify quarters as either 1, 2, 3, 4 or 10, 20, 30, 40
• “<username> <year><quarter>” to look up a user’s schedule for a specific quarter and academic year. Quarters must be formatted with a trailing zero. Example: 201220 is Winter quarter, 2011-2012

To make a long story short, getting the actual class name of the current object in an instance method is easy, just call get_class( $this ). The problem is, $this doesn’t exist in a static method, and just calling get_class() returns the current class (basically the same as __CLASS__), which isn’t helpful if you actually want the name of a child class. As an example, consider a base class which has a get_instance() function, for creating an instance of the class, or returning an existing instance if it has already been instantiated.

class Singleton {
    private static $instance;
    public static function get_instance() {
        if ( is_null( self::$instance ) ) {
            self::$instance = new Singleton();
        }
        return self::$instance;
    }
    private function __construct() {
        // Singleton constructor stuff
    }
}

This is great, but the problem arises when we try to extend that singleton class. Ideally, the inherited get_instance() method would return an instance of the actual class, but with the above implementation, it would always return an instance of the Singleton class. Here’s an alternative, which uses the get_called_class() function.

class Singleton {
    private static $instances = array();
    public static function get_instance() {
        $class = get_called_class();
        if ( empty( self::$instances[$class] ) ) {
            self::$instances[$class] = new $class();
        }
        return self::$instance;
    }
    private function __construct() {
        // Singleton constructor stuff
    }
}

class Child extends Singleton {
    // Child class stuff
}

You’ll notice that there are a few changes here. First, the Singleton class now stores instances in an associative array, keyed by class name. Second, it uses get_called_class() to get the class on which this method was invoked. This means calling Child::get_instance() will return an instance of Child, not Singleton. Pretty sweet, huh?

To understand what I’m babbling on about, consider a very basic plugin that adds square brackets around post titles.

It really doesn’t get much simpler than that, but the problem is that there might be another plugin that uses the same prefix. Also, if you decide to change the prefix, you’ll have to go through and change all of your hook and action registrations too. What a pain.

The next option is to use a class as a namespacing tool. This is fairly common among more experienced WordPress developers.

What I propose is developers use static methods instead. There’s very little difference between a singleton class with instance methods and a class with static methods, and simply using static methods can simplify your code. Check out this example.

So that’s that. All the advantages of namespacing with classes, with a small reduction in memory overhead (because you’re passing around strings, rather than objects). Static methods are a PHP5 feature, but with PHP4 support being phased out in WordPress 3.2, this seems like a reasonable next step for WordPress developers to take.

What do you think? Are there any issues with this technique that I’ve missed? Are there any improvements that could be made? Let me know in the comments!

The slides are available here.

The WP_Ajax class is available here.

The Simple Post Ratings plugin is available here.

I’ll also post a link to the video on WordPress.tv when it becomes available. EDIT: Here it is

The first thing you’ll want to do is add an alias directory to your XAMPP install. If you’re running Windows, create the following folder.

C:\xampp\apache\conf\alias

Next, you’ll need to modify your Apache configuration file. You can find it under C:\xampp\apache\conf\httpd.conf. To make changes, you’ll need to edit it as an administrator. If you’re running Windows Vista or Windows 7, your best bet is to open Notepad as an administrator and then open httpd.conf. To open Notepad (or any other application) as an administrator, right click on it and select “Run as administrator”.

Once you’ve opened httpd.conf, add the following to the end and save it.

Include "conf/alias/*"

Now Apache will look in the alias folder for additional configuration files. This way all you have to do to add an alias is add a new alias configuration file to the alias folder.

Let’s do that now. Suppose you want to add an alias called “dev”. First, download this alias template file and place it in the alias folder. Rename it to “dev.conf” (or whatever you’re alias is called). You’ll need to edit the template (again, as an administrator) and replace “DIRECTORY” (it’s there twice) with the path to the alias and “ALIAS” with the name of your alias. In our example, it looks something like this:

<Directory "C:\users\foo\programming\dev">
    #
    # Possible values for the Options directive are "None", "All",
    # or any combination of:
    #   Indexes Includes FollowSymLinks SymLinksifOwnerMatch ExecCGI MultiViews
    #
    # Note that "MultiViews" must be named *explicitly* --- "Options All"
    # doesn't give it to you.
    #
    # The Options directive is both complicated and important.  Please see
    # http://httpd.apache.org/docs/2.2/mod/core.html#options
    # for more information.
    #
    Options Indexes FollowSymLinks Includes ExecCGI

    #
    # AllowOverride controls what directives may be placed in .htaccess files.
    # It can be "All", "None", or any combination of the keywords:
    #   Options FileInfo AuthConfig Limit
    #
    AllowOverride All

    #
    # Controls who can get stuff from this server.
    #     Order allow,deny
    Allow from all 

</Directory>

Alias /dev "C:\users\foo\programming\dev"

Restart Apache (you can do so from the XAMPP control panel) and check out http://localhost/dev

The following screenshot illustrates some of the default post states.

Adding your own states is actually pretty easy. The following code snippet should get you going in the right direction.

Assuming you have code elsewhere that is adding the custom field to posts, you’ll see something like this.

That’s pretty much it. Obviously there are more useful examples, but this should be enough to get you started.

Time Honored Destiny

It had been four months since the event. Four months of hell. If Thomas hadn’t started his project years before the outbreak, he would have given up weeks ago. The world had been shocked by how quickly the sickness spread. Thomas didn’t have time to be shocked. He just kept working. Before the sickness everyone had told him time travel wasn’t possible. They tried to convince him that he was wasting his time. He wondered how their opinion might change if they knew that his work might be the salvation of the human race.

“It’s finally ready.” Thomas glanced at John, the only other person immune to the sickness. He was glad to have a companion to listen to his ramblings but wished it was someone who could respond. John was completely paralyzed. His body was so horribly disfigured it was almost unrecognizable as human. Thomas had found him a week after the first outbreak while looking for supplies at a nearby hospital. At first he’d thought the disfigurement was some strange manifestation of the sickness, but the hospital records said he’d been in a bus accident. He had never been identified by the hospital, so Thomas christened him John Doe. At the time he’d laughed at the generic name and promised to come up with something better, but with all of his creativity being poured into his work he never had the time.

The machine only had room for Thomas, but if he was successful he would change history for both of them. John had been crippled before the outbreak, but by changing the past, Thomas could at least ensure he’d receive the medical attention he needed.

Trembling with excitement, Thomas calibrated the machine to place him across the street from the CDC headquarters in Atlanta. He gave himself a month before the initial outbreak. Surely that would be enough time to prevent the world’s doom. All they had to do was find patient zero. By seizing one improperly sanitized needle from a washed up heroin junkie they could prevent the end of humanity.

“Wish me luck.” He murmured. The countdown commenced. 3…2…1…

After a few instants he immerged, dizzy and disoriented. He spotted the entrance to the CDC and stepped into the street. In his excitement he didn’t even see the bus as it raced toward him.

That’s it. Whatcha think?

The example I’ll use is the $wpdb global variable defined in WordPress. It contains an instance of the wpdb class, which wraps the database connection and provides several helper methods. The problem is that it is defined in the global scope, so when you use it in another file, your IDE has no clue what type it is.

If you add the following code snippet to the beginning of your function (or file, if applicable), it will tell your IDE the type of the $wpdb variable without actually affecting the execution of your code.

if ( false ) {
    $wpdb = new wpdb();
}

As you can see, the code inside the if statement will never execute, but every IDE I’ve ever used isn’t smart enough to figure that out, so they assume the type of the $wpdb variable is now wpdb.

When you add the snippet, you get code completion suggestions that look something like this:

The nice thing about doing this is that if you forget to remove the code snippet, it won’t break anything. You’ll lose a CPU cycle or two evaluating the if statement, but your code will run without any problems. It’s probably a good idea to remove it before you release your code, though, since it might be confusing to anyone who looks at it later.

Leslie’s never-say-die attitude allows her remain positive, no matter how insurmountable the obstacle she’s facing. Her persistence is a source of comic relief, but in the end she always seems to succeed in her endeavors.

This attitude toward problems should serve as both an inspiration and a warning to anyone feeling discouraged. When you’re stuck on a problem, don’t let the Ron Swansons of your life keep you from going after your dreams. On the other hand, don’t become so wrapped up in what you hope to accomplish that you’re unable to listen when your friends and loved ones tell about potential problems. You shouldn’t be discouraged by your challenges, but you also shouldn’t ignore them. They won’t go away just because you pretend they aren’t there. Face your problems with optimism, but be realistic. As with almost every aspect of life, the key to success is maintaining a balance.

OK. Enough pep talk. We now return you to your regularly scheduled program.

I’m taking Don Quixote as a humanity elective. It wasn’t my first choice, but it had the same schedule as the course I wanted to get into (Introduction to Film Studies).

Even though I’m a CS, I decided to take Formal Methods. Since it’s an SE course, it doesn’t count as a CS elective, but I have a bunch of general electives to complete, so it’s not a big deal. Curt Clifton is teaching it, so it should be pretty awesome.

Lastly, I’m taking Cryptography as a Math elective. This year it’s being taught by the CS department, which should mean it will be both accessible and interesting. Taking CSSE120 from Matt Boutell was what convinced me to go into CS, so I look forward to taking another class from him.

Overall, I’m pretty pleased with my schedule. My day doesn’t start too early, but it also doesn’t end really late in the afternoon. Should be peachy.

What Dougal suggests is that theme developers instead dynamically include another template file to handle the post format. I’ve seen a similar idea suggested in a Trac ticket, and the issue I saw (but neglected to comment on) there is the same one I see here. Including PHP files is a relatively expensive operation.

Now, admittedly, including files isn’t the worst thing you can do, and WordPress already includes dozens of files for every page request. Still, that doesn’t mean you shouldn’t try to get as much performance out of your theme as possible. On an average page load on an average website, using the get_template_part() method for Post Formats will cause 10 additional includes, but what happens if the blog admin decides they want to display 50 posts per page? The solution doesn’t scale very well.

The alternate solution I propose is to dynamically call functions via call_user_func. When the name of a function is passed to call_user_func, that function is invoked. Incidentally, that’s actually how WordPress does function callbacks for its actions and filters.

Here’s how the code might look.

First, in your functions.php file, define functions for the Post Formats you wish to support. You could also place them in a separate file (like formats.php for example) and include that from your functions.php. That way you can keep your logic separated from your display, while only adding one additional include instead of ten.

Next, in your template, use the post format to determine which function to call. To make your theme more robust, make sure the post format is supported before you try to call the function. If it isn’t, call your backup “default” format function.

OK, not terribly exciting, but I ran across an interesting challenge, and even more interesting solution while I was working on it.

One of the things you probably noticed is that the “price” value is updated automatically when you change the tax rate. While that isn’t a terribly complex feature to implement, it wasn’t quite as simple to do as I expected it to be, so I thought I’d share my solution in case it’s helpful to somebody else.

The JavaScript event I used to capture key events was the keydown event. The challenge of this event is that it fires before the field is actually updated, so directly accessing its value won’t work. The solution I found was to setup a timeout event to be executed a few milliseconds later. This event would be able to access the updated value of the field.

Here’s a snippet of code that illustrates the concept.

$('#field').keydown(function() {
    setTimeout(function() {
        var fieldValue = $('#field').val();
        // Code that actually uses field value
    });
}, 50);

Well, I’m not really one to obsess over my grades, but I was still faced with a choice of saying no to more freelance work, or screwing up my GPA. So, I chose option number three.

While looking over my course requirements in preparation for Winter registration, I realized that I could take 12 credits per quarter and still graduate next year. Reducing my work load by one course per quarter will hopefully mean that I can take on more freelance projects, while maintaining better grades.

It really feels like this summer has flown by for me. I find it hard to believe that it was really 3 months ago that I last posted here. It’s been good though. I ended up working on the adCenter Desktop team at Microsoft. It’s just about the only team in the Online Services Department (OSD) whose main product is not a web client. My internship with Microsoft was as a Software Development Engineer in Test (SDET) so much of my responsibility on the adCenter Desktop team was writing up and automating test cases. It may not sound like the most thrilling work, but I enjoyed it, and learned a lot while doing it. I plan to return to Microsoft next summer for a second internship, though I’ll likely be on another team (hopefully still in OSD). I look forward to being back in the Northwest.

The thing that has wowed me the most about this area of the country is how green everything is. Bushes, flowers, and trees seem to adorn every lawn and office park. It’s hard to precisely describe the difference between this area and the Midwest, but I think the best way to put it is to say that here, cultivating greenery is the norm, rather than the exception. Not everything is like this, of course. Seattle itself seems a little dirtier and grayer than the surrounding area, but even with it’s trash littered sidewalks and swarms of hobos, it has a certain classic charm.

I actually only visited Seattle a handful of times when I was here. I lived in Kirkland and worked in Bellevue, which are both on the other side of Lake Washington. Living in Kirkland was really nice. The area is nice, and downtown Kirkland is only a few minutes away, affording access to a wide variety of shops and restaurants, as well as docks and a beach overlooking Lake Washington.

Bellevue is a pretty cool city to work in. Unlike Microsoft’s main campus in Redmond, most of the office buildings are fairly new, and are built to provide lots of natural light. With lots of restaurants and coffee shops within walking distance, it’s always easy to find a bite to eat. There is also a fairly direct transit route from Kirkland to Bellevue, which made it easy to ride the bus to work (and I did almost every day).

Toward the end of the summer, I got to go to a WordPress meetup that was being held in Seattle. I got to meet a bunch of people including Joshua “Red” Russak of First Time Online and Nick Ohrn of Plugin-Developer.com. Nick is a Rose alumni who has been freelancing in Seattle for the two years since he graduated. We had been planning to meet sometime this summer to talk WordPress, and he is the one who told me about the WordPress meetup. We actually ended up meeting the next day at a coffee shop and I had a great time telling him about what I’d been working on and hearing about the stuff he’s been doing. I look forward to doing some WordPress projects for him during the upcoming school year (which is what we actually met to discuss). I also got a chance to talk with Joshua this week and discuss doing some work for him as well. Suffice it to say, it’s going to be a busy school year, but I look forward to it.

How about you? What has your summer been like?

About a day and a half has passed since my accident and, while I’m still frustrated that it occurred, I can now see the silver linings surrounding the event. First, the way the accident occurred is something to be thankful for in itself. I was only traveling at twenty or so miles an hour, and I ran into an unattended parked car. I was going to pick up my mom so we could run some errands and I wanted to clean off my passenger seat because it was covered in papers from school and old packaging from fast food restaurants. I wasn’t paying close enough attention to where I was going and I rear-ended an SUV parked at the curb. While this is an embarrassing story to tell, I’m thankful that I wasn’t hurt, and that there was nobody in the other vehicle to be injured. I was able to locate the owner of the other vehicle without too much trouble, and he was very gracious about the whole thing, as was the police officer who responded to the accident report. All-in-all, things went about as well as an at-fault teenage driver can expect at the scene of an accident.

Now of course, my car is now out of commission (It’s an old vehicle, so I only have liability insurance), and my plans to purchase a replacement need to be pushed up by a couple months, but it isn’t like I was planning on driving it for terribly long anyway. Fortunately for me, the friend I’ll be living with works in the same building as the one I will be in, and we had already discussed carpooling, so getting to work shouldn’t be a big problem. The public transit system is also apparently pretty good, so I think I’ll be able to get around without too much trouble.

So where’s the silver lining? If I hadn’t wrecked my vehicle, none of these things would have been an issue anyway.

While that is true, there are other things that are less apparent at first glance. First, I had already been having issues with my car that made me nervous about taking it on a two thousand mile road trip. Oil leaks and a flaky battery were my main concerns, but there were also about a half dozen smaller issues that my mechanic had told me I should fix, but that I hadn’t had the money or time to deal with. I’m not really the worrying type, but several friends and family were concerned about me making the trip with such a clunker. This accident makes me wonder if perhaps they were right, and this was God’s way of telling me I should fly instead. Now, I wouldn’t have complained if he’d let something less destructive and less embarrassing convince me, but I guess that’s not the point. Now I can rest more easily, knowing that my on-time arrival is much more likely to occur.

Finally, flying to Seattle has other advantages as well. Avoiding a 32 hour road trip means I’ll be arriving two days earlier than originally planned, so I’ll have more time to get settled before I start work full time next week. I’ll also have that much more time to spend with Tony and Jackie Ferrell, who I’ll be staying with during the summer. They’re great friends who I’ve missed, so I welcome the opportunity to see them sooner.

The biggest draw back in all of this is that I won’t be able to see my aunt and grandma on the way out. I was looking forward to stopping to visit with them for a little while, since I was going to be driving right past their home, but perhaps I can do so on my return trip instead.

At any rate, that’s what’s been going on with me recently. What’s up with you?

Privacy? What Privacy?

Something most people apparently don’t realize is that whenever they browse the web, they’re already giving up some of their privacy. Just by visiting a website you give them information about yourself. Your IP can be used to get an idea of where you are in the world. The server can even determine what operating system and browser you are using, and unless you disable cookies, it can even remember you if you visit again. None of these things are a real problem, but they illustrate the point that online privacy doesn’t really exist.

When you interact with any website, you give them even more information. Register an account? They now know your email address. Most websites encrypt passwords before storing them on their database, but you never know when a web site might store it as plain text, potentially giving the administrator access to any other accounts you’ve created with the same password. Any other information you provide (age, gender, interests, address) is also no longer private. For the most part that isn’t a problem, but there’s really nothing keeping the owners of the website from using that information in any way they please.

Consider Gmail. I use it (or at least Google Apps, which is essentially the same thing) for my email. Something I noticed early on was that the ads that are displayed next to my emails aren’t random. They often have something to do with the subject of my emails. Not only that, but they’ve become more and more related over the years as I’ve used Gmail. What this suggests is that Google not only “reads” my emails to figure out what ads to show, but they’re keeping track of my interests in order to provide more relevant ads. There’s not necessarily anything wrong with this. Google needs to make money just like anyone else, and offering relevant ads seems like a pretty good way to do that, but this serves as another example of online activities not really being private.

Finally let’s consider Facebook, Twitter, and social networking in general. We’ve now added an extra layer of information distribution. Not only is your information being shared with a web server, but it’s now being distributed to people around the globe. Most social networks have methods by which you can hide certain information. Facebook has detailed privacy settings and Twitter allows you to “protect” your tweets, requiring users to send you a request before they are allowed to view them. What you have to remember, though, is that any information you’ve shared with the social network is really no longer private, regardless of your “privacy” settings. So when you complain about your boss, or post embarrassing pictures of yourself online, you’ve already made the first mistake. Privacy settings are convenient, but you have to remember that no information is really private once it’s been posted to the internet.

So what’s this all mean? Should you be scared to use the internet, much less visit social networks? No. Just remember that anything you don’t want the world to know probably shouldn’t go online, especially not to a social network.

Other Issues

Ok, so I’ve explained why I think Facebook privacy is a non-issue, so why did I consider leaving Facebook? Put simply, applications and security.

If you’ve used Facebook for more than a couple months, then you probably remember the FarmVille and MafiaWars craze that so many people bought into. I remember a period of time when it seemed like every other entry on my feed was either “So-and-so needs help with a job” or “So-and-so just won a blue ribbon in X”. I refused, on principle, to participate in either of these games, and was constantly annoyed by all the noise. Today there are dozens of these kinds applications, with new ones showing up almost daily. I use social networks to connect with people socially, not to play games, so I find it annoying to be bombarded with random “news” items from applications my friends happen to be using. I understand that not everyone shares my views on this, but it seems like it should be possible to opt out completely, hiding all applications from showing up in my feed.

Facebook’s security is fairly robust. They even added a new feature recently that lets you receive notifications whenever somebody logs into your account from a different machine. My problems with Facebook security lie, again, with applications. Over the past year, I’ve seen dozens of “applications” that provide no useful functionality. If you add the application, it sends notifications to your friends, asking them to add the application. If they do, the pattern continues. These applications are little more than glorified viruses, and I think the Facebook management should be doing more to eliminate them and prevent new ones from being created.

Alternatives

Right now there aren’t very many alternatives to Facebook. Twitter is a great service that I enjoy using, but it caters to a different crowd of users. I don’t really expect my Grandma to ever use Twitter, so if I want to connect with her online I have to use a network that she can understand and use.

One alternative that has a lot of promise is BuddyPress, an open source project backed by the same people behind WordPress. It essentially allows you to build your own social network. It’s a pretty slick tool, with a lot of room for customization, but the main problem with this kind of distributed network is that almost all friend connections are lost. The great thing about Facebook is that I can visit one website and see updates from personal friends, fellow college students, and relatives. This is more difficult with a self hosted solution like BuddyPress.

There is group of students who want to create a “distributed Facebook”, called Diaspora. At the time I’m writing this, they’ve acquired over $180k in funding, which is quite impressive. The main idea for their project is to create a completely distributed social network, so rather than posting your data to a central hub, you host it on your own node, allowing you to have more control over what information you make available and who can see it. This sounds like a fantastic idea, but once again, my Grandma is a problem. I don’t expect her to ever be able to set up her own server to host her own node. My guess is that less than 1% of the Facebook population has the knowledge and resources that kind of setup requires. For an idea like this to work, they’re going to need to provide an alternative “hosted node” option. Even then, it would take a long time before enough people switched networks for a Facebook exodus to be feasible.

Conclusions

In closing, I believe that people’s privacy issues with Facebook arise from a misunderstanding of online privacy, not necessarily just from poor Facebook policies. By no means do I look at Facebook through rose-colored glasses, but I think too many people are expecting the Facebook administrators to protect their information, rather than taking responsibility for their own privacy.

Remember. Nothing you do online is truly private.

Now, just a quick note, lest you get confused. It is Seattle itself, not my experience there, that I expect to be mostly cloudy.

The reason I’m moving to Seattle this summer is because I’ll be working for Microsoft as an SDET (Software Development Engineer in Test) intern. A lot of meaning and potential responsibilities come with that title, but the most simplistic job description of an SDET might be “I break Microsoft products so you don’t have to”. I’m not sure exactly what I’ll get to work on during the three months I’m there, but whatever it is I’m sure it will be both interesting and experience building.

Beyond my work at Microsoft (which will take up most of my time) I also hope to get to see a lot of the area surrounding Seattle. I enjoy Disc Golf, so I plan on trying out nearby courses while I’m there. Microsoft organizes tons of events for summer interns, so I’ll probably take advantage of a few of those. I’m not sure what events will be available, so I don’t know yet what I’ll be interested in, but it sounds like there are enough of them that I’m sure something will pique my interest.

I’ll be living with Tony and Jackie Ferrell while I’m in Seattle. Tony is a Rose grad, which is how I got to know him and Jackie. I’m sure we’ll have some fun adventures around Seattle. I’m also excited about exploring some of the project ideas Tony and I discussed while he was still at Rose. I’ll hopefully have more news on that later this summer.

Once the summer ends, I’ll be taking 4 computer science courses next fall. Normally students only take one or two CS courses each quarter (filling the rest of their time with “easier” electives), but for the courses I’m taking it was now or never, so I decided to do it. It probably means it’ll be the busiest quarter of my college career, but being able to take all of those courses should make it worth it. I also plan on auditing at least part of one of the courses online this summer. That should hopefully reduce my workload in the fall (though it will mean I’ll be even busier this summer).

And that about sums it up. As usual, I’ve probably overbooked myself this summer, but I always figure its best to get everything you want to do out on the table so you can figure out which things are least important and can be eliminated. As the summer progresses, I’ll be able to figure out exactly which things those are.

I thought it might be interesting to talk about some of the customizations I made to the P2 WordPress theme that I’m using.

True to the WordPress paradigm, I didn’t actually change anything in the P2 theme, but rather wrote a plugin to modify its behavior with filters and actions.

The most visible change I made was the addition of a “Sleep Counter” widget that displays the number of hours and minutes I’ve slept since March 18th. This actually relies on the the other parts of the plugin, though, so I’ll leave the discussion of this widget until later.

The main purpose of this customization is to keep track of the hours I’ve slept programmatically. I could have done this with post meta values, but that would defeat the purpose of P2 because I’d have to go into the back end to create posts. Instead, I decided to simply put the information in the post and parse it using the `wp_insert_post` action.

This hook is called whenever a post is saved, right after it has been written to the database. The function is passed the ID of the post that has been created/edited. What I do is get the post content, parse it to find the number of hours/minutes recorded, and save those values to two post metas (or update the post metas if I’m editing the post). The function is as follows.

function get_sleep_times($id) {
	$post = get_post($id);
	$content = $post->post_content;
	$matches = array();
	// pretty lazy regex matching here, but it works
	$matched = preg_match('/hours[^0123456789]*([0-9]*).*minutes[^0123456789]*([0-9]*)/i', $content, $matches);
	if ($matched < 1) // don't do anything if no sleep counts were posted.
		return;
	$hours = $matches[1];
	$minutes = $matches[2];
	update_post_meta($id, '_sleep_hours', $hours);
	update_post_meta($id, '_sleep_minutes', $minutes);
}

This code isn't perfect, but since I wrote it after about 24 hours without sleep and it is only intended to be used by myself, I'll give myself a break.

The next step was to create the widget to display my total sleep count. I hadn't created any widgets since the new widget API was released in WordPress 2.8, but a great article by Justin Tadlock set me on the right track. I added some customization options for how the sleep count is displayed, but the more interesting part is the function that actually displays the widget. It gets the options for the widget, then calls a `print_sleep_count` function that I defined.

function print_sleep_count($format = '') {
	global $wpdb;
	$ajax = empty($format);
	$format = $format ? $format : SleepCounterWidget::$default_format;

	$query = "SELECT COALESCE(SUM(meta_value), 0) as totalsum FROM {$wpdb->postmeta} " .
			"INNER JOIN {$wpdb->posts} ON {$wpdb->postmeta}.post_id = {$wpdb->posts}.ID " .
			"WHERE {$wpdb->posts}.post_status <> 'trash' AND {$wpdb->postmeta}.meta_key = '%s'";

	$hours = intval($wpdb->get_var(sprintf($query, '_sleep_hours')));
	$minutes = intval($wpdb->get_var(sprintf($query, '_sleep_minutes')));

	printf($format, $hours + intval($minutes / 60), $minutes % 60);

	// exit if this is requested via AJAX
	// if we don't, WordPress will add '0' to the output
	!$ajax or die();
}

You’ll notice some funky stuff with the `$ajax` variable. I’ll explain that later. Just ignore it for now.

The main magic here is the the SQL query. I decided to use a SQL query instead of the post meta API because I wanted get an aggregate of the post metas, and doing the aggregation from within SQL is much more efficient. I built a customizable query so I could use it to get both the number of hours slept and the number of minutes slept. I use sprintf to insert the meta key I’m looking for when I send the query to `$wpdb`. The reason I decided against the $wpdb->prepare() function is efficiency. The result would have been the same, but that function does some argument escaping that isn’t necessary because I know exactly what variables I’m passing in.

Once I get the total number of hours and minutes I’ve slept, I convert it to actual hours and minutes. This takes care of the situation where the sum of minutes is greater than 60 (one hour).

The final enhancement I made was some front end AJAX that updates the widget value whenever I create a new post. It’s pretty basic, and just utilizes the Admin AJAX API available within WordPress. The way I decided to do it was to utilize my existing `print_sleep_count` function. I simply hooked to an AJAX request event. That’s where the `$ajax` variable I mentioned earlier comes in. If the format argument isn’t passed to `print_sleep_count`, I assume it is being called via AJAX and exit at the end of the function.

You’ll also notice that I’ve got a widget on my main website that displays the same sleep count. It is mostly identical to the one on my Sleep blog, but the database tables are hard coded (because here `$wpdb->tablename` uses the wrong prefix). I also decided to have the sleep count link to my Sleep blog.

And that’s pretty much it. There are some enhancements I could probably make. For one thing, I could store the sleep count as one option in the options table to save myself from having to calculate an aggregate value every time, but with the number of visits I get each day, it’s not enough of an issue to worry about.

If you’d like to see what the rest of the plugin looks like, you can download it and check it out.

When working on different projects, I sometimes have problems with plugins interfering with each other (or at least making certain features difficult to test). I can always go though and enable and disable the correct plugins and themes to get the environment I want, but this can be tedious when there are lots of plugins.

Dev Profiles takes away some of this tedium. Create profiles to store snapshots of enabled plugins and selected theme. These profiles can then be enabled later on to bring back that same snapshot.

I find this useful when I’m working on several different plugins and/or themes and need to switch back and forth between them frequently.

If you think you might find this useful, you can check out my plugin page.

NOTE: I’m not currently doing anything about plugin activation / deactivation hooks. This may be a future feature, but right now neither of these hooks are called.

I’m using P2, a theme developed by Automattic, along with a plugin I developed to automatically parse and track sleep hours from my posts. The running total will be displayed on the sidebar of the new site, as well as here.

Right then, I’ll let you get to it.

Windows Phone 7 is Microsoft’s new offering in the mobile market. It hasn’t been released yet, but the demos look fantastic, and the development tools are available. After watching a few videos and seeing how easy it is to create an application for the WP7, I decided to download the tools and try it for myself.

Here is a screenshot of the result.

I only spent a few hours developing this application, so it isn’t fantastic, but I accomplished what I set out to do. I’ve created a functional Windows Phone 7 application!

If you want to see the source code, you can download my Visual Studio 2010 project here. Keep in mind that there isn’t excessive documentation for WP7 application development, so my design practices may not be perfect.

[Note] I fixed an issue that keeps the application from launching. If you can’t get it to run, try downloading it again.

You can learn more about Windows Phone 7 and download the developer tools at the Windows Phone 7 developer website.

Here’s a direct download link if you’re really curious.

The problem I ran into is that when NetBeans uploads that file and overwrites the old version, the permissions on the file change. I did some searching and it sounds like NetBeans doesn’t set permissions on the file, which means it just takes the server default. This doesn’t make much sense (should keep permissions of previous file) and for me it meant that the permissions were set so low that I couldn’t run the PHP scripts I was uploading.

It’s simple enough to change the permissions on those files.

chmod -R 777 /path/to/files

But you have to run this command every time you make a change. This is a royal pain when you’re making a lot of small changes (like when you’re trying to track down a bug for example).

One solution might be to try hooking into the ftp server daemon to run that command every time something is committed, but that sounded like too much work. I came up with a better solution.

Instead, I decided to just run that command in a loop. The following scripts runs infinitely (until execution is stopped) and waits 1 second, then runs the permission command.

while (true); do
    sleep 1
    chmod -R 777 /path/to/files
done;

If you want to be able to do anything else in the command line while this is running, it’s probably a good idea to run it in a screen.

Also, you can end the execution of this script with CTRL+C. Otherwise it will continue executing forever.

The following script takes a folder name as an argument, creates that folder in the website root (specified in a config variable of the script), downloads and extracts the latest version of WordPress into that folder, and runs the WordPress install. Finally, it sets the admin password to the password specified in another config variable and removes the password change nag.

#!/bin/bash
#
# configuration
#
rootURL="http://localhost" # your root URL. No trailing slash
rootDIR="/home/anderswc/public_html" # your document root. No trailing slash

dbname="wordpress"
dbuser="username"
dbpass="password"
dbhost="localhost"

wppass="password" #password to use for the admin account
adminemail="email@example.com" #email to use for the admin account
blogtitle=$1 # defaults to the specified directory name. can be anything

if [ -z "$1" ]
then
	# directory name must be specified
	echo "Please specify an installation folder name"
else
	# ideally you shouldn't have to change anything past this point
	echo "Creating a WordPress instance in $rootDIR/$1"
	# create the directory to install WordPress in
	mkdir $rootDIR/$1
	# download the latest version of WordPress into the directory we just created
	wget -o /dev/null -O $rootDIR/$1/wordpress.zip http://wordpress.org/latest.zip
	# unzip the archive containing WordPress. files will be inside a 'wordpress' directory
	unzip -q $rootDIR/$1/wordpress.zip -d $rootDIR/$1
	# move the files out of the 'wordpress' directory into their intended location
	mv $rootDIR/$1/wordpress/* $rootDIR/$1
	# remove the wordpress archive and the wordpress folder
	rm $rootDIR/$1/wordpress.zip
	rmdir $rootDIR/$1/wordpress
	# make WordPress directory writeable by PHP so installation can take place
	chmod -R 777 $rootDIR/$1
	# request the configuration file generation and blog installation webpages.
	postdata="dbname=$dbname&uname=$dbuser&pwd=$dbpass&dbhost=$dbhost&prefix=$1_"
	wget -o /dev/null --post-data "$postdata" "$rootURL/$1/wp-admin/setup-config.php?step=2"
	postdata="weblog_title=$blogtitle&admin_email=$adminemail"
	wget -o /dev/null --post-data "$postdata" "$rootURL/$1/wp-admin/install.php?step=2"
	# WordPress password was randomly generated
	# change it to password in configuration and remove password change nag
	sqlcommand="UPDATE $1_users SET user_pass=MD5('$wppass')"
	sqlcommand="$sqlcommand;DELETE FROM $1_usermeta WHERE meta_key='default_password_nag';"
	mysql -u $dbuser -p$dbpass -e "$sqlcommand" $dbname
	# uncomment the following line if you're running this script on a public server
	#chmod -R 755 $rootDIR/$1
	echo "Done"
fi

If you think this might be useful, give it a shot and let me know what you think.

If you’re thinking “OK, how do I actually use this?”, then this trick probably isn’t for you.

Whatcha think?

I’ve been following the development of WordPress 3.0 more closely than previous versions, and one of the features that has intrigued me the most is the new theme being developed. Twenty Ten is shaping up to be an elegant and innovative theme and I’m excited to see what it becomes before the final release of WordPress 3.0.

One of the nifty WordPress 3.0 features Twenty Ten takes advantage of is the new API for choosing a header and background image. In previous WordPress versions, theme authors would have to include complicated forms and form handlers if they wanted to allow their theme’s header and background to be customized. Now the theme simply has to indicate to WordPress that it supports custom headers and/or backgrounds and mention any images that are included in the theme. WordPress takes care of the rest, allowing administrators to choose between the available images or even upload an image of their own. Not only does Twenty Ten utilize this API, but it includes eight (so far) beautiful images to choose from.

The really nifty thing about having this standard API for custom headers/backgrounds is that it’s easy to customize. For example, I was able to write a plugin that randomly chooses between the available headers using a standard (to WP 3.0) filter. What’s neat is that this plugin should work with any theme that uses the new API and includes custom headers (though it’s hard to test since no other themes support currently do).

If you haven’t seen what Twenty Ten looks like, you can find my own instance here. I have the custom header plugin installed, so if you refresh the page the header image should change.

And that’s about it. If you want to check out my random header plugin, it has its own plugin page.

Plugin page (basically the same instructions as this post).

Direct Download

A problem I sometimes notice, though, is that people’s gravatars don’t always match the the mood of the conversation. There are two extremes. One could be discussing a serious subject, but have a gravatar that is jovial and humorous. This might not always be appropriate. On the other hand, they could be involved in a humor thread, and then have a serious gravatar that just doesn’t fit the situation.

You can, of course, have multiple emails associated with your Gravatar account, but who wants to set up an email account for every mood they might be in? Once again, Google Plus Addressing comes to the rescue.

So what you can do is to map different gravatars to the same email, with different plus addresses.

Here are some examples I just create (note: I took two of the pictures with my phone, and yes, my hair is getting long).

will+happy@itsananderson.com

will+sad@itsananderson.com

will+surprised@itsananderson.com

This is a boring example, but hopefully you get the idea.

Unfortunately, you still have to verify each email on Gravatar, even though it goes to the same inbox. Wouldn’t it be great if Gravatar let you do this automatically? Just upload a picture and associate it with an email, then choose a tag that can be used to access that picture. I don’t think it would be that hard, and it would be totally awesome to use!

Go ahead and grab the new versions of Code Snippets and Short Comment Filter.

If you’re familiar with ASP.NET you probably know that every page has a presentation layer (a lax form of XML) and a business layer (C# or Visual Basic). This is a brilliant way of organizing a website because it separates your data (the presentation layer) from your complex logic (the business layer). This gives you several advantages (not by any means an exhaustive list):

1. Complicated logic is easier to understand when it isn’t cluttered with HTML tags
2. Because the presentation and the business layers are loosely linked, changes in one often don’t require a change in the other
3. Object oriented principles can be used in both the presentation, and the business layer, promoting code reuse

Now that I’ve sung the praises of the ASP.NET paradigm, it’s time to talk about one of the challenges it presents: AJAX calls, specifically JSON responses. If you’re not familiar with JSON, the idea is that rather than passing information as plain text or XML, it can be passed along in “JavaScript Object Notation”. In plain terms, this means that you pass a string which can be parsed to create a JavaScript object. Because JavaScript treats associative arrays and objects essentially the same, meaning that an array can be used to represent a JavaScript object. Confused yet? Here’s an example. For the sake of simplicity, I’m just setting the object string directly. Normally it would be the result of an AJAX request.

var json = '{ name : "Joe Smith", age : 23, job : "Consultant" }';
var func = Function('return ' + json);
var jsonObject = func();
alert(jsonObject.name); // Joe Smith
alert(jsonObject.age);   // 23
alert(jsonObject.job);    // Consultant

Ok, so now we’ve got the ASP.NET and JSON explanations out of the way (if you were already familiar with these technologies, thanks for sticking with me). Here’s the challenge I ran into, and how I got around it.

Let’s suppose you’re using ASP.NET and C# to run your website, and you have a directory of users. For this example let’s suppose that you have a page which lists the usernames for every member of your site. Clicking a user’s username should bring up their other information (in this case, their name, age, and job). Being an eager developer, you decide to display this information via JavaScript and AJAX because that seems like the Web 2.0 thing to do. So you’ll make a request to another ASP.NET page and parse the response as JSON. This is great, except that ASP.NET pages are generally served as HTML (which in a stretch could be treated as XML). Not exactly your optimal solution, especially if you aren’t comfortable with traversing a DOM tree. Fortunately there’s a really simple solution. Leave everything out of your ASP.NET page except for the tag to include the C# file. Then in your C# file you can print the JSON information directly. Let’s look at an example.

The first file is GetUserInfo.aspx, which is very basic.

< %@ Page Language="C#" AutoEventWireup="true" CodeBehind="GetUserInfo.aspx.cs" %>

The second file is the GetUserInfo.aspx.cs, which is also fairly simple. I’ll just show the Page_Load function for brevity.

protected void Page_Load(object sender, EventArgs e)
{
    String json = "{ ";
    //
    // Get values from database and add them to the json string
    //
    json += " }";
    Response.Write(json);
}

And that’s all we have to do for this page (except for getting the values from the database, which is straight forward). I’ll note that there are actually libraries for generating JSON strings, but I left them out for the sake of simplicity. To learn about available libraries (and JSON in general) check out www.json.org.

Now I only talked about my experience with JSON and ASP.NET. I’m by no means claiming that this is the best solution, or that JSON is even how you’ll always want to receive AJAX responses. This is simply what worked for me. If you know of some reason this isn’t a good idea, or of a better way of doing it, feel free to point it out in the comments.

Since my last post, the spring quarter at Rose has (finally) ended. I love school, but toward the end of the quarter my lack of sleep was really starting to wear me down. Now that I’ve had time to recuperate, I’m already excited for next year.

Work

I started work at ANGEL Learning June 1^st and have been loving it so far. In the 2 and a half weeks I’ve been at ANGEL, I feel like I’ve learned as much about the various aspects of software development as I would in an entire class. ANGEL was acquired about 5 weeks ago by Blackboard, a company that offers a competing product. Being able to see the way this transition occurs from a “safe”, low stakes position has been, and I’m sure will continue to be an invaluable experience.

Conferences

A little over a month ago, I had the opportunity to attend an ANGEL Users Conference on the coat tails of the RosEvaluation tool. This was about a week after ANGEL’s acquisition by Blackboard. It was interesting getting to see the different customer’s reactions to the acquisition.

During the conference, I got to give a talk about some of the things our team had learned while extending ANGEL with the RosEvaluation tool. At the end of the conference, Will Mathies, Timothy Chow, Kevin Davidson, and I received an ANGEL Impact Award for our work with RosEval. If you’d like, you can read the news article Rose put together about it.

I also got to attend WordCamp Columbus, where I met Jeff Chandler of WP Tavern as well as a handful of other well known people (the event had quite a few attenders, I’m just not very outgoing, so I didn’t meet most of them). I had a great time, and came away with a few neat tips and tricks (see previous post). Even though it’s a bit of a drive getting to Columbus, I feel it was well worth the effort and I hope to attend the next one if my schedule permits.

Family

Both frightening and exciting events have taken place in my family in the last month or so. While I was at the ANGEL Users Conference, my younger sister was rushed to Riley Children’s Hospital after slipping into a diabetic coma. She recovered fairly quickly and was able to go home, but the last month has been a trying time for everyone (especially her). Taking care of her diabetes has required some big schedule adjustments (and lots of needles).

On a more up beat note, I’m now the uncle of a second nephew. My older sister and her husband are the proud parents of their first born, Stephen Mathew Agapie.

The End

And that just about wraps it up. I’m working full time this summer, as well as working on a few side projects, but I’d really like to start posting more (both personal posts, and more technical fare). So if I go very long without posting anything (I’d say a week is probably a reasonable limit) by all means bug me about it.

First of course you need to download WordPress and then upload it to your server. This is the easy part, and there are lots of instructions for how to do this, so I won’t go through them. You’ll also need a MySQL database, but again I won’t go into that since there are lots of resources already available that explain how to do this.

Let’s suppose we’re installing two blogs on domain1.com and domain2.com (lame I know, but give me a break). These should be pointing at the same server (your domain registrar can help you with this). If you navigate to your WordPress install, you’ll be asked to go through the installation process. Let’s start with domain1.com. Enter all your database information, but specify a custom table prefix (like “domain1_”) which will be unique to that domain. Once the installation is complete (make sure you save your password!) you need to delete the wp-config.php file.

Now if you navigate with your other domain, you’ll be asked to install WordPress again. This time, use a different table prefix (like “domain2_”). This will give you two different sets of WordPress tables in your database. The only problem is that you can’t access both of them. We’ll remedy this next.

Go into your wp-config.php file and find the line that looks something like this.

$table_prefix = 'domain2_';

We need this variable to be different, depending on what domain is being used to access your blog. Using the PHP $_SERVER super global, we can do this quite easily.

Here’s the new code you’d want to use.

if ( 'domain1.com' == $_SERVER['HTTP_HOST'] )  {
$table_prefix = 'domain1_';
} else {
$table_prefix = 'domain2_';
}

So now your server will display two different blogs, depending on what domain is used to access your site. Again, the advantage here is that you only use the space of one set of WordPress files, and you only have to do one update when a new version is released.

Mike talked about some other cool things you can do with this, and talked about how you can address issues with plugins like Google SiteMaps. He promised to post his slides, so if you’d like to learn more you should check out his blog.

New Features

• Filtered Comment Count – Look at the settings page to see how many comments have been filtered
• Customizable Filter Message – Specify a custom message to be displayed if a comment is too short. Use %type% and %length% for a dynamic message

Fixes

• Email Notification – Notification emails are no longer sent out for filtered comments
• Filtering Trackbacks – Trackbacks are no longer filtered. Only comments are required to meet the minimum length requirement

Again, here’s the download link

Here’s a simple example of the “old” and “new” ways of retrieving options from the database:

old:

$option = get_option( 'Option Name' );
if( false === $option ) {
$option = 'My real option value.';
}

new:

$option = get_option('option_name', 'default');

As you can see, this simplifies option retrieval and makes your code just a little easier to read.

EDIT: Thanks Nick Ohrn for pointing out a bug in the first code snippet. I’ve updated it so nobody is confused.

It automatically spams or deletes comments that are too short.

You can view more information about the plugin and download it here.

You can learn more about CSS Naked day here.

If you like my blog design, I’d love your vote! You can see the a list of the designs that were submitted, and vote on your favorite here.

As with just about anything in WordPress, you can use Actions and Filters to customize the way WordPress RSS Feeds. I chose to filter the query because it already has built in category filtering functionality. First I hooked into the query filter.

add_filter('pre_get_posts', 'filterRSSQuery');

Then I created a function called ‘filterRSSQuery’ which does just what the name suggests.

function filterRSSQuery($query) {
	if ( $query->is_feed ) {
		$query->set('category_name', 'News');
	}
	return $query;
}

The code is pretty straight foreward, but here’s a breakdown. First we test to see if the query is for a feed (we don’t want to do this filtering anywhere else). Then we set the ‘category_name’ argument to ‘News’ (you would change this to whatever category you want to show up). Finally, you return the modified query object.

Overall, this is pretty simple, but extremely powerful.

I’ve decided to enter the contest in the Retro&Vintage category. Why did I decide to enter? My design certainly isn’t the most beautiful out there, so I’m not “expecting” to win, but this might give me a chance to get some constructive critisism about design choices I’ve made. Besides, it can’t hurt, and there’s a prize

Do you have a WordPress blog that you’ve designed? Enter it in the contest.

Good Luck!

The website actually went through several designs before I settled on the one it currently has (I might release the other themes at some point. We’ll see). I eventually created a design that I really liked, and that’s basically what’s there today.

The work I’ve been doing on the site is one of the reasons I’ve been so quiet this past month (My course load also increased, but that’s another story). When the quarter ended, I had enough time to get all the kinks worked out of the web site, so now it’s open for business. Hopefully I’ll be getting enough business to continue to be short on time, but I still plan to post more in the months to come.

Many people fight the spam bots by obfuscating their emails (for example “joe [at] example [dot] com” or “joe@REMOVETHISexample.com”). This does an OK job, but it’s also a royal pain for the visitor, who has to type in, or fix your email address if they want to send you an email. The ultimate spam fighting question arises: “How do you keep bots out while letting real users in”. GMail’s plus-addressing feature and jQuery might be able to bring us one step closer to an answer.

If you use Google’s email services (GMail or Google Apps), you have a powerful email feature available. Plus-addressing means that bill@gmail.com, bill+something@gmail.com and bill+anything@gmail.com all point to the same email address. When deciding what address to deliver the message to, Google drops everything after the plus sign. you can then actually use filters to move items to folders inside your mailbox (for example, you could use example+bank@gmail.com for your banking). Doing this also makes it just a little harder (though not impossible) for bots to get your real email address.

By default, you probably don’t want to display a “plus address” on a contact page, jQuery comes to our rescue here, allowing us to remove the “plus” part for users with JavaScript support. The following script looks for all email links and removes the “plus” portion from the href and the inner text.

jQuery(document).ready(function(){
	jQuery('a[href^=mailto:]').each(removeSpamProtection);
});

function removeSpamProtection(index, element){
	var link = jQuery(element);
	link.attr('href', link.attr('href').replace(/\+[a-zA-Z]*/, ''))
		.text(link.text().replace(/\+[a-zA-Z]*/, ''));
}

Now all visitors with JavaScript support will see something like <a href=”mailto:example@example.com”>example@example.com</a> while visitors without JavaScript support will see <a href=”mailto:example+spamblock@example.com”>example+spamblock@rose-hulman.edu</a>. Normal visitors will see your real email address while bots and visitors without JavaScript will see something else. The bots will (hopefully) be confused by the plus sign. If not, they will be sending an email to an address that you can easily filter into a “maybe spam” folder.

So there you have it. A very short piece of JavaScript (thanks to the wonders of jQuery) that you can include in your WordPress theme or install as a separate plugin.

What do you do to protect your email online? Do you have another method that works better? If so, please share it in the comments!

Normally you’d call wp_page_menu to get a list of Pages, as well as a link to the home page (providing the home page is not set to a static Page). This function was added in WordPress 2.7, so if you use it in a theme and someone with an older version of WordPress installs that theme, they will get an error. One solution is to check that the function exists before using it, but this gets a bit messy, especially if you use several functions that are new. What do you do when WordPress 23.6 is released? Everyone will most likely have a version of WordPress that supports wp_page_menu, so the check will no longer be necessary. If you are using several functions, you’ll have to go back and fix them in several places throughout the template. This isn’t optimal.

What I’m proposing instead is defining some default action if a given function doesn’t exist. These special definitions can be placed either directly in functions.php, or can be separated into a file specifically intended for them (new_functions.php for example). That way when they’re no longer needed they can easily be removed.

Anyway, I think it’s pretty cool.

Thankfully you can provide a much simpler solution for people who don’t want to edit your templates. With this solution, users can edit the 404 page contents just like it were any other page. Here’s how it works.

To use the WYSIWYG editor, the user actually needs to add a page. Of course they don’t want “404″ showing up everywhere they list their pages, so they need to set the Visibility to “Private”. This will remove it from all page listings. The permalink also needs to be specified explicitly. Using “404″ makes sense to me, and is what I use.

Now in the 404 template we want to check whether a 404 page is defined and if it is, display it. Otherwise we should have some default content to display. Here’s a template 404 file which lays out the general idea. Some tweaking might be required to fit the rest of your theme.

< ?php get_header(); ?>
	<div id="content" class="narrowcolumn">
	<?php query_posts('pagename=404' ); ?>
	<?php if (have_posts()) : the_post(); ?>
		<div class="post" id="post-<?php echo $id; ?>">
			<h2 class="post-title">< ?php echo $post->post_title; ?></h2>
			<div class="entry">
				< ?php the_content(); ?>
			</div>
		</div>
	< ?php else : ?>
		<!-- your default 404 content -->
	< ?php endif; ?>
		<!-- search form, or anything else you always want to show at the bottom -->
	</div>
< ?php get_sidebar(); ?>
< ?php get_footer(); ?>

Notice line 3 where we query for the post with page name “404″.

In line 4 we check to see if the “404″ page actually exists and if it does, we load its information with the_post().

In line 6 we use $post-the_title instead of simply the_title(). This is because WordPress adds “Private: ” to the beginning of the title of private pages when you call the_title().

The rest is pretty straight forward. Place any default content (what you want to show if there is no 404 page defined) inside the else and then anything you want to always display (whether the user has created a ’404′ page or not) after the endif.

That’s it! If you distribute a theme with this feature, just let people know what you’ve done so they can utilize it.

Private Pages can solve to a wide variety of problems. I hope to post more ways to use them soon. For now if you have any other ideas about how they might be used, or a suggestion for improving this example please add it to the comments

Associative arrays are one of my favorite PHP features. They’re simple to use and easy to understand. Sometimes though, being able to reference the values of associative arrays with variables can be more convenient. Today I’m going to share a simple way to convert an associative array to variables.

OK, if you’re not sure what we’re trying to accomplish here, take a look at the following example.

$list = array( 'var1' => 'value1', 'var2' => 'value2' );

To reference the elements in the array, you’ll have to use the array key like this.

echo $list['var1']; //prints 'value1'
echo $list['var2']; //prints 'value2'

What we’d like to do instead is reference the values using variable names like this.

echo $var1; //print 'value1'
echo $var2; //print 'value2'

This is possible through what the PHP Manual calls ‘variable variables’. The idea is that the value of one variable is used as the name of another variable. Here’s an example.

$hello = 'world';
$$hello = 'howdy';
echo $world; //prints 'howdy'

Here, the value of $hello is analyzed and used as the name of the variable to which 'howdy' is assigned.

Now it’s not a big step to expand this to an associative array. Consider the following code.

$list = array( 'var1' => 'value1', 'var2' => 'value2' );
foreach ( $list as $key => $value ) { $$key = $value; }
echo $var1; //prints 'value1'
echo $var2; //prints 'value2'

So we’ve accomplished what we set out to do. I must caution you, though, about a couple of things which might trip you up.

First, make sure the keys are valid variable names. If you don’t you’ll run into problems when you execute the script.

Second, be very careful when using this with user input. For example, the following code is insecure.

// login.php
// checks the username and password in the GET request
$realuser = 'me';
$realpass = 'secret';
foreach ( $_GET as $key => $value ) { $$key = $value; }
if ( $user == $realuser && $pass == $realpass ) {
 echo 'login correct';
} else {
 echo 'login incorrect';
}

There are several problems here. First, we don’t know that $user and $pass have been initialized. We should check this before using them. Second (and more importantly) a cleverly generated request could let a user log in without a correct username/password. For example:

login.php?user=me&pass=wrong&realpass=wrong

When the script runs through the list of variables, it will assign a new value of wrong to $realpass. Because $realpass and $pass have the same value (as do $user and $realuser), the page will allow them to log in.

Here is a way to keep things a little safer by prefixing all the $_GET[] values.

// login.php
// checks the username and password in GET request
$realuser = 'me';
$realpass = 'secret';
foreach ( $_GET as $key => $value ) { ${'get_'; . $key} = $value; }
if ( isset( $get_user ) && isset( $get_pass )
    && $get_user == $realuser && $get_pass == $realpass ) {
 echo 'login correct';
} else {
 echo 'login incorrect';
}

Now the $_GET[] values are prefixed with get_ so they can’t interfere with other variables in the script. Additionally isset is used to ensure that the values of the variables has actually been set.

To use something like this, you’d still need to check for invalid variable names, but I won’t go into that for this example.

The easiest way to facilitate this is to create Apache aliases to these different locations. Wamp Server provides an interface for creating new aliases by clicking on the tray icon and selecting Apache -> Alias directories -> Add an alias. A new command prompt window opens where you can enter the alias information.

Creating an alias like this, you’ll probably notice that in order for the URL to work, you have to have a trailing slash at the end (e.g. http://localhost/example/ will work while http://localhost/example won’t). This can be a bit of a pain, but there is fortunately a quick fix. Click on the Wamp Server tray icon again and select Apache -> Alias directories -> [alias url] -> Edit alias. This will open the alias file in Notepad. Remove the trailing slash from the relative URL.

For Example

Alias /example/ "c:/path/to/example/"

Would become

Alias /example "c:/path/to/example/"

Save the file and close Notepad. Wamp Server should restart automatically (if not, you can restart it manually from the tray icon). Once it has restarted, you’re alias should now work with or without the trailing slash. Why Wamp Server doesn’t do this automatically, I’m not sure, but at least there’s a simple fix.

Not surprisingly almost every large holiday sees a spike around the date it occurs. Here are some of the ones I looked at (letters indicate related news articles. You can look on Google Trends to see what they are).

Christmas Search Trends

Easter Search Trends

New Years Trends

 As I said, this really isn’t all that surprising. There are a couple of interesting things to see here, like the fact that searches for Christmas have gone down over the years (decreased interest or fewer people who don’t know what Christmas is?). The really interesting stuff to me is where these searches originated.

Christmas Search Distribution

Easter Search Distribution

New Years Search Distribution

Lets take a look at this. Christmas has an approximately equal share between the UK, US, and Ireland (?!). Australia is fifth in the list. If you look at Easter and New Years though, Australia is ahead by quite a bit (and both times followed by New Zealand). This makes me wonder, is Christmas not as popular down under, or do not as many people do research about it as other holidays?

Now in keeping with the general topic of this site of late, I’ll show another trend… WordPress

WordPress Search Trends

WordPress Search Distribution

 First it’s cool to see that WordPress continues to grow in popularity. Second, I’m surprised to see that the first 6 countries here aren’t really in “The West”. Again I wonder whether these search statistics also reflect usage. If so, that says something interesting about WordPress’ demographic (kinda cool that it’s really spreading internationally). Of course I also wonder if splog activity has anything to do with these statistics.

So what do you think? Any of this surprise you?

See anything interesting here that I missed or find an interesting Google Trend you’d like to share? Let me know in the comments and I might add it here.

College Acceptance

Because I was home schooled and was graduating a year early from high school, I was unsure whether I would be accepted at Rose-Hulman. Thankfully I was. This was a great blessing for the following reasons (not an exhaustive list):

1. As a child of a Rose professor, I get free tuition (this of course is the biggest blessing).
2. Going to Rose meant I could live at home. I’m an independent person, but being able to live at home has really been great because I’ve been able to save quite a bit of money not having to pay rent.
3. Rose is one of the best undergraduate engineering schools in the US. An education from Rose gives students a jump start upon graduation.

I feel that God has really brought all of this about through (just) enough intelligence to learn, parents who gave me a stellar education, and many other blessings to numerous to count.

Job Opportunities

Toward the beginning of this year I had the opportunity to help IRPA as they transferred their static website to a WordPress based CMS. Beyond introducing me to WordPress, my work with IRPA gave me a connection which has proved pivotal to to several other work opportunities that have presented themselves to me during the past year.

About the time I began working for IRPA, I also applied at Rose-Hulman Ventures for a summer internship. My connection with Dr. Williams (executive director of IRPA) helped me gain a position in the RosEval project (a project sponsored by IRPA/Rose-Hulman). Exceeding my hopes for a summer position, I was able to begin working for Ventures well before the end of the spring term. Working for Ventures, I was introduced to a multitude of things. I learned a lot about C#/.NET and Microsoft SQL. I learned a bit about general database design practices, project management practices, how to interact with clients, organizing meetings… Like I said, a multitude of things. I also was able to participate in the Leadership Academy hosted by Rose-Hulman while working at Ventures. This was an amazing experience for me. I learned a lot, got to know some people, and strengthened my confidence.

Now, RosEval is a plugin for ANGEL, a learning management system developed by ANGEL Learning (it is also the learning management system of choice at Rose-Hulman). While developing RosEval, I was able to get a look at some of the inner workings of ANGEL. Well, the summer ended and my Freshman year began. A few weeks in, I attended the Career Fair that Rose hosts each term. I only had about 30 minutes to talk to companies, but as I was walking past the booths, I noticed that ANGEL Learning had a booth up. The thought crossed my mind “Hey, I have development experience with ANGEL, I wonder if I could get an internship with them.” As I thought about it though, I decided that they probably wouldn’t be interested in a Freshman, so I moved on. After walking around for a little while and not finding anything terribly interesting, I decided to go ahead and talk to the ANGEL people. “What’s the worst they can do, turn me away?” I thought. I had a nice chat with Tim Tomlinson, gave him my resume and told him about my work with RosEval. He told me my work experience seemed promising and said they might bring me out to Indy for an interview within the next view weeks. Dumbfounded, I walked away, thinking “Wow, OK. Um, Yeah. Didn’t expect that”. Logic kicked in and I though, “Maybe he was just being nice.” As it turns out though, he was serious. I got a call from ANGEL a couple weeks later offering to bring me to Indy for an interview. I went to the interview, still wondering whether they would even be interested in a Freshman. The interview went well, and I was offered an internship for next summer (EDIT: Which I accepted! Thanks, Nick, for pointing out that I’d missed this detail). I’m not sure I would have been offered a position if I hadn’t worked on RosEval last summer. Again, everything seemed to “fall into place”.

OK, that section was a little longer than I was planning, but I think it’s really cool what God has done to provide me with work opportunities.

Beginning School

As I mentioned before, I was accepted at Rose and begin my freshmen year this fall. I’ve gotten to know many new people, learned a ton of new things, and had several opportunities to share my faith. The community at Rose is really an amazing thing and I’m blessed to be part of it.

Starting This Blog

While there are no miraculous stories associated with this site, I think it has been a great experience for me. I’ve created about 50 posts since the beginning of the school year and had 350 unique visitors (and counting). While these numbers certainly aren’t impressive, it’s encouraging that people are reading what I have to say (and that some of them are even coming back). I’ve gradually been increasing the rate at which I post and I hope to be able to write even more frequently in 2009. The theme I created for this blog has gone through several iterations and I’m pretty pleased with the way that it turned out, but I think I might do a total redesign for 2009 (If you like this theme though, let me know and I might keep it). I’ve gone from a WordPress novice to someone who, while not exactly a “professional”, is at least competent enough to get things done and to help friends who are involved in the blogging world.

In Conclusion

I can say without a doubt that this has been one of the best years of my life. It’s been amazing to see what God has done and I’m excited to see what he has in store for 2009.

Happy New Year!

I went about creating these empty index files on my site, but soon discovered that there are hosts of folders that could use them (think about the date structure in /wp-content/uploads ). Then I remembered another tool. .htaccess. Add this line of code to your .htaccess file to keep file in any directory from being listed (this works on any site who’s server supports .htaccess by the way, not just WordPress blogs).

IndexIgnore *

Really simple, I know, but I thought I’d share it all the same.

Check out the info and download it here.

If you have comments or suggestions, by all means let me know (either by email or by commenting here)!

In my work for IRPA one of the concerns we’ve had is the ability of visitors to navigate the site. Because WordPress isn’t really a CMS out of the box, navigation is sometimes difficult to facilitate on sites with deep “trees”. One of the solutions I looked at was Breadcrumbs because they can give users a quick look at their location in the site structure. Over the break (which I’m still enjoying) I’ve been able to finish implementing breadcrumbs on the IRPA Site.

I almost think something like this should be in core, but again, WordPress isn’t a CMS, so maybe it wouldn’t be useful for most people.

One of the things that took me the longest was figuring out how to handle different site settings. A site who’s home page displays latest posts needs a slightly different navigation list than a site which has a static home page. Even sites with static pages can have different structures depending on what page is the “home” page. Anyway, Here’s the code if you want to add it to your template. Simply paste the code into your functions.php file and add a call to get_breadcrumbs() where you want the breadcrumb list in your template. It outputs the breadcrumbs as list items. You’ll need to put the ul around them yourself in the template.

function get_breadcrumbs(){
	global $post;

	$separator = '  > '; // what to place between the pages

	if ( is_page() ){
		// bread crumb structure only logical on pages
		$trail = array($post); // initially $trail only contains the current page
		$parent = $post; // initially set to current post
		$show_on_front = get_option( 'show_on_front'); // does the front page display the latest posts or a static page
		$page_on_front = get_option( 'page_on_front' ); // if it shows a page, what page
		// while the current page isn't the home page and it has a parent
		while ( $parent->post_parent && !($parent->ID == $page_on_front && 'page') == $show_on_front ){
			$parent = get_post( $parent->post_parent ); // get the current page's parent
			array_unshift( $trail, $parent ); // add the parent object to beginning of array
		}
		if ( 'posts' == $show_on_front ) // if the front page shows latest posts, simply display a home link
			echo "<li class='breadcrumb-item' id='breadcrumb-0'><a href='" . get_bloginfo('home') . "'>Home</a></li>\n"; // home page link
		else{ // if the front page displays a static page, display a link to it
			$home_page = get_post( $page_on_front ); // get the front page object
			echo "<li class='breadcrumb-item' id='breadcrumb-{$home_page->ID}'><a href='" . get_bloginfo('home') . "'>$home_page->post_title</a></li>\n"; // home page link
			if($trail[0]->ID == $page_on_front) // if the home page is an ancestor of this page
				array_shift( $trail ); // remove the home page from the $trail because we've already printed it
		}
		foreach ( $trail as $page){
			// print the link to the current page in the foreach
			echo "<li class='breadcrumb-item' id='breadcrumb-{$page->ID}' >$separator<a href='" . get_page_link( $page->ID ) . "'>{$page->post_title}</a></li>\n";
		}
	}else{
		// if what we're looking at isn't a page, simply display a home link
		echo "<li class='breadcrumb-item' id='breadcrumb-0'><a href='" . get_bloginfo('home') . "'>Home</a></li>\n"; // home page link
	}
}

If you have any corrections or suggestions for this (rather long) bit of code, please add your comment bellow. I love feedback!

By the way, I’m currently listening to the Bitwire Holiday Marathon. If you’d like to listen to Jeff Chandler talk about non-WordPress related topics you should check it out.

http://holidaymarathon.com/

By the way, I found out about the new function here. If you’d like to learn about other new features WordPress 2.7 (or how to use wp_page_menu) you should definitely check it out.

WordPress gives users two tools for structuring pages. One is page parent, which allows you to create a tree structure with your pages, much like a folder system on a computer. I’ll touch on this a little later.

The other tool provided by WordPress is the ability to specify “Page Order”. What does this mean? Well, take my site for example. I have several pages that are listed at the top of every page. Their order can be decided by several things. Page ID is one option (though not really useful, since the ID is meaningless to anyone but WordPress itself). Another option is Alphabetical ordering based on title. This can be useful, but it still doesn’t give the kind of control that most would probably like. The final option is “Page Order”. This is based on a numeric value that can be set while editing a page. As WordPress says, this is a bit janky, because you have to go into every page and set some numeric value for its order. This is where My Page Order comes in.

Description

When activated, My Page Order adds an admin page in the “Pages” category called “My Page Order”. Clicking on it gives you a list of the pages you have on your site. You can drag the page names to order them just like WordPress Widgets. If you have a tree structure, you can select a page to order its child pages.

Likes

I love the ease with which you can order pages using this plugin. It’s a great tool to install for users who would feel uncomfortable using the default numbered ordering system.

Dislikes

The way this plugin handles the tree structure of pages feels a little awkward to me. Being able to view the whole structure at the same time would be incredibly useful, and being able to drag pages between parents would be immensely helpful.

I’ve still not decided whether this is a feature, or a bug, but unpublished pages are listed along with published pages. On one hand, this is good, because it means you can decide where a page will show up before you publish it. On the other hand, it can lead to a little confusion and clutter if many unpublished pages exist. Being able to hide unpublished pages would be a great improvement.

Overall Review

This plugin is really one of my favorite tools. Its function is so simple, yet it is integral to using WordPress as a content management system.

Plugin page on WordPress.org

Author’s plugin page (a bit outdated).

Check out the Release Announcement.

Download Page

EDIT: Oh yes, and here’s the dev blog announcement and the direct download link.

The first thing I had to do was find a theme that would change my interface to look like Ubuntu’s “Human” theme. I found a great (though slightly old) post that listed some of the options available. I patched the windows dll and (rather nervously) restarted Windows. After trying a few others, I finally settled on the first one on the list. (Its interesting to note that the Cursor files didn’t seem to install automatically, but doing so manually wasn’t that hard)

Next I went ahead and set the default Ubuntu Intrepid Ibex Wallpaper as my background to add to the Ubuntu atmosphere.

One of the things I love the most about Ubuntu is its Virtual desktops. After looking around, I found VirtuaWin, an open source virtual desktop manager for Windows. I tried it for a while, but while I loved its feature set, one thing I missed was the cool “Rotating Cube” effect that Compiz Fusion gives you in Ubuntu. I did a little more digging and came across Yod’m3D. As it turns out, this project is now DeskSpace, and is no longer free. Fortunately for me, an older version of Yad’m3D is still available for download. I installed it, and am now working in full Ubuntu-like glory.

Here are some screenshots:

“Development” Desktop

“Browser” Desktop

“Graphics” Desktop

Cube Transition Effect

If I decide to keep this configuration, I’ll probably purchase the latest version of DeskSpace so I can enjoy the advanced feature set, but for now the free version is plenty of fun to play with.

The first thing I want to test is ad revenue in general. Can a low traffic site like mine even make anything from text ads. If so, how much. I suspect that the answer to this question will be “Yes, but very little”, but we’ll see.
Second, I’m curious about visitor response. These days I get about 6 or 7 new visitors a week and I have about 5 or 6 people who read my blog frequently. Those numbers certainly aren’t very impressive, but they’ll do for the experiment. What I’d like to see is whether there’s a change in how long new visitors stay on my site, and if the number of people who actually return will be different.

I’m going to run this experiment for 3 months, and then at the end of that time I’ll do a report on the results. Whether I keep the ads will depend on the results of the experiment.
Now its up to you to give me my first bit of feedback. Did you even notice the ad module before reading this post? If so, did it affect your opinion of this site in any way? If you’re a new visitor, will the ad have an affect on whether you return?

By the way, I’m writing this post with “QuickPress“, one of the many new features in WordPress 2.7.

[EDIT] It turns out this project is a lot older than I thought (as in over a decade older!). I assumed it was a lot younger because the twitter feed only goes back a couple of months. You can find out more about the project here.

Of course, this is 6 months away, but I’m already looking at some of the housing options available. There’s a nice housing community called Deercross that’s only a couple miles from where I’ll be working. Their website certainly paints a pleasant picture, and their pricing seems pretty reasonable). As the time draws closer, I’ll start looking more actively and I’ll probably be visiting the places I find online. Overall, this should be an interesting adventure.

http://rubiqube.com/8-common-sense-tips-for-writing-clean-wordpress-blog-posts/

Though some of it was kind of obvious to me, the other parts were really helpful. I’m planning to begin some more intensive training of IRPA staff once the stable version of WordPress 2.7 is released, and I get the feeling this article will find its way into the training.

• I have a project due tomorrow, as well as an exam,
• I have a lab writup due Wednesday,
• I have a project I’m trying to finish by Thursday, and
• I have an interview with ANGEL Learning on Friday.

Hmm, maybe I’ll be busier this week than I thought.

Oh well, I guess that’s just part of being a student at Rose.

“… And even though we view the conservative lifestyle as an unusual way of life, we’ll still let Republicans marry each other.”

A quote from the Rose Thorn expressing the author’s opinion that the election outcome is good for everyone.

What I see (in the whole article, but especially in that sentence) is a complete moral turnaround from the values America held such a short time ago. Not only are those who believe homosexuality is wrong mocked, but the very heterosexual lifestyle is deemed as strange and unusual. We’re the one’s who need to be tolerated.

If that doesn’t scare you, I doubt anything ever will.

As for me, seeing the moral changes the past decade has brought about, I tremble to think what the next decade might bring.

My excitement and optimism ends here.

Our country seems on the verge of economic collapse, reminiscent of the Great Depression of the 20′s. Stocks are at all time lows, jobs are disappearing like rubber balls at a magic show, and foreclosures and bankruptcies seem imminent. Into this picture steps a man who promises hope and change. We don’t have to settle for the status quo. We can affect our future, and we can do it together. Our nation, looking for a savior in our time of trouble turns to this voice of hope and promise. We want change, and we feel that we’ve found the person who can bring it to us.

What concerns me about this change is both the what and how. I’m not opposed to change. I feel that change can be good on many levels, but I’m choosy about my change. One place I enjoy seeing change is in software. I often install beta versions of software before the stable releases are made because I’m excited to see what changes have been implemented (I’m writing this post from Google Chrome for example). If, on the other hand, I install the beta version of a program and I find that some part of it majorly degrades my experience, I’m very quick to go back to what I was using before. Again, I’m not opposed to change, but I don’t like change just for the sake of it.

Let’s analyze some of the ‘whats‘ and ‘hows‘ shall we?

What: New health care plans including expanded eligibility for Medicaid and Government funded insurance.

How: This new plan is estimated to cost the federal government approximately 1.17 trillion dollars. Federal health care costs will raise to around 2.7 trillion dollars a year by 2010. This cost will ultimately fall to those of the working age (about 60% of the population). We’ll assume that other plans have created jobs for everyone and that there is no unemployment, so everyone who can work is working. With about 60% of the population, or 180 million people paying for the cost of federally funded health care, each person will be paying just under 15 thousand dollars a year. Now the plan suggests that these costs will actually be paid by large companies, but ultimately, this will simply mean smaller pay checks for employees… or that companies will have to cut back on jobs and unemployment will be created. I’ll leave it to the reader to determine whether this cost is worth the “benefit” of government funded health care.

What: More research, innovation, and jobs.

How: The government will support company and university research to create new technologies and innovations, creating jobs and increasing the standard of living for American families. Another part of this plan promises publicly available broadband access in every community across the nation. While this will give rise to technology innovations and hopefully to an increased number of jobs, it will ultimately cost us more than if these services were provided by a private investor. The reason for this is that the administration assigned to these projects by the government usually has little to no experience in the fields they are investigating. They are administrators, not scientists or engineers. Because of this, it’s harder for them to determine the feasibility of a project proposal or innovation idea. Private investors, on the other hand, seem to have more people who are experienced in the science and engineering industries, so they are better able to determine whether a project will succeed or fail. Only investing in projects that are likely to succeed cuts back on costs and delivers cheaper products to the customer. Lower production costs will also allow employers to hire more workers, increasing employment.

What: Civil rights reform.

How: “Overturn court rulings that curtail women’s and minority’s rights to challenge pay discrimination”. Pass a “Fair Pay Act” ensuring women receive equal pay for equal work. (I feel that these are great and I applaud equal opportunity employment for women and minorities). Pass an “Employment Non-Discrimination Act” that prohibits discrimination based on sexual orientation or gender identity or expression. This will effectively tie the hands of employers, making them choose a homosexual over a heterosexual employee for fear of being accused of discrimination.

Conclusion

These are just three of the changes promised by our next president. He has many other changes he plans to make during his 4 years in office. I’ve just highlighted a few that I feel are particularly alarming.

Now to the real purpose of this (rather long) post. How do I as a Christian respond to the results of this presidential election? I could throw my hands up in the air and complain about how unfair the world is, but that is not what God expects from me. Here are some thoughts that put things in perspective for me. I think the may also be helpful for Christians who feel that our next president will be the savior for our nation.

First, my biblical duty is not really to elect my nation’s leaders, but to pray for them.

“As for me, far be it from me that I should sin against the LORD by failing to pray for you.“

I Samuel 12:23 (NIV)

Second, regardless of who our nation’s leader is, it is my biblical duty to submit to them, because they were given their authority by God.

“Everyone must submit himself to the governing authorities, for there is no authority except that which God has established. The authorities that exist have been established by God.“

Romans 13:1 (NIV)

Third, because our nation’s leaders are really elected by God, they are under his control. They cannot do anything He doesn’t allow.

“The king’s heart is in the hand of the LORD;
       he directs it like a watercourse wherever he pleases.”

Proverbs 21:1 (NIV)

Finally:

I welcome thoughts and criticisms of anyone, whether you agree with me, or have a totally opposing view. Feel free to comment, please just be courteous to me and anyone else who comments. Thanks for reading!

While it eliminates the need for an extra HTTP request, it also means that the CSS style sheet will have to be larger for both IE and other browsers. Still, that’s probably faster than two HTTP requests.

paulhammond.org: Conditional classnames.

I took everything that had been running when the script loaded and placed it in a function. Then added an action to run before the header was loaded (get_header). This way the function is executed after the template information has been loaded (so WordPress knows what page has been requested) but before header.php is included. Since what I was trying to do was to add a JavaScript file to the include Queue, I could have probably attached this to wp_print_scripts, but in my case it didn’t really make much of a difference.

Anyway, the most important thing I learned here was that in order to use conditional tags in a plug-in, you have to use an action hook to wait until enough information has been loaded.

Hope this is helpful to someone else!

Eric Martin » WordPress Pumpkin.

Check it Out!

Perhaps if I display a humble affect I can talk the school into making this a major. Couldn’t be much harder than Civil Engineering right?

I’ve also gotten to try out my new technical drawing pens. Unfortunately, I’ve not done anything worth mentioning with them yet. Mostly just scribbles and a couple sketches. Drawing with pens is really a tricky art. I’ve got a lot to learn about the techniques required to produce worthwhile drawings with them. It’s really quite different than graphite!

A couple other things that came in the mail are my wireless router (which hopefully extend the wireless accessibility to my room) and my 1TB hard drive. I have yet to get either of these set up properly, but I’ll hopefully have a chance by the time break is over (man that time is coming fast!).

Anyway, here are the sketches I did with the graphics tablet.

Again, I hope to eventually produce something more worthwhile, but this is what we’ve got so far

Waste of Water

Eiffel Tower Sand-Sculpture

The Pause That Refreshes

He’s also done art in other media, but I personally think his chalk paintings are the best.

Other bits of excitement… my parents are getting me a graphics tablet from WOOT for my birthday, and I’m waiting on a 1 TB external hard drive which should arive in the next couple days. I also just purchased a set of Koh-I-Noor Rapidograph Technical Pens which I can’t wait to try out. It should be an exciting 4 days

Freshman spend the week before homecoming constructing the bonfire out of about 5,000 railroad ties. It reaches 20-30 feet into the air and is crowned by an outhouse donated by someone in Terre Haute. Adding to the excitement, the sophamores attempt throughout the week to tear down the bonfire. 8:00 pm until 8:00 am is “open season”, but they have to get past the freshman who are guarding the bonfire with their lives (well, at least with nerf guns and maple syrup).

After the pep rally on Friday, everyone heads over to the bonfire to watch a fireworks display (which is amazing) and then the moment everyone has been waiting for… the lighting of the bonfire.

The fencing club decided to have some fun (it’s unfortunately hard to see here, but they’re fencing with fire brands).

Finally, “shiny guy” cooked us some marshmallows in the blaze.

EDIT: Turns out “shiny guy” is Daniel Sing, a grad student.

Watch him on YouTube

I question the sanity of putting a bunch of engineers and scientists near so much fire, but the event ended safely and everyone had a good time (nobody did anything too stupid).

Now I can’t wait until next year… when I get to give the new freshman a hard time!

He discussed some of the challenges of the web, and shared some of his thoughts on how the web will change in the future. It was an excellent time, with lots if interesting anecdotes and an informative question and answer time.

Here’s a picture if the lecture. Unfortunately, my phone doesn’t take very good pictures if the subject is more than a few yards away, so this one didn’t turn out very well. One of these days I’m actually going to purchase a real camera!

The first year, the guy who started it all wrote out Pi. He began in the Quad, went around the lake behind the Chapel, and finally went off the dock (if you’re not familiar with campus, that’s a LOT of numbers!). Last year, he did e. This year he wrote out Phi to 10,811 digits. I joined in the fun and wrote out root 2 to 2,226 digits.

Its An Anderson – Calendar.

Another big claim they made is that our country was founded on the principles of freedom from God and freedom from religion. Funny, but I always thought it was Freedom OF Religion! They ask the reader to help them “shore up Jefferson’s battered ‘wall of seperation between church and state’ “. I wonder if they’ve actually read Jefferson’s letter. I wonder what he would have thought of the lengths people have gone to twist his words to mean the very opposite of what he was trying to do.

The last thing I want to point out (I could go on for quite a while) is that they bash faith based services in the community. They claim these are destructive to our culture and should be “legally challenged”. I wonder which part of the faith based prisons (for example) they have issues with. Is it the reduced violence and trouble-making, or maybe the reduction in tax payer costs. Whichever part they’re against, I’m certainly glad they’re working so hard to make this country a better place to live.

Note: I feel it would be giving too much credit to the foundation that funded this ad if I put a link to their site on this post, but on the other hand, it would also be wrong not to let you see the other side of the argument. If you’d like to see what they have to say, search “The Freedom From Religion Foundation” on Google (or Yahoo, or whatever) and their site will be the first to come up.

Paul Stubbs : Turn off the Language Bar.

Anyway, I’m sort of geeky when it comes to pencils (and a lot of other things) but at least it keeps me off the streets!

I purchased books yesterday for about $660. Fortunately most of them will be good for a couple classes, so next term I’ll probably only have to spend about $100 or so.

I’m taking Differential Equations 1, Discrete and Combinatorial Algebra 1 (kinda cool, my dad was one of the technical editors for the textbook), Physics 1, and Intro to Logic Design. I’ve also got college and life skills, but it won’t take nearly as much of my time as the other courses.

I plan to put up my class/work schedule for anyone who cares. It might go on my Rose-Hulman page though, we’ll see.

When I first booted, it took only about 2 minutes (even with the first time configuration and everything). Now it boots even faster. Here are some changes since the previous model.

• Improved “Dashboard”Rather than the clunky buttons to control the wireless adapter and the sound, the new computers have light sensors (way cool!) so you just place your finger on them to change things (and the volume control is a light sensing slider (again, way cool!).
• Added portsThere are now 4 USB ports rather than just 3. Though not often an issue, I have had occasion to wish I had more than three ports. Now it won’t be as much of an issue.
  This model also includes an HDMI port, which could be useful for connecting to a (super) bug screen.
  It’s interesting to note that this model doesn’t include the serial port, but I no longer have the only device I ever used this with, so that’s not a problem.
• Different speaker locationWhile I’m not sure whether to call this a feature or a problem, the speaker is now on the front of the machine, rather than beside the keyboard. I’m not sure the sound will be as good because of this, but it allows for a larger keyboard and it’s also less likely to get clogged with grime.

Overall, the computing power is truly amazing. I’m now ready for classes to begin.

I also met with my freshman adviser today and finished registering for classes. I’ve got mostly morning classes, with ECE130 in the after noon. Hopefully this will let me do homework and my work at Ventures in the afternoon and then work for Noble Romans and IRPA in the evening. With 17 credit hours of classes (if you count college and life skills ) and 30+ hours of work a week, this promises to be a busy term.

Freshman orientation began yesterday. So far the activities have been enjoyable.

Probably the most fun thing we’ve done is play bingo with the squares being things one of the other 400 people has done (or can do). The squares had things like “Been to at least three continents” and “can juggle”.

The most informative activity so far has been the CSI session. The Vigo County Coroner talked about the differences between the CSI TV shows and what really happens at a forensics lab.  The talk was quite interesting and enjoyable (though not for the faint of heart. He had many graphic images and gory stories to share).

I’ve also been able to meet lots of people, which has been fun.

Check out this amazing rendition of “The Flight of the Bumblebee“.

You can read more about this in the Rose-Hulman news article.

I hope to get a chance to do some more work on it soon, but with my many commitments, it’s really hard to find time to even post, much less develop the theme.

I’m having my wisdom teeth removed this coming Friday, so maybe if I’m feeling up to it, I can do some more work on the site while I’m recovering (no promises though )

The basics of this trick are as follows.

First we need our HTML to work off of.

<a href="#" class="rollover">
Javayahtzee Icon
<img src="http://www.itsananderson.com/blog/wp-content/uploads/2008/04/av-1199.png">
</a>

Of course, if this were a real world situation, the link would most likely point to a real place (like the user profile or something), but for now we’ll leave it at this.

Now for the CSS. First we’ll make the link color gray. There’s no real reason for this besides the fact that I want to do so.

a.rollover{
color: #777;
}

we want to hide the image unless the link is being hovered over.

a.rollover:link img{
display: none;
}

and finally we show it (and set its position to absolute) if the link cursor is moved over the mouse.

a.rollover:hover img{
display: block;
position: absolute;
left: 10px;
top: 1ex;
border: 0;
}

we have effectively made the image visible again, made its position absolute (we explicitly place it), given the coordinates of its top left corner, and set it’s border to 0.

the complete html code follows.

<html>
<head>
<style type="text/css">
a:link{
color: #777;
}
a.rollover img{
display: none;
}
a.rollover:hover img{
display: block;
position: absolute;
top: 1ex;
left: 10px;
border: 0;
}
</style>
<title>Fancy Rollovers</title>
</head>
 <body>
 <p>
 <a href="#" class="rollover">
Javayahtzee Icon
<img src="http://www.itsananderson.com/blog/wp-content/uploads/2008/04/av-1199.png">
</a>
</p>
</body>
</html>

I hope this article is helpful to someone!
-Will Anderson