Arbor web solutions

Contact

Blogging in Django (or Why I'm Still On WordPress)

Django is powerful, but you have to work hard to bend it to your will.

Blogging in Django (or Why I'm Still On WordPress)

Back in September 2010, I started reading about responsive web design. I had already read Ethan Marcotte's seminal "Responsive Web Design" article back in May in A List Apart, but at the time it didn't really strike me as anything more than a bleeding-edge technique for people making mobile sites. But in September, I got the inspiration I needed to see that article in a whole new light: Bryan Rieger's "Rethinking the Mobile Web". This was the presentation that convinced me that mobile is important to everyone, large or small, personal site or business site - and that responsive web design is also important to everyone as a result. It also made me go back and re-look at Luke Wroblewski's "Mobile First" talk about designing for... mobile first.

So enough history. October 2010 was when I started making a plan to put all this reading and learning to good use - a redesign of this very site, Arbor Web Solutions, using mobile-first design and responsive techniques. Then, as now, I worked as a Django programmer and front-end developer. My thought at the time was that I would create my own Django-powered blogging software to replace the WordPress site I had been using. With Django, I would have full control over every aspect of the site's backend - what kinds of data were captured for each post; custom data types for different types of posts; even full web applications built into the backend of the site. I dreamed, I planned, I sketched; time continued to pass with no progress at all. I came across a number of existing Django projects that were doing the very thing I set out to do - DjangoCMS, FeinCMS, even Mingus, an attempt to create blogging software by combining over twenty smaller Django projects. Each solution had things I loved and things I detested, much like the WordPress site I was seeking to replace.

I decided that the best thing to do was to write my own from scratch. This might sound crazy, but writing a Django blogging engine is basically a rite of passage; the Django tutorial, after all, shows you how to build a very basic blog.

In which external constraints turn out to be a very good thing

As always, change happens. While I spent months debating the pros and cons of various approaches to Django blogging, two major things happened: I was accepted as a speaker for WordCamp Miami 2011, and WordPress 3.1 was released. My site redesign now had a deadline - March 5th, 2011 - and completing the redesign on WordPress started to seem like a very good idea. WordPress had not been standing still, either. The 3.1 release added even more custom post type options with support for "formats" like "asides" or "galleries". The custom post types of WordPress 3.0, which I had mostly disregarded at the time of 3.0's release, had been further polished, and custom taxonomy support had also improved. With each release, WordPress has been picking up new features that have helped set other software apart, and in particular the gap between what is easily possible in Drupal and what is easily possible in WordPress has been closing fast.

I worked hard and I finished my site redesign the week before WordCamp. In addition to the new design, I made custom post types for portfolio pieces and integrated the portfolio within the same WordPress install. At WordCamp, I also learned about Vanilla Forums and its easy integration with WordPress. Just today, I found out about Tweet Nest, a plug-and-play solution for archiving Twitter feeds that looks effortless compared to Tantek Çelik's Falcon project. Tweet Nest isn't specific to WordPress per se, but WordPress makes it very easy to integrate outside functionality in the few areas where it is lacking.

The result of all of this is that I stopped looking at Django as a serious contender. Once I had finished a static mockup of my site redesign, the actual coding required to use the design with WordPress basically took one weekend. I don't think I could ever get close to that with Django, mainly because WordPress has pre-built code for a lot of things that I would have to implement on my own in Django. The time crunch of getting ready for WordCamp Miami pushed me to WordPress, and I'm actually quite happy with the result.

Does this mean that I'll never write my own Django blog software? Of course not. Django is capable of some amazing things that are simply not possible in WordPress without massive low-level hacking of PHP, and no one really wants to do that. But my Django blog remains on my "someday" list; in the meantime, I'll be using WordPress.