Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage Interviews Interview with Charles Humble on InfoQ and QCon

Interview with Charles Humble on InfoQ and QCon


1. Can you tell us how you got started in technology and then what led you to join InfoQ to write in the first place?

Alex's full question: Hi. I’m here with Charles Humble at the 10th QCon London. Charles, you’d been with InfoQ for a considerable amount of time as chief editor. Can you tell us how you got started in technology and then what led you to join InfoQ to write in the first place?

Yeah, sure. So the point at which I got interested in computers I would have been about nine or ten, I guess, and that would have been the 8-bit era, so the Commodore 64 period. The Commodore 64 was the computer that I learnt on, and a lot of my kind of programming heroes at that time are people like Jeff Minter and Andrew Braybrook and some of the games programmers. And those guys would share little bits of what they were doing in magazines like Zzap!64 and that kind of thing. So you could go and learn how you put a sprite in the border or how you get more than 16 sprites on the screen or whatever it was.

So that was kind of my first introduction to computers; my father worked for IBM so I had some kind of early exposure at that way as well. And then I went off and did some other things, and then my first professional programming job was in the early ‘90s, it was about ’92, ’93. And at that time, open source wasn’t really a thing. Software was very proprietary and it was a kind of closed world. Academics didn’t really talk to industry and vice versa. It was a very different time.

So my first job was actually in publishing and hilariously where I've ended up I wasn’t actually that good at publishing at that point. But they were a company called home distance learning or HDL and they used to do management courses like management training courses, and they were semi-bespoke. So they would have sort of semi worked out courses and then they would customize them for Unilever or whoever it was. And I sat in the art studio for a while. And in the art studio, they would do the customizing bits which were mostly hand drawn at that point, and then they would lose them all.

So the next time they needed an arrow pointing that way or a a picture of a dinosaur or whatever it was they couldn’t find it. And I feel, “Oh, that’s just a database and some metadata. That’s a problem I know how to solve.” So I solved that problem for them and that was kind of the beginning of a weird change in direction in career back to IT. So I actually read English at university. I did that mainly to annoy my English teacher at school, which is a reason, and had a great time and then came out the university and didn’t really know what to do with myself because it turned out there weren’t a lot of jobs for literary critics. Who knew?

Yeah. Like I said, I did this job in publishing and then I kind of did everything in IT that you can imagine. So I worked for a helpdesk for a while for Barclays. And then I went to Barings Bank. I don’t know if people remember Barings Bank but if you didn’t look it up, it’s a good story. Kind of doing back office ERP type of stuff. I then gradually worked my way up and ended up as an architect and a CTO. So I kind of did the whole, you know, everything, from desktop support to senior job. And then in parallel of all that, I was just reading stuff on the internet when InfoQ was first started. So that would have been 2006.

I sent an email to my predecessor of the Java queue, Scott Delap, and I sent him an email because he hadn’t written about something like GWT, I think it might have been. I said, “Why are you not covering GWT?” And he gave me a very nice explanation why and then said, “Would you have an interest in writing for us?” Then I went, “Hmm, yeah, that sounds like fun.” And that was kind of how I got started, and then I ran the Java queue for a while and I recruited you amongst other things.

And then the startup that I was doing where I was CTO got bought by a large consulting firm. I worked for them for a while and that really wasn’t me, I decided, and Floyd [Marinescu] and I had been in conversation anyway and the chief editorial role was there. I thought that’s quite interesting. And I think that for me, it coincided with the point in my career where I’d kind of – I think I reached the level as a programmer where I wasn’t really getting any better and I wanted to go and do something else.

I’d programmed professionally for 20 years. It's quite a long time, and that for me personally, that was time to go and do something a little different. And the wonderful thing about InfoQ is I get to kind of keep up with what’s going on and meet amazing people and have extraordinary conversations. And I’m still sufficiently technical that I can kind of keep up with things. But at the same time, I don’t have to sit and do the job of actually physically programming anymore, which to be honest, I wasn’t enjoying as much as I have been.

Yeah. It’s been an interesting ride and that’s kind of how I ended up. So yeah, I’ve been doing, I’ve been chief editor for I think two – this is my third QCon – so it’s about two years as chief editor. And that’s been fascinating. I came in at a point when InfoQ was in need of, you know, some love and attention and I’ve been able to get a little bit of that and that’s starting to pay off. It’s a fascinating and complicated thing to run. So it’s a good challenge.


2. But how have those queues evolved over the years to focus on what changes in technology or development practices have been coming out?

Alex's full question: And the concept of queues is being able to separate out into different technologies: back in the day when you and I joined, there was Java, there was maybe .NET at the time, but there was sort of what you would call the traditional enterprise languages, as well as architecture as well which has always been a strong component. But how have those queues evolved over the years to focus on what changes in technology or development practices have been coming out?

So that’s kind of fascinating. The mainstay of the site is still the kind of big enterprise technologies. In the case of Java, which is obviously my area, so it’s the one I know the best, that evolved from being about Java the language to being about the JVM and the runtime and some of the alternative languages like Scala and JRuby and so on.

And I think interestingly now, if you look at it, we’re kind of going back and focusing in on Java again more because mainly on the back of Java 8 and getting lambda expressions in Java finally, there’s a movement back towards that. I think there’s more interest in Java than there’s been at some time. The other thing that we’ve started to do is just look at other –– you know, there are a whole variety of other languages that are interesting and bringing new and fun things. So Go is a really good example of something that we don’t have as a focus topic any more but we do create a lot of Go coverage, and some of the other emerging sort of actor languages like Pony. We’ve got someone [Sylvan Clebsch] talking at QCon who’s the designer of the Pony language, for example. So that’s the thing on the language side. On the architecture side, again, it’s kind of been a journey towards micro-services and those kind of things, as you would expect, CQRS [Command Query Responsibility Segregation] now.

So the great skill and the interesting thing about the job that I do is trying to kind of spot the trend early enough, [so] that we can be in a position to cover it at the point of people that are interested. So something like perhaps, data science, which is relatively new for us and some of the research that's going on in AI now, and machine learning techniques and those kind of things which are still relatively early. There are still a lot of companies that are talking about them but they’re not wildly adopted yet outside of the Bay area companies like the Googles and Netflix and what have you. But they’re clearly going to become one of those things that if you’re a programmer, you’ll just going to need to know at least some of that stuff.

So that’s kind of what we try and do, is spot those things early and then build up the expertise to be able to write about them in a knowledgeable way at the point of which more people become interested.


3. And of course InfoQ has multiple language translations as well. Could you tell us a little bit about where InfoQ is published and how the international group hangs together?

Yes. So we publish in five languages. The English site is the part that I run and then we publish in Chinese, Japanese, Portuguese, and French. So essentially, we have teams that are local teams in each of those areas, and they can take content from the English site and translate it and that’s a percentage of what they do. But obviously, they can also go and write local stories. The China team, for example; China has an incredibly active startup scene now that has some parallels to the Bay area and what have you.

So they will do those kind of local stories and sometimes we’ll bring those back and publish them on the English site as well. But then they can also take stuff from us and translate them. So on the English side, we have about 50 people that write on a regular basis, and that’s kind of my responsibility to manage that and check that. And then as I said, the other regions take them and they will do some translations and will do some local stories. And occasionally, we’ll pick one up and bring it back.


4. And these writers that you talk about, are they all full-time staff on InfoQ?

No. We don’t employ any full-time writers at all. Everyone who writes for us is an engineer or is a manager of a software team. So one of the things that makes InfoQ InfoQ – its kind of the secret source, if you like – the magic thing that makes InfoQ "InfoQ" is everything on it is written by an engineer. So what that means is that everything that’s been written has been written by someone that genuinely knows what they’re talking about, or at least has used the thing that they’re talking about. And what tends to happen a lot with professional journalism, professional IT journalism, is you get people that are great writers, but they’re not necessarily that conversant with the stuff that they’re actually talking about. They may have been at some point in time but they’re perhaps not practitioners is the term they use anymore.

We use the same model on QCon, in fact. Again, we’ll go and find engineers rather than professional conference speakers to talk about what they’re doing. And I think that gives the site quite a unique feel and just as I say, it sort of represents an amazing body of real world experience that we can then share.


5. You’ve also been putting together some mini books recently and collected papers for things like the iOS 9 release. Where did those come about and what kind of success have you seen with that so far?

That’s been hugely successful. There's a publishing term which is "repurposing content." Basically, the idea is you take content you’ve already got and publish it in some other form. With InfoQ, one of the challenges that we have at the moment is because we publish quite a volume. If you’re not coming to the site every day, it’s really easy to miss useful and important stuff.

So one of the ideas with the eMags and the mini books is to take effectively a collection of related articles and put them together in some kind of downloadable format like a PDF or an ePub or .mobi if you're on kindle or what have you. And those do incredibly well for us. There are a group of site readers who maybe come to the site once a week or once a month or something rather than every day. And for those people having something that they can go to which is the – "everything you need to know about iOS" or "everything you need to know about Java 9" or "what's this Cynefin thing that I keep hearing about" or whatever it is, those are just really fantastic resources.

We try and keep them very concise. So the general idea is that you can sit down and read them in a morning and kind of get a very quick brain dump of “Here are all the important things that are hot, that you need to know about this technology or this new version of iOS or whatever it happens to be.” And then you might need to go off and do deeper dives onto articles on the site or elsewhere and that’s fine. But it’s like a kind of crash course.

We have the advantage that we’re not selling physical books. So if you're selling physical books, you need wide spines on the shelves because when people are browsing that's just how it works. We don’t care about page length and width. So we can do this very, very concise thing which is kind of a quality of basically being an on-line publisher rather than physical publisher.

So yeah, those are incredibly successful 10,000, 15,000, 20,000 downloads is not unusual, which is great.


6. Sort of like a PDF version of a podcast then?

Yes. I mean podcast is something that we haven’t done to date. We do the video interviews (like this one). But yes, it’s that kind of thing. Again, it’s getting experts and doing some sort of condensed, concentrated version of – an intense version or a very concentrated version of “This is the stuff that you need to know.” Podcasting is quite similar. As I say, podcasting isn’t something we’ve done – I would quite like to. It may be something that we do at some point in the future. But yeah, there are certainly parallels.


7. How many readers come back to InfoQ then?

Worldwide right now, it’s about 1.5 million. The English site is 750,000, typically. So that’s a pretty good number. It’s hard to know. I don’t really have a feel for what the top-end number might look like. We kind of plateaued at about a million for quite a long time and then we’ve suddenly seen a huge growth in the last year or so as I said about 1.5m. Some of that, quite a lot of that, is coming from China at the moment. There’s a lot of momentum in China just now.

The reason it’s difficult to know what the maximum number is it’s because it’s very hard to get reliable stats on how many programmers there actually are, how many developers are there in the world? No one really knows, right? Most of the stats you can find appear to be made up by people with nothing better to do. They don’t have a lot of real basis; 10 million number I’ve heard. I don’t know how real or otherwise it is.

But again, we skew our content towards senior people. And one of the consequences of that is that you’re not going to get all ten million. You’ll get some slice of that as yet not entirely clear. But I still think there’s room to grow and I think now there’s some really interesting things that we can potentially do in terms of trying to make it easier for people when they come to the site, to find the stuff that they need quickly.

So one of our underlying assumptions is that most of our readers are very busy. They don’t necessarily have time to kind of absorb, you know, to spend ages clicking around the site trying to find the stuff. So the ideal thing would be you arrive at the site and the things you need to know are in front of you right there. And that gets you into some interesting territory around personalisation and machine learning and those kind of things which we’re doing experiments on at the moment and you’ll see some of that on the site in time. And yeah, that’s a big drive for us.

Alex: And something else I have noticed over the past five or so years is that there’s been an increase in the number of presentations that get recorded, not just interviews like this one but from conferences all across the world as well.

Yeah. We do – roughly half the money that we spend on content is spent on presentations currently. They are a relatively expensive format. We film at about 30 conferences a year, 30-35 (it varies a little bit). So yes, obviously, we do our QCons as you would expect but we also do huge numbers, number of others as well; we did SpringOne for the last couple of years, for example, and smaller events as well.

So for example, we’ve recently started – we don’t actually film them but we take the video from the [Docklands] London Java user group and publish that on the site, and those kind of things. Being able to do that is wonderful. And having a platform that you can do that is a really fantastic thing. There are percentage of – I call them readers but it’s slightly unfortunate term – there are percentage of readers who are really watchers, who would watch video as their primary thing. And there’s another percentage of readers who are readers in the conventional sense and they'll tend to gravitate towards articles and news posts.

And obviously, there are people in the middle who do both. But there is kind of a split there, so we kind of try and cater for both groups. And yeah, we are constantly on the lookout for the new events and events that strike us as being interesting that fit to the mission and the remit that we have for ourselves. And then yeah, we try and partner with them. And from the event point of view, we do a lot of this, so we kind of do the logistics for them. And then we’ve got this incredible platform with this huge number of readers who can come and watch the presentation. It kind of everyone wins. It’s a nice situation to be in.


8. How did the QCon conference brand come around?

Alex's full question: And of course you have interviews like this one which you’re watching right now with some of the greatest people in computer science and presenters as well. Like you say, we pick up the Docklands.LJC feed. I think this week, we’re recording both at EclipseCon and at QCon London as well. But let’s talk about QCon London. It’s now in its tenth year. It’s bigger than ever. It’s been at the QE2 I think pretty much since the beginning, if not the beginning. How did the QCon conference brand come around?

So InfoQ and QCon started at about the same time. QCon London was originally a partnership with Trifork who run the GOTO events in Europe now. And I think the first one – if I remember rightly – the first one was actually kind of done at a JavaOne event ten years ago. And then QCon London was the sort of first formal one of with the QCon branding. I go to a lot of conferences as you would expect; it's kind of part of my job.

Genuinely, honestly, I really, really enjoy QCon. There’s something very special about it. And I think what’s special about it is for one thing it's cross-discipline. So I can be sitting next to a DotNet person or a JavaScript person or a Java person. But equally, I can be sitting next to someone who’s an Agile coach or done really interesting neuroscience research or whatever it happens to be. So we kind of try and do some quite different sorts of things from the kind of things you get at a typical tech conference. You know, we have this formula that seems to work really well.

This year, I think is the biggest QCon London we’ve ever done. So we’ve got about 1,300 attendees at the QE2 for three days, which is amazing. I have the extraordinary privilege both mornings actually of doing little tiny two five-minute intro things on the keynote stage and you stand on the keynote stage and look out. There are 1,300 people looking back at you. It’s actually an amazing experience.

It’s a hugely successful thing. I think there’s something really incredible that’s happened in the industry. So going back to what I was saying at the beginning about starting work in the ‘90s and people didn’t talk to each other. And now, you can have an event like this one and you can have Facebook and LinkedIn, and we had Apple here for the first time in London. These huge named companies coming and talking, sending their engineers, their actual people working at the coal face, to talk about what they’re doing and sharing knowledge. You’ve got the BBC and Netflix who kind of complete at some level, collaborating on stuff.

I think it’s just incredibly exciting and that’s kind of what we try and do with QCon. For me, being involved in the committee and helping to choose some of the speakers and all of that kind of thing is a very rewarding part of the job. I mean obviously, the other thing as the editor of a website is to actually be able to come and physically meet some of your readers, and people who aren't your readers and talking to them about what they’re doing and what they’re interested in. It’s just an amazing opportunity.

So I try and do the three English QCons that we run. So we run London, San Francisco, and New York, and I try to do all three of those. You know, it’s like a huge focus group for me. I can just go around and have these amazing conversations with people about all sorts of stuff. And it’s really exciting. And hopefully, that’s the same experience that attendees get as well is stand around in conversation with someone who’s here because the chances are they’re doing something interesting and smart and engaging, and you’ll have an amazing conversation.

Again, we put a lot of time into thinking about the user experience of the conference. So we have long breaks between sessions, for example, we have areas where people can congregate and there's 25 minutes between sessions. And the reason we do things like that is to encourage people to grab a coffee and just get chatting because so much interesting stuff happens – obviously, the sessions are important – but so much interesting stuff happens in the hallways. So yeah, it’s an amazing thing, I think. It's always a joy.

Alex: And of course the conferences are organized into tracks where you have one room and one track host that’s looking after a fairly similar set of topics, but then you replicate that five or six times across each of the days. So you get people who are coming in being able to choose kind of like a smorgasbord of events or just stay on particular events that they’re interested in.

Yeah. So it’s kind of like having multiple vertical events going on in the room. That’s kind of how we think about it. So I was hosting the Java track yesterday. So I sat there and obviously put a Java track together. And you can come and spend the whole day in that track and there is a kind of like a coherence in the plan that runs with the whole thing. But equally, you can be looking at the schedule and going, “Oh, there’s someone talking about Erlang or something and that’s not a thing I know anything about. So I’ll go to the Erlang session because that will be interesting.” And that’s the brilliant thing about having that kind of cross, as I say having a cross-discipline event is that you can do that sort of thing.


9. Now, you talked about the English speaking QCons there with New York and San Francisco. How long have those been running?

San Francisco was the second one, and I think that started about a couple of years after the London. So it’s been about eight years now. And then New York is the newest. I think it’s about five years old.


10. And what about in non-English communities?

Okay. We do I think seven worldwide. We did a couple in China. There are a couple in Brazil, one in Rio. There’s at least one in Japan. So yeah, quite a spread. I think there are seven or eight worldwide now and we’re always looking for – again, we’re constantly looking at the possibility of doing other ones. So hopefully, that will be more over time.

Alex: One of the things that I enjoy about coming back to QCon London is that because we have this track-oriented structure and because we try and keep on top of what’s new, they change over time, and it doesn’t feel like you’re coming back to the same conference. You find that you’ve got a lot of familiar things like the Java track, for example, but then there are also other ones on human factors or new languages or optimization. So it feels like there’s a lot of changes as these things go on. And I guess that’s one of the things that keep the events feeling lively.

Yeah. We make a huge effort to do that. I’m not sure – you may remember the chasm graph, crossing the chasm that we did a few years ago which we kind of use as a model. And we aimed at the first little bit of that graph, which is kind of the innovators and the early adaptors stage. And we make a huge effort as the committee to find new speakers and people to talk about new things. So yeah, I really hope that’s the experience that people have is that what you want to get from an event like this. And I think hopefully, this is what most people get from it, is that you come our way and you feel inspired and interested because you’ve learned something new or been forced to think about something new.

And there’s a very interesting thing that I’ve noticed as well, which is you get these kind of emergent trends that weren’t really planned but just happened. So for example, I’ve lost track a number of times over the last two days I've heard someone citing an academic paper. And that’s something that wasn’t happening a couple of years ago. Suddenly, as an industry we’ve gone "ooh".


11. You know what they wrote about in the ‘70s? That was actually quite important.

Turns out they knew what they were talking about. Who knew? Yeah. And I guess because suddenly we're all building distributed systems and distributed systems are frankly still quite hard. So something about the Lamport paper, or whatever it is. If you haven’t read these things, go read these things. So as I said you got this kind of – the event has these emergent properties which are quite interesting which we sometimes plan for but often we don’t plan for, they just happen. And that’s just a very interesting quality of it. So a couple of years ago, it was Conway's law, for example, is the one that no one planned for and everyone – it’s almost like we have some rule where you have to have your Conway's law slide in before you’re allowed to speak. And as I say this year, it seems to me all about papers, and that’s great.


12. What’s your favorite memory of QCon London?

So about four or five years ago, I did the functional track here. One of the things I did the functional check was I invited Phil Wadler to speak. And Phil, at that stage, he’s told me, I was genuinely surprised, but he said he’d never done a non-academic conference before. So doing like an industry type conferences was novel for him. And he was – I was chatting to him in the room before we started; like twenty minutes before we started and there was no one in the room and he’s, “How many people do you think will get?”

I was like, “You know, there’s like 100 capacity room roughly, and I thought we might fill it.” And he was thinking we might get like four or five people. We absolutely packed the room and he gave the most brilliant, brilliant presentation which was very – you can find it on InfoQ, actually, it’s on "Faith, Evolution and Programming Languages;" go look it up, it’s awesome. But yeah, so he gave this very interesting, very provocative talk about, essentially about Lambda. That was just a wonderful wide-ranging talk. One of the points that he was making was that he was Jewish and the whole Lambda calculus thing was invented by a German Nazi, and that was therefore quite interesting.

So it was a very, very kind of just a really fascinating talk. Yeah, he absolutely killed it. That was a really lovely moment for me personally because he took a risk and we took a risk and we weren’t sure that it will come off and it totally came off. We got so many good comments from that afterwards. So that was very, very satisfying. But there have been a lot over the years. There are always these moments that you're just sitting and watching the keynote and someone is just absolutely killing it for one reason or another. That’s a real, real wonderful experience.

Alex: And also look forward in the future as well. Charles Humble, thank you very much.

Thank you Alex.

Apr 18, 2016