If you’re reading this post, you’ve noticed that we recently migrated our blog system out of Drupal into Wordpress. This is the first step in our initiative to decentralize our various web properties and replace them with loosely coupled systems. Our aim is to make it easier to publish information and increase transparency into the Songbird project. Stay tuned for more changes to come soon.
In the process of migrating data, we wrote a set of Ruby scripts to facilitate ingestion of content. We’ve created an ActiveRecord model for Drupal 4.5 and Wordpress 2.0 schemas. The scripts take care of migrating user accounts, content, comments and categories. A set of mod_rewrite directive is also generated to help ensure that permalinks are maintained.
In the hope that it will be useful to the community, we’ve released the sources for the scripts. You can find them on the drupal2wordpress Google Code project
Enjoy.
1 Trackback
[...] Songbird migrated from Drupal to WordPress which is something I’ve seen more and more of lately with WP 2.5. [...]
10 Comments
SubscribeGood job, Wordpress is very nice
Thanks for mentioning this migration!
I would be interested to hear more about why you chose to go with Wordpress instead of Drupal.
Also, I wonder if you could clarify what you mean by
“decentralize our various web properties and replace them with loosely coupled systems.” ?
@Ezra
Many factors were considered. Wordpress provides a rich array of plug-ins, clean templating and strong spam filter via Akismet. It’s also more familiar to most birders as it matches their blogging system.
Drupal has been the center of our content management solution for www, blogs, forums and user authentication for a while. Instead of continuing to centralize all these function into one platform, we’re considering the needs for each one of these applications and decided to migrate to those systems that are the most tailored to our needs. Where it makes sense, we’ll be tying those together using OpenId.
Your post made me curious, georges. I more then once heard about all the OpenId shebang, but never looked into it. Now I did. Personally, I think it has as many downsides, as upsides. A unified login system for the internet use has its security issues that must be dealt with, but has the huge plus, that you only have one ID and thats it. Done.
As I looked into it though, as it now stands, almost no one supports it. At least the big players in the IT seem to be very reluctant to use it. Beside of tiny (personel-)blogs and sites no one seems to use this new system. I found no single site, that catched my interest to try my newly created ID.
Guess we’ll have to wait and see, how that develops.
Behold WordPress, Destroyer of CPUs:
http://www.codinghorror.com/blog/archives/001105.html
@tylerstyle
Currently most large organization such as AOL, Google and Yahoo are acting as OpenID providers, i.e. they allow their user’s credential to be used on other OpenID sites, but don’t necessarily allow users to login with OpenID on their properties.
The sites that accept OpenID are perhaps not as prominent but their number is growing. You can look at the my openid directory and openid directory. Most notably, 37 signals the provider of the popular Basecamp, Backpack and Highrise applications supports it.
I think the only valid arguments I see is of familiarity birders have and the fact that songbirdnest.com is built with Drupal 4.5.
Drupal is currently at version six and supports OpenID (bundled in core), has a healthy module ecosystem, allows choice between many template systems (there’s even a LOLCAT template engine), and also supports Akismet through a third-party module.
After reading Dagur’s link, I’d like to point out something. It is argued that if caching is so commonly needed for WordPress, then why is it that wp-cache isn’t bundled in core as part of the distribution? In this field, Drupal shines. As major revisions carry on, important functionalities that are considered standard are simply merged with core into a core module. For anything else, you can simply build your own tailored system with more third-party modules.
This is in fact the same mechanism of Firefox; there’s a healthy ecosystem of extensions, and if some functionality that an extension provides is deemed to be worthy enough to be part of the default user experience, the extension is merged with Firefox. Session restoring is one prominent example.
I’d also like to make one final argument about moving away from Drupal. Drupal is a very flexible system, to the point that it’s almost like a base for lego blocks; you get the core and you build anything you want. If you were to decentralize services and tie them up loosely using a unified authentication service, it is much easier to manage several customized Drupal installations than to tailor all sorts of different systems. Drupal’s not just another CMS; it’s a platform.
Very interesting migration.
Ubuntu, IBM and Yahoo run their websites on Drupal.
You should have upgraded to Drupal 6 instead of Wordpress. It would have been easier to port.
I agree with kourge on this georges. I don’t see a clear reason to migrate from Drupal to WordPress.
Perhaps you could elaborate a bit more to explain the reasons if you don’t mind.
@Juanzo, I don’t know that I have much more insight to offer. Drupal has served us well, and is a robust solution with many uses. Similarly, Wordpress is a solid contender in the narrower segment of blogging. As “kourge” points out, we’ve been running an older version of Drupal (4.5) for some time now. Our configuration is less than optimal and we simply don’t have in-house competencies to take on an upgrade or to fix some of the problems we’ve been encountering. The path we chose made sense for us given our resources and needs. I encourage everyone to make their own decision based on their particular circumstances.