Five experienced technology journalists gathered to a standing-room only audience at LinuxCon Tuesday to discuss "Hype vs. Reality: Today's Linux Story from the Media's Perspective," moderated by Jennifer Cloer of the Linux Foundation.
The panel consisted of:
- Joe "Zonker" Brockmeier, Linux.com & Ostatic
- Jason Brooks, eWeek
- Sean Michael Kerner, InternetNews
- Ryan Paul, ArsTechnica
- Steven Vaughan-Nichols, ComputerWorld
Note: What follows is not a perfect transcription, but as close as this attendee's fingers could type for you live.
What are the biggest Linux stories of the last decade?
Sean: The first stories were about SCO, and it's the "story that keeps on going." It was FUD and is still FUD. The bogey monster in the corner.
Steven: Linux is real. It was IBM making Linux the center of their business operations. Now over half of large enterprises are using Linux. We're now the majority, and that started when IBM said Linux is real.
Jason: The birth of Red Hat Enterprise Linux. It was something that had to happen, where distros forked out into the kind of stability that was needed for where Linux has gotten to be. And it set up an example of a business model and created a space where Red Hat Linux had been, which Fedora filled, and in that space, Debian filled part of it, and CentOS has risen. It was a move that had to happen and an important step in the development of Linux.
Ryan: The growing ubiquity of the Linux platform in the mobile device ecosystem. It's practically dominant, whether it's your TiVo or Kindle. And other key components of the open source stack are there.
Zonker: The rapid ascent of Ubuntu and forcing companies to really focus on community. Where Red Hat and certainly Novell weren't focused on community and were concentrating on the enterprise, Ubuntu's rapid ascent forced companies to look at community in developing their products.
Kerner: Fedora was born about the same time.
Zonker: Fedora was a failed attempt until they were forced to do something.
[editor's note: These remarks spurred a lot of comments from the audience, the panel, and the Twitterverse. I couldn't begin to keep up typing out the back and forth.]
In contrast, what's the biggest story today?
Steven: Android, obviously. ChromeOS. Love it or hate it, whether it turns out to be the next Google failure, it's going to be a big deal. And the other is the announcement the Linux Foundation made this morning (Open Compliance Program). By making it so much easier for businesses and developers to get straight with the licensing, it will speed up the adoption of Android and other types of Linux.
Ryan: Mobile technology in general. Portable computing and ubiquitous mobile Internet technology are where computing is going. To have it with you all the time. Part of that is the cloud, and Linux has a strong presence there. Enabling rapid development of mobile apps and lowering the barrier to entry to a customized mobile experience for specialized devices is going to be important. Whether it's Meego or Android, that's definitely the trend.
What about analysts in your coverage? Are the big three's numbers accurate? How do you deal with that in your coverage? Do you consider analyst's numbers?
Ryan: I would say the numbers aren't entirely useless, but they're not illustrative of the bigger picture. On the rare occasion I use analyst numbers in an article, I say take it with a grain of salt. There's not a lot of objective analysis anyway, which is hard to do in a consistent, reproducible way.
Sean: Unless it's a specific analyst story, I don't trust them because I don't know who their clients are, and you can't always know that. As far as the numbers, they're always wrong because there's no way to count the total number of Linux users. There are so many mirrors, how can anybody know? Any time somebody from Canonical says, "We have X users!" we write that, because that's all we have to go on. What would be great is a standarized way to report on downloads and usage. Analysts don't have vendor neutrality.
Jason: It isn't shipped with server hardware. You can't get a handle on the free usage, because who knows what happens to a given ISO, but I'd love to know what happens with SUSE subscriptions or Red Hat subscriptions. That's a piece of information I'd be interested to see.
John "Maddog" Hall interjected from the audience: Bob Young estimated systems by putting down numbers with a high and a low possibility by presenting all the facts he had on the matter, and that's about the best you can do.
Jason: I think as long as the numbers are cited clearly, it's better to have some numbers than nothing.
It's harder to get reporters' attention. Linux is not the biggest story in tech right now. What kinds of stories are you guys telling?
Sean: I'm always going to cover the big vendors. But it's got to be something that's interesting, unique, or differetiated and not just a commodity. In my inbox every day, I get inquiries about apps that run on Linux. That's great, but it's got to be something innovative.
Zonker: On one hand, you have more people covering Linux because it's become so mainstream, so you have some reporters without a real depth of knowledge. I've worked on both sides, PR and reporting. I watched the attrition rate of my press around 2007-2009 because the people who were full time covering these beats were getting laid off. As Linux becomes more mainstream, some of these stories are no longer stories.
Steven: This happens with all technologies. Linux is becoming commodotized and invisible. Linux is running on things that most poeple would never think of as having Linux. Remember when a phone was just a phone, and you didn't know what it was running? The success story of Linux is that it's going to be everywhere, but from a news perspective, that means Linux is nowhere. In the 1920s there were a lot of magazines about radio. Do you remember when there used to be lots of Internet magazines? Now the Internet is ubiquitous, so nobody writes about it. Occasionally somebody writes about something like IPv6 (ready or not, here it comes). Now it's happening with technology reporting, period. PC World is the last of the big, old time, print publications of that type. And there was a time I couldn't have imagined that happening. So it's a natural progression that's annoying to those of us in the business, and we'll still be trying to bring you news, but it's harder to find stories people don't already know.
"Maddog": I think we're looking at this the wrong way. Anything that is everywhere is by definiton a commodity. But the business problems are all unique. And what's different about open source is that you can change it to meet the business problems. So what you need to be writing about is how open source is going to help a specific industry.
Steven: Popular stories, like what Lindsey Lohan had for lunch today are where people go. Reporters have a problem today because we're driven not by what we think is important but by what people will read. That means if we want to make a livng doing any coverage, we have to spend time on popular stories, not important stories. Now, all of us on this panel fight to write the important stories, but it's a challenge as time goes by.
Jason: This isn't about tech publishing or Linux publishing. We are still way in the thick of digesting the changes that came along with the Internet. You can say a story is popular because people are reading it, but which readers? And we don't have a good way of answering that.
Question from the audience: Who do you write for? How do you see your role? What about who's reading those articles vs. who's advertising on your site?
Ryan: At ours [ArsTechnica], we have an extermely diverse audience because we're not a dedicated Linux site, which makes my experience different. I work alongside people who do gaming and Apple and Microsoft coverage, and our readers tend to read all of it. But at the end of the day, it's the tech enthusiasts driving the traffic. As far as the advertisers, the value we offer to them is that the readers take what they read with them into the workplace. So say we write about virtualization at an enthusiast level, but maybe there's something they take to work and decide what to put in their server room.
Jason: We ask people to at least tell us they're enterprise IT. but the truth is we don't know who is coming, so that has an impact.
Zonker: It's not about my role. I write for multiple publications with different audiences, so it depends entirely on who I'm writing for. It's entirely different if it's a review of WordPress for eWeek or for LWN. One might want to know if a CIO can count on it to do x, y, and z. LWN wants to know what changes they've made that will help somebody write a plugin.
Ryan: The challenge is that you'll get one person saying you covered APIs too much, and then the next person wants to know if they can root it and put a compiler on it. So it's hard to create a balance to serve all these technology enthusiasts with different interests. So I try to get the poeple who are non-technical excited about the programming things.
Steven: I write for a range of audiences ranging from the highly technical IEEE to more popular pieces for the Washington Post. Mostly I go to the enterprise audience. One thing that should be said is that whomever I write for, there is and should be a wall between advertising and editorial.
How are you dealing with the idea that companies are becoming their own publishers through social media or a website or a blog? Is it competitive? Is it good, a resource?
Steven: It's a resource, but one thing, not just in Linux or tech, is that people are getting confused between what they're telling you and what a reporter is telling you. They're telling you the company line, which may be exactly what you want to hear, but remember, they have to write with a bias. They want you to use the product. Mozilla, for example, does a really good job of telling you about Firefox. But Thunderbird's latest indexing will kill your machine--are they telling you about that? People are more willing to accept information and not ask the reporter's first question: How are you lying? They might not even be trying to lie, but you need to know what a person's bias is to keep from being tricked. Maybe journalists have a cycnicsm stamp on their heads, but people tend to believe "it must be true, I read it online!" My dog could be online. She's a great dog, but still...
Ryan: The big distros tend to be really straightforward, and you have enthusiastic community members who can eliminate the need for certain types of journalism. It used to be that my job was to dig through bug reports and commit logs to see the things people would overlook. Now it's all out there. Mozilla publishes the commit log for the week. Everybody here probably reads LWN. There are places with really good community journalism that eliminate the need for the traditional top-down approach. It's bringing down the number of places we can provide value, but one is to give a broader analysis.
Sean: There's always something they're not telling you. And that's our job: To investigate. For example, Mozilla 3.6.8, they had a bug they fixed, but they didn't put it in the advisory. Digging through bug reports is still critical. Everyone is still hiding something, whether they know it or not.
Jason: And since they wrote about that release and didn't disclose that fact, it tells you something right there. Companies think they can just toss up press releases or fluffy blog posts. You'll see something up for six months. Zero comments. Nobody is reading it. But there are vendors who give you meat to chew on. I think we're going to see more warring between vendors. You write something that isn't true, somebody will call you on it.
Zonker: That happened last week with the thing over Red Hat contributions to GNOME between Greg DeKoenigsberg and Mark Shuttleworth. With open source, anyone who knows where to look can find the bug report. What's missing is context, and that's what reporters are paid to do. A lot of people dropped into that GNOME conversation in the middle and didn't get any context.
Question from the audience: Is there a concern that with Linux becoming popular that it will be more vulnerable to security threats?
Jason: That's the conventional wisdom.
Sean: At Black Hat last week, I saw not a single Linux vendor on a single panel. It's such a green field for attackers. It's too easy.
Steven: Fundamentally, whether you believe in Linux or Windows or whatever, Windows is buit to be a single-user, non-networked system. Linux is meant to be networked, which makes it more secure from the beginning But yes, we should be concerned. We should have people at Black Hat. It will never be as easy as Windows. Will it be attacked? Yes. More often? You betcha.
Question from the audience: How do you think society can be altered? For example, we've heard pros and cons about making the desktop more comfortable, how the future is IPv6, and getting my fridge on the Internet. So do you have any proposals for how to advance it?
Jason: Get it in more people's hands. Linux is cool in that it just flows into all the available cracks. The important thing is that the foundation has been laid to put the tools in people's hands.
Sean: I think what it will take is money. There's never been a Linux co-marketing approach. If a consortium of Linux vendors, like say, Linux Foundation, put out the money Shuttleworth paid to go into space--$20 million--we could make a big push.
Ryan: Linux is not a consumer brand. ChromeOS, Android, these are consumer brands. Linux is a developer brand.
Sean: Think of Cisco. Network plumbing. Unless you're a network plumber, you don't need to know about it, but they're spreading the word to make it into the mass market.
Steven: It's not really our job to do that as journalists. It's your job out there to decide where Linux goes next and to make it happen.