A brief sketch of the “full stack” (intellectually speaking…) news and information company.

Meaning: it has its own way of doing things, its own ideas about what is worth doing, and it implements them from end to end.

21 Jan 2015 6:04 pm 12 Comments

I was talking to a media executive the other day and he mentioned his ambition to create a “full stack” company. This is a software term. Full stack refers to the different layers of tech that when combined make for a workable product. A full stack developer is competent at all these levels, from server technology to user experience. According to this Chris Dixon’s post, a “full stack start-up” is one that tries to control all the interlocking pieces. He names Buzzfeed and Netflix as two examples of successful full stack companies.

“Suppose you develop a new technology that is valuable to some industry,” Dixon writes. “The old approach was to sell or license your technology to the existing companies in that industry. The new approach is to build a complete, end-to-end product or service that bypasses existing companies.” A full stack start-up can “completely control the customer experience, and capture a greater portion of the economic benefits.” The hard part is that you have to be good at so many different things:

Software, hardware, design, consumer marketing, supply chain management, sales, partnerships, regulation, etc. The good news is that if you can pull this off, it is very hard for competitors to replicate so many interlocking pieces. (More on Dixon’s concept.)

As I listened to my media executive talk about owning the content management system, and the content itself, and the analytics tool that tells you how users are interacting with it, and the user experience layer, I thought: “full stack company… what a great metaphor.”

So let’s do that: Let’s push the metaphor. OriginalPancakesImagine a newsroom and information company, a journalism site, that is full stack, intellectually speaking. Meaning: it has its own way of doing things and thinking things, its own ideas about what is worth doing, and it implements them from end to end. From defining the editorial mission to deciding what constitutes “news” to designing the look and feel. Instead of borrowing what the industry does, it makes these products itself, and not just at one layer of the enterprise, but at all the “thought” layers.

Bear with me as I try to explain.

Grantland: what is the niche? You tell me! You can’t say “sports” because it’s more than sports. You can’t say “popular culture” because it’s so heavily grounded in sports. It’s more like: sports, plus what shows up when you map the gravitational pull of sports. That’s the niche. But that niche isn’t borrowed from anyone. The industry didn’t make it. Grantland made it. This is the beginning of a full stack company in news: not a borrowed beat, but an original one.

ProPublica: what is the mission? Not to be “the number one provider and news and information” in blah, blah, blah region. Not: everything you need to know about… Or “all the news that’s…” No. It’s more tightly defined than that:

Our work focuses exclusively on truly important stories, stories with “moral force.” We do this by producing journalism that shines a light on exploitation of the weak by the strong and on the failures of those with power to vindicate the trust placed in them.

From whom is that statement of purpose rented? No one! They made it themselves. “Journalism that shines a light on exploitation of the weak by the strong” is a piece of pressthink original to the editors and reporters at ProPublica. People who work for the Associated Press or the Washington Post might like to think that they got into the business to “shine a light on exploitation of the weak by the strong,” but that is not how their responsibilities are defined.

So imagine if every (intellectual) layer in the newsroom “stack” were made that way: original to the editors and reporters involved. What would that even look like? Here it helps to imagine the extreme opposite, where every layer of coverage is derived from the industry standard, from current practice, from the way things have always been done, from what others are thinking or will soon think. Pack journalism, in other words.

A good example is Bloomberg’s new politics vertical. It’s almost impossible to find a more consensus mind than the mind of Mark Halperin, co-managing editor of Bloomberg Politics. (With an annual salary said to be north of $1 million.) His stock-in-trade is conventional wisdom, packaged for easy consumption. Halperin is like the first essential layer in a “collapsed stack” newsroom: the guy you would go out and get if you wanted to do exactly what everyone else would later think of doing.

The “full stack” (intellectually speaking) newsroom is populated by anti-Halperins: People who have their own ideas about what is worth covering. They command beats and produce stories that don’t obey pre-existing industry categories. The mission is different, too. The sections of a full stack news site will section off the news in a way you isn’t familiar to you from your grandfather’s newspaper. In a full stack newsroom, the code of conduct that prescribes and proscribes what individual journalists can do contains a lot of original programming— different from what students might learn in a typical J-school.

My point is: if you want to succeed in news and information provision, a smart play is to go “full stack” on all your competitors, intellectually speaking. That means defining the beat the way no one else defines it, and coming up with a mission that differs from the industry standard. If you’re not willing to go it alone, your best bet is to admit to this up front and then compete for scoops with dozens of others who are trying to score in the same way that you are trying to score. If that’s your game, then own it.

Photo credit: Jack and Jason’s Pancakes.


Quartz’s “Obsessions”?

A good start.

I think this post represents a very good exercise. (A similar exercise is “thinking about your ideal journalism organization”).

A thought: The “full stack” metaphor/hypothetical is complicated (though not negated) by organizations’ reliance on user-generated and outside content. For example, if you’re a full stack journalism organization that exclusively uses the AP for your images, are you really full stack? Obviously how you choose to display and use those AP images will deeply reflect your unique journalism mission, but unless that mission is Remixing Outsourced Journalism (Reported.ly, perhaps), one could argue that you’re not really full stack.

There might be a key difference here: in the developer world, it seems (and correct me if I’m wrong) like there are very clear and precise definitions of what constitutes “full stack.” (Obviously, different developers will have slightly different competencies in each part of the stack.) But in the journalism world — and what you’re advocating, it seems — is that being full stack means being truly unique and mission-centric, not arbitrarily sticking to industry conventions.

Is a proprietary CMS a requirement for full stack? Does full stack and the ambitiousness associated with its model ultimately only benefit and work for a select few (and big) organizations? Are freelance journalists full stack in the sense they do everything — marketing, bookkeeping, working, website building, etc.?

I think the mindset associated with this is great: it’s not necessarily throwing the journalism rule book out the window, it’s more like using the Spark Notes (Mad Libs?) version of the journalism rule book. Each rule is limited to only a sentence, and you write the rest.

Thanks, Veasey. I guess what I’m really saying is there’s very little gain these days in being a commodity producer of news and information. That point has been made a lot, by a lot of people. But then I am saying one thing more. The work of getting “commodity production” out of the value chain has to extend to everything in that chain, including the intellectual work— the pressthink, if you will. Thus, full stack.

Cool. Yeah. Some of that stuff (“commodity production,” “pressthink©”) might have gone over my head, so hopefully I’m not butchering this discussion topic.

Obviously it helps if your organization is firing on all cylinders and working toward a common goal and talking to each other — not only when it makes sense in the workflow, but also during the other moments. It helps for different people to brainstorm stuff and bounce ideas off each other.

To me, the most exciting thing about full stack in journalism is it seems like it allows more room for experimentation. I think it’s a tricky balance between trying to constantly innovate (which demands failure) and trying to develop a workflow that allows for consistency in a daily news environment (which demands success). It helps to have enough people to be able to work in both realms at the same time, and the right kind of people that they can switch seamlessly.

I think there’s a danger of “mission creep” in a full stack culture. (And I’m not talking about sponsored content. I don’t really mind that.) There’s much to knock legacy media for, but I think they were pretty clear about what their goals were and what you could expect from them. ProPublica seems to have a pretty fleshed out mission. But does Buzzfeed? I have no idea what Buzzfeed News’ goal is. Report everything around the world?

Christopher Krug says:

Veasey Conway,

A CMS supports verticles. A CMS is “full stack”. You follow?

Here’s more info: http://www.lynda.com/Web-CMS-tutorials/What-CMS/74535/82417-4.html

Dan Mitchell says:

Well, again, these theories are all great and everything, but they still don’t address the core economic problem afflicting newspaper-style journalism, which you seem to think isn’t very important, given the fact that when you’re not expressing disdain for it, you’re ignoring it entirely.

I like that we can now have sites like these — ProPublica in particular. And these ideas are fine for sites like that.

But no “full stack” is going to replace ongoing, daily statehouse coverage, city hall coverage, coverage of local business and labor, coverage of schools, coverage of cops and courts. For that, we need something like, yes, “the No. 1 provider of news and information in blah blah blah.”

That “blah blah blah” you wrote there? Those are cities and communities where people actually live and work and where the actions of local and regional power structures affect those people’s lives most directly. Blah, blah blah.

None of this stuff means much unless you can map it to how it might make up for the astonishing loss of coverage (in both amount and quality) of those topics and institutions I listed above (not to mention a whole bunch of federal agencies).

“Mainstream journalism” isn’t tantamount to “coverage of national retail politics.” Not even close. Mark Halperin is a hack, but he’s essentially meaningless. He’s easily ignored. The vast bulk of newspaper-style journalism, and the most important kind of it, is (or was) the kind directed at the local and regional issues and institutions I mentioned above. And that’s precisely the kind of journalism that has been harmed the most — decimated — by the Internet’s unbundling.

And yet, that’s the kind of journalism that the theoretical, theory-driven theoreticians tend to ignore the most. The irony is that most such theoreticians like to pass themselves off as champions of the people in the face of arrogant old-media gatekeepers (blah blah blah) even while worrying over Mark Fucking Halperin or what’s happening with the Washington Post or the New York Times. Irony so thick it’s impossible to see through. We need to worry over what’s happened to the St. Louis Post-Dispatch, The Miami Herald, and the Seattle Times.

Way fewer working reporters are, on behalf of “the people,” monitoring (just for one example) local schools than were a couple of decades ago. And what are we supposed to do about that? I don’t think Twitter and comments sections are sufficient – or even serious – answers. This most important aspect of journalism has gone essentially ignored, because the theory class finds its leisure in pretending that Mark Halperin represents what has gone wrong in media. He represents what has gone with the small sliver of media that obsesses over Hillary’s plans for 2016. So he’s obsessing over her, and you’re obsessing over him.

Meanwhile, maybe some contractor pal of an Illinois state legislator just got a billion-dollar paving deal, and nobody will ever know. Or maybe the salaries of teachers in Michigan have been sinking below the national average for a decade, but nobody has any idea. Or maybe prisoners in Arkansas are being systematically abused, with nobody to tell us about it. But who cares about that boring shit, as long as we get to make fun of Mark Halperin for being a tin-eared spewer of conventional wisdom, and we get to talk about how Twitter is the new newspaper. What’s really important is what sells — another deep irony, coming as it does from a purported champion of the people.

I like how you demand a solution to a problem that no one has the solution for —because it’s a really hard problem — as the price of your willingness to take seriously anything else one has to say.

No, I don’t how to deliver a workable business model to fund “ongoing, daily statehouse coverage, city hall coverage, coverage of local business and labor, coverage of schools, coverage of cops and courts.” The reason I am not solving that problem in this post is that I don’t know how. Do you?

If you do, then spell it out, man. If you don’t have the new model for how the St. Louis Post-Dispatch, The Miami Herald, and the Seattle Times can rebuild their reporting staffs after years of cutbacks, why is that?

Perry Gaskill says:

This is a faulty metaphor. The original use of the term software “stack” most likely dates back to the early days of the web when it started to become common to use a “LAMP” stack (Linux, Apache, MySQL, and PERL or PHP) as the basic foundational components for a server. Each of those components was considered at the time to be pretty much best of breed with the additional advantage of having synergy when used together.

Nobody owned the stack. It was open source. That was the point.

News publications can better be defined as a matter of verticals. Grantland, for example, can be considered a topic-based vertical no matter how much it chooses to push the scope of sports, while the Seattle Times is a locale-based vertical concerned, at least in theory, with all things within its circulation area.

One of the bigger problems with locale-based verticals is something William Allen White warned everybody about around a century ago with the adoption of the teletype. You can provide either a mirror of the local community or a window on the world, but if you swing things too far over to the window side, you risk losing what differentiates you as a provider of local news.

Okay. So it’s a faulty metaphor because….?

Yeah, thinking about news publications as “verticals” is easier and fits better for me. Verticals have clear voices and original definitions of what their beat constitutes. Verticals can be small. For me, “full stack” represents a much broader idea: it represents 1) intellectual originality and 2) ownership of much of the hardware, software, labs and human power required to sustain that intellectual/industry originality.

This is exactly what we’re trying to be with De Correspondent. It’s a lot of work though, since we have to build almost every solution ourselves (from the editor to member admin, and from book publishing to speakers agency, how you rewards your writers, etcetera), but also a very thrilling and rewarding thing to do.

And maybe, in the long run, we could license things like our CMS.