Josh Davey

Process

Thu, 04 Feb 2010 01:54:17 -0500

Today Dave Senior and I had the amazing opportunity to talk to Bill Buxton, Principal researcher at Microsoft Research and give him a demo of the product we’re creating. It was refreshing to finally talk to someone who wasn’t interested in how we were going to make money, but how we could simply create a great experience.

All of Buxton’s ideas and feedback were incredibly useful and invaluable. I was most intrigued in Buxton’s interest in not the content we were creating but how and why that content was being created and representing the story leading up to its creation to frame its meaning and relevance.

His comments remind me of an essay a friend of mine once wrote about the importance of process that I have always dwelled on. For example imagine creating something as simple as a perfectly round circle. Depending on how you create that circle changes how people react to it. If you created it with photoshop you probably spent a few minutes and impressed no one. However if you drew the exact same circle with a pencil that you held in your teeth, while being held upside down, to demonstrate infinity, you will probably turn a few heads and create interest.

The thing that makes one perfect circle stand out more than the other is HOW and WHY it exists.

Facebook Connect Fail

Fri, 01 Jan 2010 17:39:20 -0500

To start off 2010 I quit a few things that I thought were taking up too much of my time. The first thing I quit was PHP which I blogged about yesterday. The second thing I quit which, I did not blog about but, I will now, because of recent events, is Facebook.

I quit Facebook not because it was an addiction but because I decided I didn’t need it. Now unfortunately I can’t completely quit Facebook because I’m a developer and the Facebook platform is undeniably powerful. I have removed my personal account but, created a new account for testing. However as I just recently discovered its going to be harder to quit Facebook than I thought.

You may be thinking its because my friends have no way of contacting me now (for some that is true and I will miss you) or because I won’t receive event invitations (99% of the time this is a good thing) but, its actually my past laziness which has put me in an awkward position.

Most of the new services I have signed up for in the last 6 months offer the option to sign up with Facebook Connect. It’s so easy! Just click on Facebook Connect and you’re ready to go. Well yea… but what happens when you deactivate your Facebook account like I have? I am now locked out of some of my favorite services like posterous.com.

Facebook actually has a system to deal with this situation where Facebook sends you an email when you deactivate your account with links to all the services you signed up to with Facebook Connect. By clicking on each of the links you can re-claim your account. In theory this is a great idea but how many developers really care about this use case? Do they really expect all their users to be as crazy as me? No, and thats why of all the services I am signed up to only foursquare.com had a way for me to re-claim my account.

So if you see me sign on to Facebook today just pretend nothing happened.

I'm Quitting PHP

Thu, 31 Dec 2009 19:19:41 -0500

I’m excited. Really excited.

But, First The Ugly…

I’m a self taught PHP developer. I’ve been programming PHP for 7 years and in that time have started some really cool projects and even started a few profitable businesses. It was all in good fun until a couple of years ago I decided to build my own framework (Madeam). The goal was to develop a tool that I really enjoyed creating all of my applications with. At first this makes a lot of sense because as you’re developing your applications you are constantly improving your tool and every time you use it, it gets better and your applications get better. Wrong!

What happened is I completely lost focus on building great applications and instead focused all my energy on creating a great tool. Don’t think it will ever happen to you? Time may be infinite but your time is finite. By exerting so much time on improving the tool there is no way you can give the final product the attention it deserves. Already roll your own framework? Look at all the time you just wasted figuring out the optimal way of handling HTTP requests.

I’d recommend using one of the hundreds of existing PHP Frameworks but, unfortunately they all suck. Perhaps this isn’t their fault though and the real problem is the PHP community in general that can’t agree on naming conventions or coding standards? No. The PHP community has always had PEAR. The problem is that everyone decided to completely ignore the PEAR coding standards and invent their own. Even worst most of the time they don’t even namespace their classes so you end up with classes called “Cache” but then you try to use a third party library by the same name and everything explodes.

There are literally thousands of PHP classes and libraries out there but you can’t use them because they clash with each other. Why does it need to be this hard to use other people’s code? Shouldn’t the framework developers be focused on creating interoperable platforms so we can all share in the wealth? Instead they’ve decided to re-write their own libraries for everything and made it impossible to plugin third party code. Everyone has taken the “It’s my way or the highway approach”.

So why am I excited? Because this is the internet and there are plenty of other programming languages to use.

Happy New Years!

This year I’m quitting PHP and MySQL for all of my personal projects, and kicking things off with a new technology stack; Python, Django, Postgresql and Nginx. I’m not going to sing their praises just yet (although, I do thoroughly enjoy them so far). I understand that all technologies have their problems and its possible I’ll become even more frustrated than before but at least now I’m not wasting my time trying to create the perfect framework. Instead I’m working with the tools I’m given to create some really exciting products. I can finally focus on the end product’s experience instead of my own.

As I continue to play with these new toys I’ll be sure to document my experiences on this blog. All of this excitement gives me the urge to blog blog blog. Looks like 2010 is going to be a great year!

Working at Ryerson’s Digital Media Zone (DMZ)

Wed, 09 Dec 2009 20:37:00 -0500

Working at Ryerson’s Digital Media Zone (DMZ)

PHP Standards and Frameworks

Fri, 30 Oct 2009 20:32:00 -0400

Recently a number of former CakePHP developers announced a new framework called “Lithium”. The project is built from scratch to take advantage of PHP 5.3’s latest and greatest. The current state of the project is exciting as it no longer looks like the mess that CakePHP was.

In a recent blog post @nateabele says:

“Back at php|tek in May, I had the opportunity to meet with lead developers and community leaders from several major PHP projects, including Solar, PEAR, Agavi, Symfony and Zend Framework. The result of that meeting was a set of naming and organizational standards that would allow developers to very easily integrate components from each framework or library in any other. This is a very big deal, and I’m proud to say that Lithium is the first framework to implement this naming standard.”

I find this rather amusing because I’ve been promoting this very idea for ages. Madeam has always been based on naming conventions identical to those of PEAR’s. The reason to use these standards are as follows:

Consistency
Autoloading
Namespacing

I’m glad the former CakePHP guys have finally figured that out. Hopefully now everyone will be able to share resources more easily.

Finally, I got one more laugh out of this from the Lithium website “Lithium is a lightweight, fast, flexible framework for PHP 5.3 and up.”. Lightweight? Flexible? What ever happened to being a full stack framework? The truth is its a stupid idea to have everything integrated into a single framework. It just doesn’t scale and it discourages open communities. The framework should do a few things and do them well. Leave the rest to the library creators. This is the type of thinking that has always been behind Madeam. I’m glad Lithium is hopefully going to move away from being another CakePHP disaster (benchmarks).

It wasn’t long ago when @nateabele tweeted that Madeam wasn’t a real framework when it was featured in PHP Architect Magazine. Its funny now to see them adopting many of the same ideas.