the kdmcinfo weblog

Is Your CMS a Bottleneck to Innovation?

Hundreds of journalists have moved through KDMC's digital media training programs, in which we show them how to produce cutting edge news content in non-traditional formats, from video to map mashups, custom widgets to slideshows, social media integration to standalone multimedia packages. Once the journalists return to their home publications to put their new skills to work, they often run up against technical "limitations" in their content management systems that prevent them from being able to publish certain media types to their sites. 

Many publications are sitting on top of content management systems that are either antiquated or weren't thought out well to begin with. Journalists may be told by their web teams, "Sorry, we don't support that kind of media" or "No, you can't use a custom template for your standalone package" or "We do not create custom database tables for data-driven reporting stories" or similar.

So journalists are being told, on one hand, "Go forth and innovate - we'll even pay for your training," and on the other, "Our systems don't support anything but simple text-based stories and images."  In these cases, technical limitations in the publishing system become the bottleneck limiting progress. But technology in journalism is not standing still, and publications that can't deliver new media content quickly and without friction will lose traffic to those who can.

To be fair, I know how complex content management systems can become. Some of them represent hundreds of thousands of lines of code, with delicate interdependencies that authors and editors never see. It can be difficult to make significant changes without affecting old content, or affecting related systems, or stepping on the toes of another department (like advertising). Not all CMS changes are easy, and I don't automatically blame the people who build and run them.

At the same time, there are many cases in which the technical staff is the problem. They may not realize how critical it is for the organization to have a flexible system, or they may lack the will to take on risky or difficult programming challenges. Often, management doesn't understand how the CMS really works, and may not be able to ask the right questions or give the right orders.

In fact, many organizations are dependent on the CMS of a parent organization. For CMSs that drive multiple publications, changes can have  even deeper ramifications on shared data structures, templates, etc.  In these cases, the publication may not have the ear of the decision makers who could make a difference, and control over the CMS is even farther removed, making change even more difficult, if not impossible.

The key is to ensure that upper management understands the real-world market ramifications of being saddled with an inflexible system. While the challenges of making CMSs more flexible are genuine, the expense of not doing so can be great, especially as smaller, more nimble news startups appear on the scene. Startups have the advantage of a clean slate, and are able to build new CMSs that are more flexible to begin with. They're able to learn from the mistakes of the past and get off on the right foot. Which means they can produce more innovative content more quickly and more easily, without having to work through multiple levels of bureaucracy to get things done. 

Here are the things I think have to happen to prevent or eliminate the CMS bottleneck:

  • Decision makers need to appreciate the importance of providing modern, flexible publishing tools for their staff. Even if it means spending more on technology for a while.
  • Web staff who manage CMSs need to work closely with journalists, not be cordoned off in the basement of a building in another state. Web staff need to be invested in the needs of the organization. Make sure web staff are working side-by-side with journalists - this should be a partnership, not an adversarial relationship.
  • Someone with both knowledge and power needs to be in a position to respond to "NO" answers from technical staff. Often those "NO" answers are unfounded, and really mean "We've never done it that way before" or "That's going to be too hard."
  • If you have the luxury of starting with a fresh CMS, consider building it yourself on top of a framework like Django or Rails. Frameworks are inherently more flexible and accommodating of change than off-the-shelf or custom content management systems.

Is the CMS a bottleneck at your publication? What steps are being taken to improve the situation?

Comments

1) Hunter Vurbeff, October 2, 2010 at 11:02 a.m. [Link]

The CMS at my publication was originally designed for print only and was later reformatted to add web content. In short, CMS is a hindering factor to say the least as Web Editors have to focus their time solely on working their way through the slow process of uploading a story. This hinders innovation or featured items that could make our online site flourish.

2) Scot Hacker, October 4, 2010 at 12:27 p.m. [Link]

Peter -

Joomla and WordPress work fine for basic publications, but not for advanced ones. WordPress in particular assumes way too much about the structure of your data. WP3 finally adds the ability to include custom data types, but the feature is new and still has some maturing to do. I defy anyone to build something like an equipment checkout and resource tracking system, a course review system relationally attached to instructors, courses, semesters and students, etc. etc. in WordPress. I can't speak to Joomla - I assume it's better than WP in this regard, but the general consensus on Joomla amongst professional developers is that it sucks. For our part, we use WP for sites that resemble a basic publication, and Django for sites with more sophisticated sites. Django is just an incredible framework with infinite flexibility (because you build the parts you need rather than trying to strip away an CMS's opinions and assumptions).

All of that is an aside though - for virtually all journalists, they're stuck with the CMS provided by their publication. Creating a fresh one is simply not in the cards (and may be specifically disallowed).

3) Scot Hacker, October 4, 2010 at 12:28 p.m. [Link]

p.s. Would you try and build nytimes.com with Joomla or WordPress? No freaking way.

4) Scot Hacker, October 6, 2010 at 2:05 p.m. [Link]

Hi Pete -

In my opinion, you're underestimating the power of WordPress and overestimating the quality of Joomla. That aside, I don't doubt that nytimes.com *could* be built with Joomla -- the question is whether it would be a wise choice.

For my money, when developing a complex site, there's NO WAY I would use a prefab CMS like Joomla or Drupal - I don't ever want to go down that road again. Frameworks like Rails and Django make the development path so much cleaner. I don't have time to get into here, but please have a look at an article I wrote on this topic a year ago:

http://birdhouse.org/blog/2009/11/11/drupal-or-django/

Best,
Scot

for more clarity on what I'm getting at here.

5) , October 10, 2010 at 12:01 a.m. [Link]

We're a tiny publisher trying to convert our website using CMS, particularly for 2 of our on-line publications. Our programmer is a fresh grad with basic knowledge of Joomla. I wonder if we should abandon our efforts to use Joomla and use Drupal instead or attempt to use Django? He does have Php knowledge.

Any advice?

6) Scot Hacker, October 11, 2010 at 10:42 a.m. [Link]

Vicky - It depends how much time you have on hand, and how good of a learner the developer is. I certainly wouldn't wish Joomla or Drupal on anyone! But if he only has PHP skills, there's going to be a bit more learning curve. In exchange for going through that curve you'll end up with a much better constructed system that will let you make modifications in the future more easily and cleanly, without fighting against all of Drupal/Joomla's opinions and assumptions. With Django, you spend your time building the features you want, rather than ripping out or trying to work around the things Drupal/Joomla have "provided" for you.

7) J. Heasly, October 14, 2010 at 12:05 p.m. [Link]

We have one of those innovation-bottlenecking publishing systems (IMHO) and the strategy with which I've found success is to daily extract the print publication from the publishing system database and insert it into the Django system I've built.

I basically use Django as a "helper engine." It's so much easier to make AP ingestion feeds, Google sitemaps and other custom feeds in Django than using our IBPS (innovation bottlenecking publishing system). In fact, I wouldn't even want to try! And Django's flexible enough that I can make the feeds produce IBPS-compliant URLs. Basically, the two systems are kind of interwoven. When I have to do something innovative that requires access to our print product, I use Django to take the data out of our IBPS and then produce Web products that point back to the content on the IBPS. Does that make any sense?

And if the innovation doesn't require print product data, I do it completely in Django.

8) Scot Hacker, October 14, 2010 at 3:09 p.m. [Link]

J. Heasly - Love the acronym "IBPS" :) Right on to finding ways to augment (or work around) the IBPS with Django. Thanks for sharing that.

9) J. Heasly, October 15, 2010 at 11:46 a.m. [Link]

Thanks Scot!

Now that I think about it, pretty much *any* commercially available/proprietary print/Web publishing system is going to have the flaws of a WordPress/Drupal set-up in that it's built to do things a certain way. In fact, it's even worse in that the IBPS probably cost hundreds of thousands of dollars to install/maintain/upgrade.[1] Trying to do anything any other way is going to involve a lot of fighting the assumptions/prejudices of the system. At least that's been my experience. (Somebody ought to get some Knight money and do a study!)

[1] The effect is compounded by newspaper decision-makers avoiding the free and open-source because they're invested in justifying the cost/overhead of the IBPS!

10) Albert Web, November 3, 2010 at 9:14 a.m. [Link]

I have used both Joomla and WordPress and I would definitely go with WordPress without doubt. Of course depending on the complexity of what you want to build, you will have to work one way or another, but at the end you should be able to reach your goals using WordPress.

Of course knowing php and how to use a MySQL database with some commands for data retrieval would be a great advantage.

You must be logged in to post comments.