CMS: pre-made or roll your own?
August 31, 2009 in College Media, Tech Talk, Websites
Image via Wikipedia
Madison McCord, web editor of The Communicator at Spokane Falls Community College, makes the case for college media organizations developing their own content management systems in a comment on a previous post: #
Where are the sites that are not built using any type of template (word press, College publisher) but instead are building and managing their sites in-house? I am a strong believer that we are all students, and should be learning everything about site design and management instead of using a plug and play system. Granted, it takes more time, but the experience of doing it puts those further ahead of everyone else. #I am the web editor, and co-designer of The Communicator Online, the student news site at SFCC. This last spring myself and the other designer sat in our newsroom and built our website from scratch. We enter every item by hand into the code, a tool that we can now use in the future. And just to prove that our site is no worse than any using WP or CP, we are currently an ACP Online Pacemaker finalist. #
I just think that too many sites don’t care about what their students are learning, which is the whole point of being a college student #
Edit: Someone pointed out that it’s not clear from the e-mail whether McCord is talking about a CMS, or hand-coding HTML using something like Dreamweaver or a text editor. If they are creating static web pages, then that’s a step backward, rather than forward. I’ll discuss that problem in a future post. If they are static web pages, then the CMS discussion below doesn’t pertain to SFCC, but it’s worthwhile. #
There are actually a number of schools that are now running on, or are in the process of developing, a home-grown content management system (Va. Tech, Michigan State, NYU’s Washington Square News, Yale, UCLA, Florida International). Some of them are using Django to build their site. Others are working with Ruby on Rails or PHP. Then there’s the mysterious Populous project, which was supposed to be released this summer. #
I do think it’s a little unfair to say that “too many sites don’t care about what their students are learning …” Developing a CMS takes a lot of time and careful thought. Time and thought that have already been expended on CMS’s like WordPress or CP or whatever. #
I first used WordPress when it was in 1.xx form, and to see how it has changed since then amazing. The amount of plug-ins and add-ons (not to mention themes) that are available is also something that can’t be dismissed. #
And wrangling a CMS to do what you want to do with it (whether WordPress or College Publisher, Drupal or Joomla) can be just as challenging for a developer as building something from scratch. There are plug-ins to write, ad management systems to explore, features to test and themes to tweak. A CMS can be “plug-and-play” if that’s what you want. But it’s also a canvas to explore skills in HTML, CSS, PHP, Ruby, or whatever language you care to learn (I’m speaking to web dev types here). #
There is a larger issue with building a CMS from scratch, as well, which I’ve seen up close: developer continuity. #
Let’s play out this scenario: You have a talented web developer who spends almost a year developing a content management system from the ground up. It’s well-documented and has many of the features of contemporary open-source CMS’s. #
But after a year or two, the developer graduates and takes a job in the industry. Now he/she is no longer around to continue to add to the code, and the web staff who come along afterward are busy keeping the site running, so they can’t update and fix things that are broken or could use some extra code. They can’t adapt to new technologies and social media add-ons that come down the pike (social media additions, for instance). #
Or suppose there’s a security hole somewhere in the code? Who’s going to fix it? Who’s going to poke around to make sure malicious hackers can’t get into your database or server setup? The biggest CMS names out there have people who are doing just that so you don’t have to (just look at the number of .xx updates WordPress pushes out. #
You could spend some money on an outside developer to address those issues. But in this economy, who’s got money for that? And that developer would have to spend time learning the set-up as well. #
Like I said, I’m happy for any school that can develop their own CMS. I wish them the best of luck. But I’m less interested in reinventing that particular wheel and more interested in pushing the storytelling envelope and getting legacy media staffs working in a “web-first” mentality. #
Edit: McCord writes, “And just to prove that our site is no worse than any using WP or CP, we are currently an ACP Online Pacemaker finalist.” I don’t know that CMS goes into the decisions of judges. Dennews.com won an online Pacemaker last year, and we’re a finalist again this year, but we’re still running on CP4. In some ways, this reminds me of the Quark Xpress/InDesign debate from 2001. It’s not the tool, but what you do with it that counts. #
Update: Max Cutler in the comments points to courantnews.com – an open-source college media cms built on Django for those who are interested in participating in a roll your own style CMS project. #
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=306ef44a-8b9d-4bb2-8701-a417bb0323bf)
A group of us college news developers has been working on an open-source Django-based CMS currently called Courant News ( <a href="http://www.courantnews.com)” target=”_blank”>www.courantnews.com). The Yale Daily News is launching on it this week, and we are hoping to get other papers on it as well this semester. I agree with you that it doesn't make sense for each news organization to reinvent the wheel, but doing collaborative efforts like Courant News can help both the organizations and individuals involved. And by having it open-source and community driven, the turnover-effect you describe can be at least partially mitigated.If anyone out there is interested in learning more about Courant, I'd be happy to talk with you.
Thanks for the tip, Max. I'll keep an eye out for that. There are several projects going on right now that I'm aware of. Not to take away from the development that's taking place on these things, but it's helpful to remember that each colleg media outlet is different in many ways. Resources are frequently limited, and I don't fault people for taking the route that best suites their situation.
A few points I'd like to make:For most student newspapers, it would be a terrible idea to build a CMS. There are several options out there that fit 95% of the needs and most student newspapers won't have the consistent tech talent they need to maintain the CMS. Even the Stanford Daily migrated to WordPress after a few years with a home-built CMS.If I'm not mistaken, The Communicator website are manually updated HTML files. A "content management system" generally stores content in a database, among other improvements. It's even worse of an idea to try to run an entire news website off manually updated HTML files than it is to build your own CMS.Considering all of the conversations we had in the last year about building CMSes vs. collaborating on code, why no links?
Some juice for your next post about a hand-coded HTML page: Madison doesn't argue for an in-house CMS, but for no CMS at all, which is, as Bryan said, a definite step backward.I've had this debate many times with editors from my former high school newspaper (which recently switched to Joomla — a work in progress). I've also had this discussion via e-mail with Madison about the Communicator site. As Daniel guessed, the Communicator site is indeed managed by manually-updated HTML files (or at least it was in March when I last spoke with Madison. A quick dive into their source code looks like it's still the same process). Knowing how to make a new HTML file in Dreamweaver doesn't put anyone "ahead of everyone else." It's the way you implement tools that is truly valuable. If the only place your skills get you is to a static page of text with some images, that's not valuable in this fast-paced industry. Why spend time learning how to do something that everyone else can do at the click of a button? It might feel nice to know you built the whole site from scratch, but the only value there is for you personally, when really you should be creating value for your readers and community. There is no database, no form of backup, no way of organizing the data, no RSS, no comments, no pingbacks… and that just scratches the absolute surface. But if the real debate here is in-house CMS vs. pre-made, I side with Daniel in that it's better to build on top of existing CMSs. In doing this, you're letting other developers worry about the basics while you focus on the tools that will really set your publication aside from the rest. Instead of wasting increasingly limited resources on maintaining development of your in-house CMS, newsorgs should be using those resources for staff training or unique plugin development (EditFlow, for example) or social media strategies or multimedia production… anything else, really…
[...] here to see the original: CMS: pre-made or roll your own? – Innovation in College Media Share this on del.icio.usDigg this!Share this on RedditStumble upon something good? Share it on [...]
All great points. I agree that pushing the storytelling envelope and getting papers to work in a web-first manner are incredibly important, but I think that part of that (pushing the ways in which journalists tell stories and provide context) can best be achieved with the experience you gain working on a CMS.Take the NYTimes – they're producing incredible media content and are mostly web-first. A smaller college paper could easily train to produce similar works even if their paper uses WordPress. But some of the truly incredible web-based journalism projects of the past few years (Everyblock, PolitiFact) require a more advanced programming skill set and experience working on projects of a larger-scale than WordPress could ever hope to provide by itself.For schools large enough to aspire to projects of that size, building an in-house CMS (or participating in a project like Courant) is the most sure-fire way for your staff to gain the necessary experience.
I agree that projects like Everyblocki and Politifact require higher level programming skills than just hacking WordPress. BUT, i don't think the way to get those skills is to reinvent the CMS wheel for a main news web site.A better approach would be to sandbox a project outside the main web site. If you want to learn databases and Django, then build something that uses those tools creatively and add it to the main site as a subdomain, or spin it off as its own domain. Politifact didn't start out as the St. Pete Times CMS. EveryBlock started out as ChicagoCrime.org.As I mentioned above, building a CMS from the ground up introduces a lot of variables that can put the main web site at risk of more headaches than necessary. Certainly, Courant is a better approach than going with an in-house project to rebuild your entire web site.Another thing I would mention here is that for many schools, when you're talking "staff" on the web, you're talking fewer than 3 people – maybe – who are dev. types, and a lot of multimedia/online folks.
Here at FIU, even though we are rolling out our CMS, the plan is to open source it and have it be the option for college newspapers that want to experiment with Ruby on Rails as a development and deployment platform. I think that's the same goal as Courant News and Populous. You can already download the source for FIUSM.com at GitHub.com
There are now and probably will be for the foreseeable future, some significant differences between what was called “the place I can go to find what I need to know” and university accredited training such as one might find is a growing variety of MBAs. Nonetheless, the “sloodle” concept of e-Learning in a virtual world that accurately reflects a global marketplace may already be here. As one example. a large private school in Indiana is offering graduate courses in just such a context even as we write and dream about what may be in the future.
Wow. I had no idea my comment would turn into a story. But if it starts people talking its even better. I want to specify a few things that I did not make clear in my original post (since i again didn't think it would be made into a story). We do not use any type of CMS (pre-made or built by us) we do manually update our html pages. I just want to address a few points brought up in the story and in the comments. The term "a step back" is being thrown around a lot, and in terms of "cutting edge technology" we may be a step back, but for us A GROUP OF COLLEGE STUDENTS, the last thing we care about is how difficult is it for us to update our pages. We just want to learn how to do it . a CMS doesn't give the STUDENT a chance to LEARN, this is backed up by Lauren Rabaino when she said, "Why spend time learning how to do something that everyone else can do at the click of a button?" It again comes back to the fact that it is important to learn the process, and when I said, "that puts us ahead of everyone else," I meant us as in the students learning the process, not the site.I will be the first one to admit that The Communicator site, or many other manually updated sites ARE NOT the greatest there are. Take for example, the site that Lauren helped with, and I believe is still working for, The Mustang Daily at Cal Poly. Built on what I think she said was a Word Press system (as I recall from our March discussions) the Mustang Daily site is VERY IMPRESSIVE. And it shows from the wall of awards they probably have in their newsroom.Had the opportunity been given to me to clear up some of the things I said to make this story more clear, I thing everyone would have had a better idea of where I'm coming from. But just to make it clear one more time:It is more important to me that my staff and the people who take over the site after us LEARN about the basics of html building. And as basic and pre-historic as our system of doing things seems, it is the way that we have done it at The Communicator Online, and we feel it is done very well. I think that a CMS is perfect for a huge daily college paper, but at our by-weekly two-year newspaper, we chose to take the time and learn html.
Lauren,Please read my response to this post at the bottom of the page. I completely disagree with your stance regarding the quote, "Why spend time learning how to do something that everyone else can do at the click of a button?" My staff and I don't do it because, "It might feel nice to know you built the whole site from scratch," but because at the time of building for what we wanted and needed as a news site, it made as much sense to let our staff gain the experience of building and managing by hand. As you know, we are a small two-year school, and 99% of the people who have or ever will walk through the doors of our newsroom will not become journalists. But by showing them how to build a site from scratch, they gain the experience to use in their future careers. And even though it may be more difficult and time consuming to do it our way, it gives us the opportunity to learn, not boost our egos.-Madison
Well put Madison, it's important for people to realize that we are only a two year school with a bi-weekly paper. And not only are most of us not going into Journalism, probably none of us are going into web development. With our site, students can learn a thing or two on the basics of building pages and will be better equipped to build one for themselves in the future. A lot of those CMS sites are really great and I admire them. But having manually updated HTML pages fits our needs and students get to learn something valuable by working with them. It's important to us that everyone can feel involved in the entire process. Marshall Moore Graphics Editor, The Communicator
A content management system does not automatically mean students will lose the opportunity to learn Web skills that I agree are very important. What a CMS does do, however, is allow student to focus on the skills that are most important to today's journalists. A very small number of students will go into Web production (working exclusively on the Web site as a developer), and even if they do they'll be expected to learn how to manage a CMS, not a static site. More students would benefit from more traditional HTML, such as learning how to code links, do basic CSS and use styles. With a solid CMS behind you it's easier to learn the basics with less pressure and then forge ahead into more advanced multimedia and interactive feature production. A CMS isn't just for a daily newspaper. It's for any Web site that wants to ensure an easy-to-use, easy-to-operate, accessible newspaper (including adding RSS). You don't want to spend all your time creating static pages, you want to spend it developing online features or writing stories.
[...] CMS: pre-made or roll your own? (collegemediainnovation.org) [...]
William,That is possibly the best response to this post I could have hoped to see. I appreciate that you understand our reasoning for doing our site this way, and I do understand your stance and reasoning for the implementation of a CMS. The way we have set up our site though is very simple in itself though. All of our archives are put into PDF formats and placed on a hosting site, Issuu.com as far as the RSS feed, we just ran out of time last year, the RSS has been built and will be implemented in a few days, and one of the biggest focusses of our staff is multimedia. By having the staff members do their pieces, all we have to do is upload it to the site. Many college newspapers use a CMS, and believe that it is the only simple way to manage a website, but if a static html site is done properly, it's honestly just as simple.As far as us having other "cool" features, our site is about 6 months old.
Hey Madison,My apologies if my first comment came off a bit troll-ish and negative. It wasn't intended as a snipe towards your website and the work you've done, but rather a small bit of frustration with Bryan re: running a post that questions whether a student newspaper should build a CMS or not. We've been talking (i.e. Bryan, I, and a few others) for the last year or so about this question and the focus of our efforts now are to consolidate student newspaper efforts around one or two open source CMSes (like Max's or Rick's). Building a website in static HTML vs. having your content in a content management system is an entirely separate question that I'd be happy to talk with you further about if you'd like. I think there's a bit of miscommunication happening in the comments.Thanks,Daniel
Daniel,Its not a problem, I could tell that there was something else that you meant in your post. The HTML vs. CMS discussion is one I am looking forward to having, I have spoke with Bryan, and it is a post I want to be involved in. It seems like I'm one of the only people who feel strongly for the use of a static HTML site. And just to clarify with you, I do know the positives of using a CMS, in my above posts, I explain my reasoning for using a hand-built HTML site.I look forward to discussing this topic with you, bryan and anyone else who feels strongly one way or another.Thanks,- Madison
As Madison noted below, I have invited him to write a post for the blog about HTML vs. CSS and he has agreed to do so. That post will be forthcoming soon.
Quite frankly, I think this whole blog post is ridiculous purely for the fact that the author neglected the bare essentials of journalism and journalism ethics because no attempt to contact McCord was made prior to publication to clarify whatever the author deemed to be unclear. The Communicator Online is hand-coded HTML. I am currently the Editor-In-Chief of The Communicator and have therefore personally seen McCord and Graphics Editor Marshall Moore co-design it. At best, the author should have e-mailed McCord as he would have avoided his mistake which has vitiated his point. Secondly, from reading these comments, McCord clearly is not arguing that he is a better journalist or manages a better site than anyone else. Rather, I think his point is that college journalists can have an equally good site and have a more enriching experience if they do it themselves. Having a good site and taking a little extra time to learn something new is not a step backwards by any stretch of the imagination.
Your concern is duly noted, and I'd invite you to read the follow up blog post by Madison as well.
[...] wrote a post about using a pre-made content management system vs. building one from the ground up (link). Madison McCord represents a different perspective on the debate. I asked him to share his [...]
[...] CMS: pre-made or roll your own? (collegemediainnovation.org) [...]
Unless its a basic site, I find it easier to build a custom CMS. If they only need a basic site though, something such as wordpress works fine.
Unless its a basic site, I find it easier to build a custom CMS. If they only need a basic site though, something such as wordpress works fine.
Unless its a basic site, I find it easier to build a custom CMS. If they only need a basic site though, something such as wordpress works fine.
Unless its a basic site, I find it easier to build a custom CMS. If they only need a basic site though, something such as wordpress works fine.