Google Buzz is new, but you can still use good old coreylib to pull your buzz feed into your PHP application.

Step 1: Make sure you have PHP 5 with cURL and SimpleXML installed and enabled. (If you’re using MediaTemple or eApps, you’ve got it.)

Step 2: Get coreylib – http://github.com/collegeman/coreylib/zipball/v1.1.1

Step 3: Get src/coreylib.php out of the ZIP and put it in your project.

Parse! Here’s an example

<!-- load coreylib -->
<?php require_once('../src/coreylib.php') ?>
<!-- two levels of debugging, should something go wrong... -->
<?php
//clAPI::configure('debug', true);
//clAPI::configure('trace', false);
?>
<?php if (!clCache::cached('google.buzz.feed', '1 second')): ?>
	<!-- create a new API parser with the feed address -->
	<?php $api = new clAPI('http://buzz.googleapis.com/feeds/108964711519495307614/public/posted') ?>
	<!-- parse it  -->
	<?php if ($api->parse()): ?>
		<!-- uncomment this line if you want a pretty tree to explore -->
		<?php //$api->info() ?>
		<!-- use xpath to grab all of the <entry> elements -->
<ul>
		<?php foreach($api->xpath('//feed:entry') as $entry): ?>
			<!-- for each entry element, grab the alternate link and the content -->
			<?php
				$content = $entry->first('feed:content')->__toString();
				$content = substr($content, 5, strlen($content)-5);
				$href = $entry->first('feed:link[@rel="alternate"]/@href');
			?>	
<li><?php echo $content ?> <a href="<?php echo $href ?>">&raquo;</a></li>
 
		<?php endforeach; ?>
		</ul>
 
		<?php clCache::save() ?>
	<?php endif; ?>
<?php endif; ?>

To learn more about coreylib, check out the project page.

After reading Seth’s praise for custom printer CustomInk on more than one occasion, I decided I’d give them a shot.

Now, that’s not the whole truth.  The whole truth is that at first I couldn’t remember the name of the company.  I could, however, remember the name CafePress.  The funny thing is I’ve never used CafePress before – I’ve only ever heard of it.

What was it that made me consider CafePress before gritting my teeth and plundering Seth’s blog for the forgotten name?

Perhaps it is because I’ve heard so many people talk about CafePress, my mother in particular.  My mother is a pastel artist, and she does beautiful work.  She once considered using CafePress to distribute tiny prints of her work.  The samples we received in the mail weren’t any good.  Still, when it came time for me to put brand to cotton, I still went back to the inferior service first.

All I’m saying is that there is marketing at work here in a way that I don’t fully understand. I don’t typically find my mother in tribes to which I am a member.  Go figure.

Long story shorter, I was finally able to recall the name, and let me tell you: I am one satisfied potential customer.  The t-shirt design workshop CustomInk provides through their Web site is top-notch.  The interface is a little 2003; but the usability is unparalleled.  I had only one issue with the tool: it took me a while to figure out how to allow the white to stay in my uploaded art.  (There was a checkbox – I overlooked it.)

The t-shirts are for promoting coreylib – my universal API client (with caching built-in!). Ten t-shirts in two sizes for $251. That’s not bad, considering I’ll be creating ten community billboards the size of Large and Extra Large chests.

Best of all: the CustomInk workshop lets me share my designs with my audience ahead of time. Here’s hoping you guys like it.  Please note: I’ll probably be changing the tag line on the back – the current one is a bit too confrontational. (See, I’m trying to learn from my mistakes.)

This article is about a new WordPress plug-in that I have just released in ALPHA.  Download and use at your own risk: it is far from finished.

I love WordPress. But writing plug-ins for WordPress sucks. There are a lot of plug-ins out there, all of them using WordPress’ smashing plugin API to create some great functionality. But at the end of the day, the native API leaves me wanting more.

I want to develop WordPress plug-ins using MVC principles. I want for security to be easy to remember and easy to implement. I want to be able to write plug-ins that provide body content on the front-end with minimal modification to my themes. I want to be able to utilize AJAX without implementing esoteric, hack-like features in my plug-in.

And I don’t want to have to refer to the gigantic Codex every time I need to one-off a new project: give me a manageable API.

With these goals in mind, I have written a plug-in for WordPress that creates a new plug-in framework. This framework, based on CodeIgniter, has far exceeded my personal expectations for the project (and it’s still in alpha!). Not only did I manage to satisfy my own requirements, I managed to do so without hacking either of the proven stacks, all while taking advantage of some very cool features of PHP 5 (annotations).

I named the plugin WP-CI.

A bit on how it works

Explaining the nuances of MVC development in CodeIgniter is beyond the scope of this article, and with good reason: they have some great documentation.  But for the uninitiated, know this: CodeIgniter divides application functionality into Controllers and Actions.  Each Controller gets its own PHP file, and each public function on the Controller is exposed to the Web as an Action – discreet units of application functionality, each with a specific purpose (like display a form, save some data, delete a record, or return some JSON).

In WP-CI, each controller file (stored in ./application/controllers) is, in effect, a special WordPress plug-in that I call a WPCI Plug-in.  Within the body of each PHP file you have complete access to the entire WordPress API, as well as complete access to the CodeIgniter framework.

To learn more about the structure of a WP-CI file, download the latest release and have a look at ./application/controllers/hellosparky.php.  A more complicated example exists in the WPCI Plug-in Administrator, which is itself a WPCI Plug-in.

Using annotations

Once again, for the uninitiated, annotations (in PHP) are special comments added at various locations in a PHP file.  This is a concept specific to PHP’s object-oriented programming syntax, and allows developers to add meta-data to class definitions.

WP-CI makes extensive use of this PHP 5 feature, making this framework incompatible with PHP 4 (sorry guys and gals: welcome to the future).

Annotations look like this

/**
 * @user_can(edit_plugins)
 */
class Admin extends WPCI_Controller {
  function index() { /* ... */ }
}

The annotation featured above, @user_can(edit_plugins), is a cue to our framework to require any user accessing this plug-in to have the built-in WordPress capability of editing plug-ins.  (This requirement implies a user be an administrator.  You can read more about user capabilities in the WordPress Codex.)

There are other annotations too, like @menu and @submenu which serve to create administrative menus and menu items from your plug-in actions.

For more examples of annotations in action, have a look at the WPCI Plug-in Administrator, ./application/controllers/admin.php.

Early adopters beware

The current release is ALPHA and so you should expect the core API to change with the next release, and possibly even break your application in complicated ways. You have been warned.

This software is not warranted in any way.

So, have you tried it out yet?

While designing the Web site for Winchester’s Bright Cowork, my friend and colleague @coreyweb used a service called crowdSPRING to crowdsource the logo. As a spectator to the process, it seemed like an efficient and relatively inexpensive way to get some design work done. (You can see the final product here.)

For those of you unfamiliar with the concept:

Crowdsourcing is a neologism for the act of taking tasks traditionally performed by an employee or contractor, and outsourcing it to an undefined, generally large group of people or community in the form of an open call. For example, the public may be invited to develop a new technology, carry out a design task (also known as community-based design[1] and distributed participatory design), refine or carry out the steps of an algorithm (see Human-based computation), or help capture, systematize or analyze large amounts of data (see also citizen science).

Commissioning work through crowdSPRING (almost) couldn’t be easier: just describe the project, and set a competitive price. The price you set is what will be paid to the winning designer. crowdSPRING then charges buyers 15% on top of the stated price – for a $300 job, this fee is $45. All money is paid ahead of the launch of the project, and is held in escrow until the project’s close – by default, one week. “Creatives” submit designs, and at the end of the project, the buyer selects one to be the project winner.

I was encouraged by the results Corey got from his crowdSPRING experience, so I decided to create my own project there (a new logo for my coreylib project).  So far, so good: I am very excited about some of the responses I have received.

But despite what I would call success, it should come as no surprise that the design community is very vocal in their disapproval of crowdsourcing and services like crowdSPRING. There is a status quo being challenged here. Crowdsourcing introduces competition on a grand scale into what has traditionally been a very selective process.  While this does yield a net benefit to the buyers, the work generated does exhibit a few negative consequences of the community process: namely, degraded quality and consistently, and a drastic reduction in the value proposition for the designers.

Business analysts and the service brokers themselves are claiming that crowdsourcing is the future of design, given the reduction in cost and the buyer’s ability to be extremely selective and critical of the end result.  Wrapped too in this new capacity to criticize is a departure from a time of designers being treated like doctors: that is, rather than being given carte blanche, they are instead often forced to assimilate the sometimes-arbitrary opinions of their customers (”I think it would look better if it were more balanced: centered vertically.”).  While this assimilation probably does happen quite a bit in the old world of design, I imagine that this play comes at a much higher price than $345.

It is easy to understand how this new marketplace could be perceived as an assault on the authority of designers and their maintenance of common (and important) aesthetics: an assault that, were it authentic, certainly should lead to a very cynical reception.  But I completely disagree with the idea that crowdsourcing is the future and will replace the traditional marketplace for design. The community-driven process of a service like crowdSPRING could never surmount the true authority and expertise that comes with a trained hand and mind, and the dedicated consultation.

What crowdsourcing does achieve is the creation of a niche market: one at which someone like myself, with limited time and a tight budget, can easily and efficiently vet a large number of ideas.  But these services are far from perfect.  Of the two I know about – 99designs and crowdSPRING – I chose crowdSPRING simply because I knew of a local success story.  In hindsight I am relieved I chose crowdSPRING over 99designs, because the manner in which 99designs manages the purchasing process is completely unethical. Unlike buying designs from crowdSPRING, 99designs does not appear to require the buyer to award a project. This leaves designers highly vulnerable to idea theft, and nefariously forces many into providing free consultations. Frankly, for a thing like community-based design to be fair and productive, both the designers and the buyers must have some skin in the game.

So how could using crowdSPRING be easier?  Well, for one thing, it would be nice to be able to quickly view Creatives’ previous submissions.  It is possible to get to previous submissions from a Creative’s profile page (a feature built-into crowdSPRING), but once on the project page, the user’s submission is hidden amongst all the other submissions made.  Some more guidance given to Creatives on the process of submissions would be very helpful: rather than having to score and review seven small variations on a single concept, it would be far more efficient to have all seven submitted as a single submission.  It took me a while to discover that I could score and comment from the gallery page – making this feature more prominent, and adding more controls for navigating between zoomed pieces would dramatically increase the frequency with which buyers provide quality feedback.

So if you’re on the market for a new logo or t-shirt design, I recommend giving crowdSPRING a try.  Make sure to read and follow the tips given to buyers at the onset of the project, namely the one that advises buyers to send private messages to creatives inviting them to work on the new projects.  Using this push approach to getting traffic to your project definitely increases the frequency with which you receive submissions. That and providing as much feedback as possible – most of the crowdSPRING designers thrive on constructive criticism.

If you’re looking for a place to start, I recommend these Creatives:

• genuine
  
• ninet6
  
• davebowman
  
• simply
  
• BMdesign
  
• majosa
  

Have you had success with crowdSPRING or crowdsourcing? Please share your experiences in the comments.