A quick thought before heading out the door today:
What if news websites already running on open source content management systems made their entire sites — content, databases, applications, templates, etc. — freely available to the community? It could simply be a matter of packaging their nightly backups (which any smart organization should be making anyway) into a downloadable file and putting out an open call for users to submit code contributions. This could also serve the twofold benefit of defraying the costs of web development and providing an educational resource to j-schools and other organizations.
I’m thinking that college news outlets are the more likely candidates to adapt such a practice first, but as we continue the trend toward increasing transparency, others in the professional world might be willing to try it too. Can anyone think of possible downsides to this?
I think it’s a very interesting idea, and somewhat parallel’s the idea of making all of your content available through an API. I believe the big concern, however, would be in regards to the desire to keep intellectual property under lock. Heck, what if the entire news organization was open source in the sense that anyone could download, tweak, and contribute to its code (both in terms of their website, but also organizational structure)? You should check out this post if you haven’t already.
re: intellectual property, I guess it’s similar to concerns about Creative Commons and the like. People who want to steal will always find a way, but standardizing access to information makes life easier for people who want to use it for legitimate purposes.
I’m a fan of open source organizations. Transparency is going to be the way of the future.
Very interesting idea.
If there is a clean separation of a site’s app code and content (and other logical components), this would be a very doable thing. I would agree with Daniel that making a full database dump (for example) available like this could be problematic because of intellectual property concerns, but why not make at least the schema itself available? And having content syndicated using RSS and exposed via APIs is even more effectively open than a downloadable tarball.
I would also caution against blending this kind of practice with backup and disaster recovery procedures. True, spreading multiple copies of site components around will prevent incidences like what befell ma.gnolia.com, but there are better ways to do backup. Then again, it never hurts to have redundant backups either.
If you haven’t already seen it, check out the Development pages of http://openlibrary.org/. If I’m reading you correctly, they’re doing pretty much what your idea sounds like.
Yeah, clean separation of code and content is a good idea, especially considering that folks who plan on using one probably won’t be needing the other — i.e. feeding content into a new app, or borrowing an existing schema for their own content.
I hadn’t seen the Open Library project before; thanks for the link.