It's just in the service of journalism
Earlier when I argued that news sites should regularly change the design of their content area, I was following up on a theory that readers get bored of “sameness” and don’t realize there is much new on a site without new content — and especially a slightly different look for that new content — on the front page.
That’s not the main reason I’m behind more flexible layouts for news site front pages, though. As a journalist, I just want a site’s front page design to be able to accommodate my news organization’s best content at the moment, rather than restrict the type of that content in order to satisfy a static front page design. For instance, the content area of nytimes.com almost always looks the way it did about 10:30 tonight, with a vertical photo on the right side:
Just two hours earlier, though, the Times was using a strongly horizontal photo of snow in Central Park that would not have retained its impact had it been cropped into a vertical. In order to use it, the site altered its typical front page design, to put the wider photo in a wider column:
This is the exact kind of flexible design that I was recommending. It is in the service of journalism, letting the site showcase its best content. Many people commenting on my previous post asked a very important question: Will a change in the layout of a content area confuse readers? There’s no personal doubt in my mind that the vertical-photo design looks prettier on nytimes.com, but I can’t imagine the switch from one to the other of the above designs will “confuse the hell out of most visitors,” as one person commenting put it. What I can imagine is that most visitors will take a second glance, figure it out, and be exposed to the best photo nytimes.com had to offer at the moment.
I think the Time's front page is a perfect example of what I was talking about.
I don't think we give readers enough credit sometimes. The visual language of big headline indicates the lead story, whether it's in a paper, online or in a magazine.
One of the most important principles in design, that some "Web designers" forgot is that good design is always content-driven.
(The "Web designers" I'm refering too are the ones who do cool things because they are cool, and for no other reason.)
A horizontal photo is a horizontal photo, making a vertical to fit the design breaks the principal tenet of content-driven design.
I think most sites, and most CMSes, should be flexible enough to create several standard templates for differnt types of content -- strip with no art, strip with art, vertical lead with vertical art, horizontal lead with vertical art, etc.
We're not talking about 365 new designs a year for the rest of the sites life, but a few different ones that reflect the change in content.
The Times example is amusing -- Nathan, you're not old enough to remember it, but back in the days of 8-column newspaper layout, the joke about the New York Times was that all of their photographers had broken wrists, forcing them to hold their cameras in a vertical mode at all times. The newspaper seemingly would go for weeks with the same 3-col by 7- or 8-inch photo position on the front page.
I think all Web sites have three basic homepage components: branding, navigation, and recommendation. The first two should very rarely be altered. The third should be driven by timeliness (urgency and appropriateness to the moment) and the content that's being recommended.
That said, it's possible for relatively static geometries to work well for news -- look at MSNBC as an example. By altering the relative positions of image and type in the lead graphic, the designer can respect the integrity of the photo without actually altering the page layout.
We adapted this approach for the Charleston Post and Courier http://www.charleston.net/ -- a commercial customer of our design studio, not a Morris newspaper. The paper has only a two-person online unit, but it has been able to execute the central graphic very well. The site always looks fresh because content is central to the presentation.
We have created a homepage management tool, called SiteStitcher, that lets a fairly nontechnical editorial user swap out various layout modules, and we have a companion facility for scheduled layout changes (dayparting). Our upcoming site redesigns will take more advantage of that functionality.
Steve, that’s a great point about the print Times from another era. I guess it was an era in which the technology hadn’t yet matured enough (to cold type and then desktop publishing) to have any kind of “graphic designer” role in a newsroom. Maybe I just need to wait for our online medium to mature a little bit more.
Can your CMS do different layouts?
Chris Heisel says news Web sites ought to change the design of their content area regularly, rather than having the same look every day:
The system that most newspapers use for updating their sites involves pushing a new headline, blurb and maybe a photo to the front page.
This would be the print equivalent of running every story in the same position as it was the day before, and just putting new headlines and new photos on the page.
Sites need to change their visuals so visitors know that the site has been updated, Heisel argues, citing an E-Media Tidbits post saying readers get bored of “sameness.” And that doesn’t just mean new photos, that means switching to a different layout and changing the sizes and shapes and positions of stories and photos. This is what I tried to do back at my college paper’s online edition: compare the Web front pages for Feb. 14, 2002, March 28, 2002 and April 18, 2002.
A few days ago I complained that many sites did not change their layout for a big story, the loss of the space shuttle Columbia. I theorized that rigid content management systems were behind some Web sites’ ordinary look and feel on a news day that was anything but. Heisel also believes content management systems are one reason most news sites don’t change their design on a daily basis — or preferably several times a day.
I agree with Heisel that replacing those dumb content management systems with ones that support many different front page designs — or better yet, working with raw HTML for maximum flexibility — will increase reader interest in news sites. That’s a very good reason to make your front page more fluid on 364 days of the year, just in case you weren’t convinced of the need to use a banner headline the day the space shuttle disintegrated.
How do you balance flexible layouts with the tendency for users to scan Web pages habitually, expecting stories in the same place?
I kinda like the idea of changing the homepage on a day to day basis. Say, on Friday cover entertainment and 'stuff to do' more prominently and on Mondays play up the big feature stories from Sunday (for the people who only surf at work...)
Doing this, you could let readers know what to expect on certain days while still not making it the 'same' everyday.
At some point the 'industry' needs to take the web as seriously as it does print. When they do, maybe they'll invest on some real innovation for CMS systems. Until then, expect run-of-the-mill turnkey solutions heavily molded (moldy?) in certain forms/designs.
The problem with working raw HTML is that you have to devote manhours to it - something a lot of media companies are loathe to do at the moment. (Personally, I think it's because of the looming lessening of cross ownership laws - ie a media feeding frenzy later this year...) While I don't agree with the strategy, people tell me it's called 'reality' ;)
Adrian, that’s a very good point to bring up about scanning. On the other hand, I think readers scan print edition front pages, too, before deciding what to read, and no one complains about stories and photos being shifted around from day to day in print.
But maybe the “cognitive burden” factor that you’ve identified trumps the “boringness” factor online. I know I generally spend several minutes with a front page in print (because I begin reading stories on the page) whereas an online front page might only get several seconds of my attention. And you’re right, I have trained myself to look for stories on the right side of washingtonpost.com and the left side of nytimes.com, because of where those sites put their photos. The New York Times did change the position of their photo for about a week recently, though, and that didn’t confuse me too much.
Here’s one potential compromise: What if sites had a consistent “story area” and “photo area” but allowed for more variation within those spaces? Rather than just containing a single vertical photo, for example, the photo area could also allow two horizontal photos. The story area would need to allow for several different headline sizes including banner headlines, grouping of related articles, identification of stories as “special reports,” as well as the use of one or two smaller photos to go with specific stories.
Kpaul, I was originally only thinking of varying the design to swap different pieces of content in and out, but you have an interesting idea there for swapping different types of content in and out as well. Especially if you’re talking about putting entertainment news up prominently over the weekend or in the evening or something, there would definitely need to be some design elements that clearly convey the difference between a breaking international story and a concert preview.
Barry Parr has a good summary of what types of content are accessed at different times and on different days. I’ve heard some talk recently of changing around news site front pages to reflect that.
I read your article in Dr. Web`s Magazine.
On the site www.bullfrog-bulldog.de you see my ideas concerning this.
Best wishes for the future,
smiley by MaJa
I think it is worth while remembering why CMS was developed, the idea was to offer a possibilty of filling in content without he CI of a Company. That was in my opinion one of the reasons to develope CMS. And therefore it does not make sense to me critizising this wanted effect. If it is meant to be another area of CMS some sort of 'evolution' this would be a good idea, but to use exactly the reason for developing it does not make sense to me.
I do appologize for my bad English, but im German and not in practice of speaking and writing.
why not to change a Page like you did. because the user are confused when they see the page next day. there is no design straight line to follow, you just changed the order of topics and pics. you cant call this design change. what you must do is a special color for "tuesday", eg. all links have this color on "tuesday". of course you can change the order of design but a little bit more recognizable. dont confuse your users with changed order of articles. (my english is not the best but i am learning) bye daniel
Sigrid, outside of the newspaper industry I think you’re right that “filling in content” is a task that can fall to non-designers. But my opinion is that the staff of news Web sites need to have the tools (and the training) to design a front page, not just throw it together. You’re also right that it’s foolish to criticize a content management system for operating exactly as planned — I’m taking issue with the plans themselves, not the programming. As kpaul alluded to, sometimes it seems to be business decisions that get in the way of the journalism.
cms were not only developed for keeping up with the CI of a company. another reason was to make the process of content-creation simplier - in a way it's what blogs are for today for smaller pages.
so going back to coding raw html cannot be the solution.
but changing the style of a big entrypage depending on some event (there are: planned events - e.g. olympics, elections; and unplanned catastrophes - e.g. columbia; ...) is an important issue.
cms are to blame, but there are ways to overcome their deficits without going back to coding html (once more mixing up content with structure and design).
Everybody talks about Content Management, but most of the CM-systems also try to manage the structure. Structure Management is much more complicated then Content Management. If a client ask a specific type of structure on a website, individual programming is necessary. This is the moment, where most of the CM-System fail. Some few CM-Systems have the ability off managing stucture. But most of them are so complicated to handle, that we have to ask ourselves, wether it is not easier to learn the programming language itself...
While it is true that some layouts get boring with the time, changing Layout regularly is for sure going to confuse the hell out of most visitors. As a webdesigner and webprogrammer of CM-Systems myself I like to play around on websites, but I think that is not the case with most of the actual customers of a site. When I go to the big news sites I go there with getting my information as quick as possible and not to play a round with some new feature. Worse I should not have to learn the navigation everytime I go back to a site.
I think the solution to this would simply be for the CMS to only spit out the markup that is really needed and let the *CSS* do the rest. Not beeing able to change layouts imho has really nothing to do with CM-Systems.
All the best
This is a very interesting discussion and I don't think this issue is limited to "news" sites. I can't help but thing how this relates to an Intranet site, where you may have many different types of content, news included. I know that for the CMS I use down at my day job, there would need to be HTML manipulation to achieve this so in effect either the Web team would have to maintain it or train the content owners. Maybe some sort of content template would solve the problem, but I can't help but think it might be more trouble than it's worth.
What about blogs? Could they benefit from this as well? Hmm... In any event it's interesting to think about.
I suggest to have a look at typo3 (www.typo3.com). This Content Management Framework is the most flexible application for content management i've ever seen. Check it out :-)
Sasa, you have a great point about using CSS to provide all of the site’s design. That separation of design from content has many benefits to news sites, as my friend Adrian Holovaty has repeatedly pointed out. And if the CMS we use at work could spit out plain content ready to be styled with CSS, you’re absolutely right — I’d quit complaining. Perhaps there is hope for this taking off in the news business eventually, as Adrian just reported on ESPN.com’s new CSS-driven and almost tableless design.
Shure. A CMS should never stop any layout- or designideas.
Here are some examples for using the www.CONTENTCENTER.de, written by myself:
The separation of the content ("contenu") and the look /design("contenant") have a recent evolution on the Net with the use of CMS tools. I think that it is more simple today to have a good design, but there is more a long work to adapt the good design for the content (CSS is one way) ...
I do appologize for my bad English and for most french articles on WEBPublication.org
Found this blog recently.
Maybe I miss the point.
As a matter of course a CMS should enable quick changes in layout, and should be based on CSS, as was pointed out here.
But does it really make sense to change layouts say every day? Or let each editor decide what color he likes today?
If you go for corporate identity or similar, your web design should be constant and recognizable, IMHO.
Thomas C., for most kinds of Web sites I agree with you. However, what I’m interested in here are online news sites, which generally do rotate content every day or multiple times per day. Some pieces of content need to be displayed in slightly different ways than others. Yes, the few interchangeable designs a site would come up with would be consistent and recognizable. But that doesn’t mean the “corporate identity” of a news site should require always using a horizontal photo or always using a vertical photo. Gratuitously changing design just to make existing content look new is unfriendly to the user, but changing the design to show off your new content most effectively is not. I’m not advocating changing colors and navigation all the time — that would be horrible — but I am advocating the ability to present two stories and two photos one day and three stories and one photo the next.
A different kind of media consolidation
Last night at work, one of the automated processes in our content management system broke down and seriously slowed production for several hours. The point of saying that is not at all to bash the IT department — problems do occur, and they got on it right away.
The point is that automated process happens to run on a server hundreds of miles away at our corporate parent’s Web hosting facility. It’s a process that is also used by, I think, about a dozen other news Web sites in our chain, who were affected by the same problem. Luckily, things recovered enough by the time I finished working on the site for the evening for readers never to know the difference.
But what if?
- What if this problem wasn’t resolved soon enough last night?
- What if this problem affected more of the site than it did?
- What if a serious problem prevented me from updating the site at all?
- What if an even more serious problem prevented readers from accessing the site?
- What if any of those problems had occurred during a big breaking story?
- What if there was a catastrophic failure at our hosting facility — a fire, a flood, or even a terrorist attack?
While the probability of a catastrophe is incredibly low, the probability of some of the other problems is not. Of course, having a centralized hosting facility for many online news sites probably reduces the likelihood of failures, because greater efficiency allows more resources for the “what if” scenarios than any one site could muster alone.
What worries me about centralized news site hosting, however, is that even if the likelihood of a problem is reduced, the impact of any problem is increased. The issue I had last night affected not just the site I work for, but news sites in six states. Traditional media are fairly resilient to failure — if one newspaper’s press breaks down or one television station’s transmitter fails, other media outlets are not affected. But a failure at a large media company’s Web hosting facility could disrupt online distribution of “local news” on the other side of the country.
We who work for news Web sites want readers to depend on us for information, so we need to fulfill our end of the bargain. We need to make sure our medium still operates in the event of minor problems or catastrophes, rather than failing. The Internet, with its famously distributed architecture, is designed to do exactly that. But a centralized architecture of hosting many news organizations’ Web sites in one physical location on one content management system is not.
It’s often argued that media consolidation into a small number of large conglomerates will bias journalism and reduce media diversity. But — in the event of a catastrophic problem at a conglomerate’s Web hosting facility — it’s possible centralization could temporarily but significantly reduce media availability. Readers deserve better.
Can your CMS do 150-point headlines?
Take a look at Poynter’s nationwide collection of print edition front pages from Sunday, Feb. 2, the day after the space shuttle disaster. Some of the bold banner headlines read: “Columbia is lost,” “It’s Gone,” and “What Happened?”
All of these headlines are big — almost universally over 100 points and probably almost 300 points at several papers. There’s a good reason for this giant type, and it is not just increased single-copy sales the next morning. It is about using design to give readers context. The headlines’ words are about the shuttle tragedy. But what their big bold blackness really says is, “Pay attention. This is one of the biggest stories of the year.” I believe readers are not only acquainted with this system, but expect it: If there’s no enormous banner headline on the front page, then it was a fairly normal day in the world yesterday and I can move on to the sports section and my bowl of cereal.
Did online newspapers take advantage of this well-established design principle on Feb. 1, using banner headlines to communicate the story’s importance? Generally, no. In Poynter’s gallery of online front pages from that day, there are a few news sites that played the story appropriately for its magnitude: nytimes.com, tbo.com, washingtonpost.com, latimes.com and daytondailynews.com, among a few others. But many sites treated the story like herald.com, ocregister.com and even cnn.com — switching the shuttle story into the top position on their front page but leaving their site with a standard design. Nothing about those sites’ look that day — not a very large headline, nor a very large photo, nor a very large set of related articles — conveyed the importance of the story the way newspapers’ print editions did the next morning.
Of course, screen sizes force headlines on the Web to be far smaller than the largest headlines in print. And some Web sites — perhaps ones with a local focus — genuinely chose not to use a banner headline. I support a conscious design decision like that, even if I probably wouldn’t have done it that way.
But I suspect that at more than a few news sites, online producers did deem the Columbia story worth special treatment until their content management systems stopped them. Too many content management systems dictate an inflexible design on news sites’ front pages. Photos can only be certain shapes and sizes and placed in certain positions. Teasers to related articles can only go in certain places. Packages for special projects or big news stories can only be created in certain ways. Headlines can only be one of two formats, and they certainly cannot be a 36-point banner headline. Such are the constraints of our content management system at work, which is why we built most of the front page manually on Feb. 1 and three days thereafter.
However, many other sites run with content management systems did not or could not shift into manually-coded mode that Saturday. I believe the rigidity of those systems prevented some news sites from telling the disaster story appropriately. A Web front page should be an empty canvas to paint on, not a pile of predefined pigeonholes waiting to be colored by number. That unimaginative coloring by number shouldn’t be acceptable any day of the year. Certainly not on the day seven spacefarers died pushing the limits of human possibility.
I hear ya and definitely agree with you. Alas, though, I don't think the big companies 'get it' yet - other than the revenue numbers...
Print papers, are largely manually laid out - particularly front pages. This happens whether or not there's a '150pt headline' day or not. So there's no additional effort involved in making that change.
Unfortunately, web editions are not manually laid out, and newspapers cannot afford to do so - even if you reduced them down to a small number of editions a day (rather than the continuous publishing flow demanded by users). The cost of the people required to do that reliably without screwing up is not sustainable, especially as people don't pay for online editions.
You're right - companies only get the revenue numbers. That's why they exist. They're legally mandated to maximise investor returns.
It's very important to realize that the newspaper is a printed medium with 100+ years of display history that has evolved very slowly. So readers understand what headline size means. Also, of course, headline size is important for selling newspapers from racks, but I agree that is not the newspaper's major reason for the very large headlines. Large headlines in newspapers are studiously tracked ... "we did 125 point on that story, so this story deserves 150 point", and so on.
Online, I am not convinced that LARGE headlines are needed or really meaningful. Online front pages - no matter the site - present a summary of the inside content. The presentation is offerings, not the full story. Very large headlines can actually get in the way of presenting information (pushing good content too far down on the page, for example).
But your point is true about the need for a 'clean slate' that allows news Websites to abandon the usual in the case of the very unusual, such as a major story like the shuttle crash.
One point about Websites and database tools is that the existing tools are probably best for very fast udpates. In other words, if you want to get a story up fast -- important on the Web -- you should use the existing publishing system. Get the story up now and worry later how it looks (within reason). I would prefer fast, accurate reporting over impressive display in the first several hours of a big story.
After the first few hours, display becomes more important, and the same-size-fits-all display hampers many news Websites.