The Evolving Newsroom

Dith Pran, the photojournalist who survived and documented the devastation wrought by the Khmer Rouge in Cambodia and whose story is told in the movie The Killing Fields, has died in New York.

The New York Times has done an impressive job bringing together his obituary - including adding a recent interview with him to their video project, the Last Word - with stories about him, by him, slide shows of his images and audio of his reactions to the first public trial of a Khmer Rouge figure. (You'll need to register with the NYTimes site to see some of this. Registration is free).


Seeing his images again is a reminder of how powerful pictures can be in storytelling.

Which reminds me, NZ journalist Chris Bourke wrote a couple of posts on photojournalism recently which are well worth a read. In one he reminds us of some of the biggest moments in photojournalism history...











... and he laments the decline of photojournalism: "I despair when I see quality magazines that once championed photo journalism now making do with endless bland images from agencies. The publishers are tight-fisted, so the photo librarians cut corners (there are plenty of gritty agency shots that aren’t “posed by models”). The result is a vanilla magazine, from cover to back page."

In the other post he writes about an extaordinary interview he had in London with John G Morris who, among other things "managed the legendary photo-agency Magnum in its early years, and was photo-editor at the New York Times and Washington Post during the tumultuous 1960s".

He talked to Morris about the front page images that helped turn Americans against the war in Vietnam, about the difference in coverage of World War II and the war in Iraq - “Let’s face it, [in the Second World War] we were propagandists. The American and Allied press was not neutral, we were fighting the war and we were an instrument of propaganda” - and much more.

Well worth a look.

Nothing like a bit of Silicon Alley Insider gloom for a Monday morning. So here's a piece about US newspapers experiencing their worst drop in paid advertising revenue for 50 years.

It serves as a reasonable opener to Eric Alterman's excellent essay in The New Yorker: Out of Print: The death and life of the American newspaper. (Thanks to Nathan and Jim for the link.)

Alterman notes in the opening paragraphs that, "Independent, publicly traded American newspapers have lost 42pc of their market value in the past three years, according to media entrepreneur Alan Mutter. Few corporations have been punished on Wall Street the way those who dare to invest in the newspaper business have... The New York Times company has seen its stock decline by 54pc since the end of 2004, with much of the loss coming in the last year."

He goes on to examine the impact of the web on newspapers.

"Until recently, newspapers were accustomed to operating as high-margin monopolies. To own the dominant, or only, newspaper in a mid-sized American city was, for many decades, a kind of licence to print money. In the internet age, however, no one has figured out how to rescue the newspaper in the United States or abroad. Newspapers have created websites that benefit from the growth of online advertising, but the sums are not nearly enough to replace the loss in reenue from ciruclation and print ads.

"Most managers in the industry have reacted to the collapse of their business model with a spiral of budget cuts, bureau closings, buyouts, layoffs and reductions in page size and column inches. Since 1990, a quarter of all American newspaper jobs have disappeared.

"The columnist Molly Ivins complained, shortly before her death, that the newspaper companies' solution to their problem was to make 'our product smaller and less helpful and less interesting.' "

Along the way Alterman sweeps through the history of newspapers since their beginnings as political broadsheets and touching on the ideas that led to the press taking on the mantle of 'objective' recorders of important events.

And he does a good job characterising the somewhat fraught relationship between bloggers and newspapers, spending a good bit of time on the Huffington Post, as seen in this excerpt:

"In October 2005, at an advertisers' conference in Phoenix, [New York Times executive editor] Bill Keller complained that bloggers merely 'recycle and chew on the news,' contrasting that with the Times' emphasis on what he called 'a journalism of verification,' rather than mere 'assertion'.

" 'Bloggers are not chewing on the news. They are spitting it out," Arianna Huffington protested in a Huffington Post blog. Like most liberal bloggers, she takes exception to the assumption by so many traditional journalists that their work is superior to that of bloggers when it comes to ferreting out the truth. The ability of bloggers to find the flaws in the mainstream media's reporting of the Iraq war 'highlighted the absurdity of the kneejerk compairson of the relative credibility of the so-called MSM and the blogosphere,' she said."

Well worth a read.

I've had a few interesting conversations this week about quality control in newsrooms, in part sparked by the diagram I posted last week showing how many pairs of hands various kinds of news stories go through before being published.

One thing that came up was how irritated people are by the number of typos regularly seen on news websites - including nzherald.co.nz and stuff.co.nz. As Nathan said in a recent comment, "I generally expect better of professionals."

I haven't worked at either of those sites so I can't speak for what goes on there. But in my experience the problem is twofold: inadequate systems (Content Management Systems (CMSs) that require users to undertake multiple steps for simple tasks - increasing the likelihood of mistakes); and poor workflows that were designed decades ago for print and haven't been revised to account for the website.

Some typos simply come from typing too fast - you'll find plenty of these in blogs and comments (including mine) where people type quickly and give only a cursory check before hitting 'send'.

More prevalent in news publishing are cut and paste errors - where a sentence has been edited or moved and during the process a word/punctuation mark/sentence has been inadvertently cut out, cut in half, transposed or reconstituted without due care for syntax or grammar. (That's why you see double words in text and missing full-stops).

How does it happen? Easy. Take a breaking news story. They're often bits of wire stories spliced together with a few paragraphs filed by an inhouse reporter. Very often those paragraphs will have been filed by email and copied and pasted into the web CMS. Just as often, the wire story will have been copied and pasted into the web CMS.

Take all that copying and pasting and re-working of sentences, throw in some time pressure and you've got mistakes waiting to happen. Speed is of the essence so the story goes live without anyone else reading it. Even with less urgent stories there's often only one duty web editor anyway, so there's no one else around to have a look at it before it goes online.

So why doesn't someone read it after it's gone live? Good question. Some structured peer review could go a long way here. And as I keep saying, readers would gladly help if you made it easy for them.

Most stories, though, are prepared for the newspaper first before finally being loaded online late at night by a tired person. On the way, they'll go through nine or more pairs of hands, typically, and two CMSs.

Most likely both CMSs will make hard work of cutting, pasting, categorising and styling text - increasing the likelihood of mistakes being made along the way. Cyber, a system widely used for print in NZ, would fall into this category I suspect, based on my brief encounter with it. If you have a different view, I'd love to hear from you.

What's the answer? One, news organisations need better CMSs which have been designed for neutral input and multi-platform output - no more cutting and pasting, easier editing and mark-up, and the fewer steps the better.



Two, they need to take a critical look at their workflows and redistribute resources: essentially take a couple of those people in the right-hand columns and move them to the left.

A few tips on using social media from BBC blogger Robin Hamman:

"Social media isn't something you add to a website, it's something you do. When I look back over the social media projects I've been involved in over the years, it's obvious that the key variable upon which success, or failure, is dependent is to what extent social media has actually been integrated into the overall editorial proposition."

Among the examples Robin gives are:

Works: A chat room where the presenter dips in and out of the conversation, reads comments from the chat on air, uses the chat room to actually drive part of the programme works because the effort expended in hosting the chat is well spent.
Doesn't Work: A stand alone chat room fails because someone with better stuff to do has to sit there and moderate it.

Worth a look.

Another one for the outsourcing archive: a Fort Worth, Texas, newspaper, the Star-Telegram, is joining the ranks of US newspapers outsourcing advertising artwork to India. The paper is transferring 26 ad artist jobs to a US company working out of New Delhi.

And here's one for the 'well, that makes sense' archive: AP is consolidating its print, broadcast and sales divisions. 'in an effort to streamline its business dealings across media formats'.

On a similar tack to yesterday's post about when a story is 'finished' enough to go online, I was reminded while planning a sub-editing lecture recently of how lopsided today's newsrooms are when it comes to handling stories for web and print.

The image below is a rough snapshot of how many pairs of hands various kinds of stories might pass through in your average newsroom - from reporter through chief reporter, news editor, web editor, chief sub-editor, page layout, sub-editor, check sub, proofreader, editor, edition controller and so on.

Hard to imagine anyone allocating resources like that if they were designing a newsroom today, isn't it.

On the upside: there's lots of pairs of eyes on those newspaper and shovelware stories to catch mistakes before they're published.

On the downside: there's lots of pairs of hands on those newspaper and shovelware stories to introduce mistakes before they're published.

Not to mention the extraordinary length of time it takes to get a story online using the shovelware method.

Time for a rethink, surely. A rethink, a critical analysis of newsroom processes, a redesign, re-training and roll-out.

I meant to link to this a while back. For those of you interested in mobile developments, check out the BBC's latest offering of the mobile version of bbc.co.uk.

Shane Richmond (the Telegraph's Communities Editor) does a good job revisiting the issue of what 'finished' looks like in the fast-paced world of digital publishing.

In print, a story wasn't published until it was 'finished': written, edited, edited again, sub-edited, sub-edited again, proofread and so on. Only then did we hit 'send' and hand it over to the printers - at around 9.30pm, the deadline for the first edition of a daily newspaper.

But, as Shane points out, it wasn't so much that the story was 'finished', more that it was "as good as it could be" at the time the print deadline rolled around.

"The years of habit, the mere notion of a final edition, have created the illusion that news stories can at some point be considered finished," he says.

"All we ever do when we publish a newspaper is run out of time. Given more time there would always be more to add to a story, more angles to pursue and newer stories to find space for. We can never be completely correct, only as correct as possible in the time available. We check as much as we can but we still get it wrong sometimes, all of us.

"The internet removes the time barrier. Without it, we never stop, and still we're never completely right. But everything can be changed. Publish what you know now; learn more, add more. It's never finished and it's never completely right."

Worth a read. As ever.

I've been reading Jason Ryan's blog post about experimenting with the use of semantic mark-up in government press releases. I think it's well worth a read for two reasons: one, he does a great job explaining it in simple terms, and two because it's a development that's nice to see in NZ and worth watching.

I'll be honest, this is an area I don't know much about. Web code for me is like the language of a country I've visited once or twice - I recognise phrases and get the gist but can't speak it very well.

But I'll hazard a broad outline: in a nutshell, semantic mark-up adds information to the underlying code of web pages which makes them easier to read - for people and computers.

The extra markup might mean, for example, that people with vision impairments who use screen readers can see more of the page (images, pointers and directional headlines are spelled out as well as the main text).

It could also mean that pieces of information on the page can be retrieved for further use - such as plucking out key dates and adding them to a calendar, or populating news/updates/events sections of other websites.

Jason points to a Firefox extension (due to be part of Firefox 3) which "allows you to click on microformatted information in web pages and it will initiate application sequences. So, in the case of hCard, clicking on the microformat will open your address book (in Thunderbird or Outlook, say) and you can save all the information then and there. Geo formatted information will open Google Maps and show you exactly where the place or event is, hCalendar will open your calendar application and allow you to save the event."

Bring it on.

Interesting to see this taking shape in the public sector (Jason edits the Network of Public Sector Communicators weblog) - government, in other words. Will the news business keep up with the pace?

"We were being asked to do things that we just couldn't bend the system any further to do."

Sound familiar?

That's PA's IT development manager Paul Berman talking to Silicon.com about the company's decision to use an open source platform, Nuxeo, to build a better content management system.

Like everyone else on the planet PA is handling an increasing volume of video, audio and slide shows as well as text and they need a system that can manage that simply and efficiently.

Anyone who works in a newsroom will know that the legacy systems we have simply aren't up to the job. Older web CMSs weren't built for multimedia in any volume and no matter how many scripts the good people in IT write to make the ageing print CMS talk nicely to the ageing web CMS, it's not a smooth process. In most newsrooms reporters still write into the print CMS.

Invariably getting something online involves multiple steps, things to remember, repetition and miles of mouse clicks and keystrokes.

That makes it difficult to distribute the burden of marking up and loading web copy among authors and editors, who are the natural candidates for the role in a web-first operation.

If it's a really complicated process, they're not going to want to do it, might not understand it well and cut corners, and will be distracted from the business at hand of researching and producing good stories.

"We have some really very specific requirements for our editorial applications in which we want to deliver a very effective tool for our journalists to do things with as few keystrokes as possible, efficiently and quickly," Paul Berman told Silicon.com.

On the reason for choosing open source, he said: "It's a mixture of the flexibility, the cost and the potential to scale it and make it really adaptable for our environment."

Interesting to see a major news group choose open source given the customary preference in such organisations for proprietary 'best of breed' systems.