How did you become interested in open source and community building?

I was contributing to an open source web publishing system for digital archives called Omeka. The primary goal of Omeka is to make publishing digital archives of historical photographs and stories as easy as publishing a blog. We patterned our community strategy around Mozilla and WordPress, trying to create a ladder of contributions where people of varying skill levels could get involved, and I was helping coordinate developer community growth. Shortly after launching our first public beta, we realized that the community of interested users was more diverse than we imagined, not only from museums and archives but also libraries.

For me, community building began mostly as a way of understanding and negotiating the differences and needs of these institutions. You need direct, personal connections with your users in order to understand their needs; in the process, you start to draw connections between the work of others and play a role of matchmaker.

My interest in community building led me to help co-found THATCamp, The Humanities and Technology Camp, an un-conference. THATCamp is a BarCamp-style event, bringing together technologists and humanists to create sessions related to digital humanities. Sessions vary from event to event, but my favorites have always been ones that focus on building. And since 2008, there have been over 100 THATCamp events around the world.

You’re involved in open web projects through the Mozilla Foundation, right?

I’ve been working as an Integration Engineer Contractor with the Open Badges team at Mozilla, mostly helping third-party developers integrate with APIs to create and display badges. Open Badges is a standard to recognize learning online through the open sharing of digital badges, It’s an exciting approach to informal learning and using badges as a way to capture achievements that are otherwise not visible on a resume.

One of my contributions to the project has been creating several WordPress plugins to make it easier to issue and display badges; it’s important that a variety of platforms adopt the standard to give the community a variety of ways to hook into our infrastructure.

You’re also interested in hacking hardware, such as drones. What has this taught you about coding?

This semester I helped organize a group of fellow graduate students at UC Berkeley to form what we’ve called “Drone Lab”, an informal group that has met weekly to hack, discuss, and investigate creative and problem-solving uses of consumer-grade quadcopters. These are hobbyist toys that you can buy at your local shopping mall, but the ability to control them using software that you script unleashes the potential to tap into their cameras and sensors from heights and hard-to-reach places that are new and exciting. What we ended up focusing our hacking on were new ways to control the quadcopters, including voice and tracking head movements.

What I found fascinating the last several months was introducing several of my classmates to Node.JS through programming these drones. Learning to program can often be a frustrating and unrewarding experience, but with just a proper development environment and a few lines of Javascript, you can fly a copter. Programming shouldn’t be limited to terminal windows, and the feedback of seeing the drone fly can be very rewarding. This also fosters creativity and unexpected things – sometimes you’ll see the drone do something in flight that seems odd, which prompts new questions about your code and experimentation that can be less common in programming.

So were you part of last year’s TacoCopter stunt?

TacoCopter is a project that I’m not involved with; I believe it’s meant to be more of a joke than a real thing. Still, there’s something intriguing and futuristic about a flying robot delivering Mexican food that gets people’s attention. We joke a lot about delivering tacos via drones.

What do you see on the horizon for programming and the open source movement?

In the age of GitHub where it’s easy for anyone to share code online and gain a following, the proliferation of projects both big and small can come at the expense of a clear way to integrate various codebases together. In my experience, it’s often the “glue code” and examples that are most valuable to users who want to use your software; the last 10 percent, so to speak. To be effective in open source community building, understanding those needs of integration is crucial and something I’ll be spending a lot of time working on.

In general, I’m excited to see more companies using and releasing open source software, not for the goal of selling it but in an effort to develop better services and give back to communities that they benefit from. The precise model for how this software will be supported, grown, managed, and sustained is still to be defined; these are often projects without a software foundation. I hope to see more coordination and partnerships among companies regarding open source contributions.

Finally, what’s next after you finish your Master’s?

I’ll be joining Twitter as an Open Source Advocate in June. I’ll be responsible for building relationships with communities to drive adoption of our open source projects and APIs. Twitter has over 100 open source projects, and as an organization has made a big investment in using and releasing open source software.

time to make it official with a tweet: in June I am joining @twitteross as an Open Source Advocate. I’m pumped!

— Dave Lester (@davelester) May 8, 2013

Images via OpenBadges.org, UC Berkeley School of Information

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Survey: How apps can solve photo management
• Sector RoadMap: Social customer service in 2013
• Sector RoadMap: Content personalization in 2013

The impacts of the purely JavaScript-based system are multiple: for end users, the ability to run native PC apps on any device with an internet connection and to purchase and protect content without digital-rights management (DRM); for content creators, cheaper, faster rendering and the ability to distribute anywhere viewers can type in a URL; and for open web or cloud-computing advocates, a push away from proprietary or legacy plug-ins and an embrace of HTML5. With the presence of William Morris Endeavor CEO Ari Emanuel at the launch on Friday, the creators of the ORBX.js technology were also seeking to emphasize its piracy-fighting powers for the movie and TV industries: with video streams or apps watermarked in the cloud, DRM in the browser becomes unnecessary.

OTOY and Mozilla came together recently with the realization of a shared goal: trying to turn the web into the platform for all apps. Mozilla’s effort to implement H.264 in Firefox inspired OTOY to rewrite their own codec to run in JavaScript, said OTOY founder and CEO Jules Urbach at the launch event in San Francisco on Friday, and the partnership has now culminated in an optimized rendering experience that is approaching native app speeds in Firefox. Among the capabilities demonstrated at the launch were a virtualized Windows desktop running in Safari, lag-free gaming in a browser and streaming that can be adaptively encoded based on a user’s bandwidth.

“Web is the medium,” said Autodesk CTO Jeff Kowalski, who was very upbeat about the possibilities of the new tech for increasing work collaboration and creativity, and reducing delays through real-time rendering. Besides investing in OTOY, Autodesk’s interest is in providing 3D apps to their customers using cloud resources. The implications for agility — both for individuals and for enterprises — are freeing: a low-power home device can drive the centralized, high-power cloud machine, eliminating the need for a high-end workstation or provisioning of hardware assets to employees or contractors. Kowalski’s suggestion, in fact, was that such a move will allows users to downgrade their hardware, because it no longer has to match the needs of the software.

So what is needed for ORBX.js to work? Any HTML5 browser (Chrome, Safari, Firefox, IE10 or Opera) will do, but it needs to have WebGL technology to take advantage of the codec’s full decoding speed. Mozilla CTO Brendan Eich predicted that Apple will eventually come around to more fully accept WebGL. When asked if Apple, Google, or big streaming providers can do anything to stop the use of ORBX, Urbach said nothing short of getting rid of the browser would stop the tech from being used.

The central issues with streaming all of your computing are bandwidth and money. Video seemed to stream well on an iPhone over 4G, and with the adaptive streaming and superior compression of ORBX, Urbach projects a 25 percent bandwidth savings for, say, Netflix streaming. For that to happen, Netflix, Amazon and other providers have to adopt ORBX, something that the Mozilla-OTOY partnership is actively working on. They are hoping that their solution will be the one to put the format wars to rest, and allow consumers to collect the highest-definition content possible in a way that is format-agnostic. With respect to pricing, a ballpark figure suggested at the launch was $300 per year for OTOY’s cloud-rendering engine to take over one person’s computing needs. Pricing is still up in the air, but Urbach expects an AMI to launch later this year with the second generation of ORBX that will also include HDR encoding capabilities.

The videos below show streaming video and gaming through a browser using ORBX (via Mozilla).

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• NewNet Q4: Platform mania and social commerce shakeout
• NewNet Q4: Platform mania and social commerce shakeout
• Connected Consumer Market Overview, Q2 2010

Imagine what Famo.us can do to an Amazon.com experience or even Walmart.com? Hermes could create a new, unique shopping experience, and Warby Parker can use data and design to create a personalized virtual store. Yup, I know I am getting ahead of myself … but hell, it is better than getting excited about yet-another calender app.

Now Famo.us is all set to show off its HTML5 development platform, which at least in demos, removes any advantages of the mobile-native operating environments. The native operating systems such as Android and iOS have enjoyed an edge because they provide a smooth and satisfying experience on their platforms. Famo.us wants to do that with its platform, which is entirely HTML5 based.

Newcomb will outline many of the details of the Famo.us platform at the HTML5 Developer conference in San Francisco today. As part of the announcement, the company will announce plans to make its platform free for individual developers and give them the option to create and host many Famo.us apps on its platform. Believers in Famo.us are going to be the same people who believe and want to build HTML5 apps.

The company in turn wants to make money by charging hardware makers — the companies that make phones and televisions, for example. In addition, Famo.us will also charge enterprises that want to use its platform for services.

Famo.us had impressed me with its technology, and co-founder and CEO Newcomb laid out a compelling vision for his platform. It resonated with me mostly because the company had come up with a way to tackle what I think is a going to be a growing issue in the future: building apps for a data-rich environment. That requires a new approach that looks beyond documents and instead thinks in terms of data streams and apps. It needs web standards, as it is becoming difficult to create, deploy and maintain apps for many of these emerging (hardware) platforms. Famo.us has come up with the right approach.

According to a Forbes.com report, the company has developed a new physics engine that when combined with Famo.us’ previously announced surface rendering technology will allow developers to create apps and interfaces for data-rich environments using javascript and HTML5. Just as gaming engines enable game developers to create multifaceted worlds, Famo.us is betting that with its platform, developers can create similar data-rich applications. “Famo.us will let data itself be visually and tactilely expressive,” writes Anthony Kosner.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

Twitter down for you right now? Fear not, we open sourced our JavaScript component framework so you can build your own. twitter.github.com/flight/—
Ben Ward (@BenWard) January 31, 2013

The Github post describes Flight as a “lightweight, component-based JavaScript framework that maps behavior to DOM nodes.”

It differs from other frameworks because it doesn’t dictate the approach developers must take to providing data to the application.

“It’s agnostic to how requests are routed, which templating language you use or even if you render your HTML on the client or the server. While some web frameworks encourage developers to arrange their code around a prescribed model layer, Flight is organized around the existing DOM model with functionality mapped directly to DOM nodes.”

DOM refers to the Document Object Model – which is the standard representation of a web page and its elements — links, images etc. —  in the browser available to JavaScript.

Flight is the latest of several projects Twitter has put up on the shared Github code repository and versioning system.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• A near-term outlook for big data
• There is more to Node.js than buzz
• Dissecting the data: 5 issues for our digital future

Of course, as a commenter on a Hacker News item, pointed out, numbers can be sliced and diced many ways. If you look at Indeed.com’s chart showing “absolute” number of job postings, it’s clear that demand for Java skills reigns supreme still.

“The fastest growing languages also happen to be the languages with the least jobs,” wrote commenter Zandana. “I imagine Scala’s growth will tail off a long time before it can compete in [the] sheer number of jobs with JavaScript or Java.”

There’s little doubt that the use of Scala, the brainchild of Martin Odersky, is on the rise. Twitter and LinkedIn are big Scala users. One advantage is it enables Java programmers — and there are millions of them — to keep using familiar libraries and develop web-scale applications.

But then again, these charts only show what parameters are put in. If you add Objective-C to the mix, as another Hacker News commenter pointed out, you get still another view of what’s going on.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Infrastructure Q1: Cloud and big data woo enterprises
• Infrastructure Overview, Q2 2010
• New challenges for the IT organization

The new Meteor 0.5.0 version — released Thursday– adds server-side authentication APIs and user account controls which should make it easier for developers to write more secure applications.

Meteor’s aim over all is to help developers write truly webscale enterprise applications faster and easier than before. Developers especially like Meteor’s ability to “push” hot code updates to users without interrupting the users’ work.

The beauty of Meteor’s framework overall is that it lets developers use the same APIs on the client and server side of the application divide, speeding actual development.

According to Meteor’s blog post, the new Meteor.allow API:

“controls which data a Meteor client is allowed to change in the database, and hooks that give the Meteor server control over what data it sends to each client. These core APIs operate at the wire protocol layer, so they establish a strong foundation for security.”

Also new is Meteor Accounts which is built atop the core Meteor authentication to provide higher-level APIs to manage user accounts.

Also new is support for the Secure Remote Password protocol which lets users securely log in to a server without having to send that server their unencrypted password.

In the past six months, San Francisco-based Meteor has built an impressive developer base which helped it raise $11.2 million in Series A funding in July. But as with any new tool set, there were gaps to be filled, and security was chief among developer concerns. A question posted to Stackoverflow summed them up.

“What is the time-frame for adding authentication and data validation to Meteor? This validation/authentication is vital for Meteor to be taken seriously for anything other than toy projects. Full write permissions for every client is obviously unacceptable for most (if not all) applications.”

It sounds like Meteor,  co-founded by Matt DeBergalis, founder of the ActBlue fundraising platform, is listening to their concerns.

Feature photo courtesy of  Flickr user Dominic’s pics

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Infrastructure Q1: Cloud and big data woo enterprises
• A near-term outlook for big data
• There is more to Node.js than buzz

The reason?  It’s time for the scrappy language to grow up to better support the creation of enterprise applications, according to a blog post by S. Somasegar, corporate VP of Microsoft’s developer division.

Somasegar wrote:

TypeScript is a superset of JavaScript that combines type checking and static analysis, explicit interfaces, and best practices into a single language and compiler. By building on JavaScript, TypeScript keeps you close to the runtime you’re targeting while adding only the syntactic sugar necessary to support large applications and large teams. Importantly, TypeScript enables great tooling experiences for JavaScript development, like those we’ve built for .NET and C++ and continue to innovate on with projects like “Roslyn”.  This is true whether you’re writing client-side JavaScript to run on Windows, Internet Explorer, and other browsers and operating systems, or whether you’re writing server-side JavaScript to run on Windows Azure and other servers and clouds.

TypeScript adds static typing to the mix. That means the compiler will flag an issue up front if the developer assigns an incorrect data type to a variable instead of waiting till the program runs, as dynamically typed languages — including JavaScript — do. That delay can lead to problems when the program does run.

An early preview of TypeScript is now downloadable from this site

For more on TypeScript check out the video by Microsoft Fellow (and programming language superstar) Anders Hejlberg.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Infrastructure Q1: Cloud and big data woo enterprises
• A near-term outlook for big data
• There is more to Node.js than buzz

Collison sparked a discussion about Go’s prospects when he tweeted that he felt Go will become the dominant language for systems work in infrastructure as a service, orchestration and platform as as service.

Prediction: Go will become the dominant language for systems work in IaaS, Orchestration, and PaaS in 24 months. #golang—
Derek Collison (@derekcollison) September 11, 2012

Reached by email, Collison said Go or “Golang” appeals to people working above the kernel or the driver layer which is typically written in C or C++. Go’s support of static typing is key and makes it suitable for  building high-performance systems.

“Static typing essentially means explicitly declaring everything prior to a compile,” said RedMonk analyst Stephen O’Grady via email. “Dynamically-typed systems are much more loose, and therefore generally faster to code in. The advantages of static typing tend to be in high performance systems, because there is no decision to be made about the type at run time, or systems of substantial complexity, because looser [dynamic] typing can lead to difficulty in debugging errors.”

I have now completed two projects in Go. I predict that it's going to be the dominant language for server work. #golang—
Tobias Lütke (@tobi) September 12, 2012

Spurred by the discussion, O’Grady did his own analysis using Github and StackOverflow data. The resulting RedMonk language rankings found JavaScript, Java, PHP, Python and Ruby still ensconced as the top five languages overall although Go has made some progress, moving from number 32 in 2011 to number 30 this week. 

O’Grady wrote that this progress may sound modest

” … but means that in that time [Go] has improved more in popularity than Scala or Haskell and as much as Java, at least from a rankings standpoint (obviously growth becomes more difficult the more popular the language becomes). Second, there’s its age. At a bit less than three years of age, Go’s position as a solidly second tier language is enviable, given the fact that there are much older languages like Smalltalk that have yet to break that barrier.”

Apcera is using Go, Collison said, as are Heroku, CloudFoundry, Google where the language got its start,  and other companies. In his view, C will always have its place but Ruby and Python code bases will go to Go.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

But journalism students show what she sees as a lack of desire to learn about JavaScript, HTML, CSS and other tools to help tell a story on a web page, rather than the front page, Mulligan wrote in the Nieman Journalism Lab blog. It’s probably never been easier for students of any age to actually learn to code — there’s free or near free online coursework from Codecademy, the MIT/Harvard EdX program or Coursera. But Mulligan’s recommendation is that J-schools need to integrate these coding courses — or at least teach students about how web pages deal with or render their stories — into the base journalism curriculum.

Mulligan wrote:

We need to innovate our curricula, really looking at what we are teaching our students. Learning, or mastering, specific software is not properly preparing our future journalists for successful, life-long careers. No one can learn digital storytelling in a semester. Mastering Dreamweaver and Flash isn’t very future-friendly, and having a single mid-level “Online Journalism” course offered as an elective does more harm than good. We should be teaching code in all of our journalism courses — each semester, each year, until graduation.

There is growing belief in many quarters that software programming is no longer just for programmers – New York City Mayor Michael Bloomberg has promised to learn how to code; 6-year olds in Estonia may soon do the same. And with more news flowing through the web rather than print, what Mulligan proposes here is not surprising at all.

It’s our job as educators to remove fear of learning, a fear notoriously prevalent in journalists. HTML is not magic. Writing code is not wizardry; it’s just hard work. Learning to program will not save journalism and probably won’t change the way we write our stories. It is, however, a heck of a lot more fun being a journalist on the web once “how computers read and understand our content” is understood.

What is somewhat surprising, in my view, is the reluctance she sees among young would-be journalists to learn these skills. From what I’ve seen over the past few years, many young reporters are impressively proficient in these skills. It’s the geezers (ahem) who have a hard time with coding. But here’s the thing: Even geezers can learn. And if they want to stay employed, they will do so.

Feature photo courtesy of Shutterstock user argus

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Content monetization: News licensing and syndication still need marketplaces and infrastructure
• Infrastructure Q1: Cloud and big data woo enterprises
• A near-term outlook for big data

Meteor Development, the hot startup that aims to remake enterprise software development, picked up some pretty impressive coin in a $11.2 million Series A funding round led by Andreessen Horowitz, with contributions from Matrix Partners.

The company’s goal is to help developers write enterprise applications for the webscale era. Meteor’s real-time JavaScript framework – which has generated considerable buzz — lets developers use the JavaScript skill set they already have to build these applications on their local PCs without having to learn and knit together non-JavaScript code (Python, PERL, etc.)  that typically runs on the server.

“We give you a consistent programming environment for both the client and server code. You need to know JavaScript and you will learn some APIs but you no longer have to think two different ways,” he said

GigaOM first reported on this funding round before it closed, in late May.

Meteor will use the money to build the open source community around its offerings. “That means writing software, hiring engineers to help us do that faster and working with developers who are using it,” said company co-founder Matt DeBergalis in an interview. The company now has 7 employees. The long-term revenue opportunity is for Meteor to build commercial add-ons for the existing products, DeBergalis said.

Along with the money, Meteor gets a ton of enterprise software expertise from its new backers. Rod Johnson, founder of SpringSource, now part of VMware, is joining Meteor’s board.  Both Matrix General Partner David Skok , who helped build the JBoss sales strategy and brought JBoss to Red Hat and Andreessen Horowitz General Partner Peter Levine, former CEO of XenSource, will advise the Meteor board.

What Meteor hopes to give enterprise developers is the ability to build the fast, interactive applications they’ve come to appreciate from their experience with Google and Facebook although precious few companies have the resources of those tech giants.

To be sure, Meteor is not alone in this quest. Parse does some of the same things but for mobile app developers. Yahoo’s open-source Mojito framework attacks some of the same problems and got good early reviews, but does not seem to have gotten much traction.

But given this funding and the amount of interest in making enterprise applications as dynamic and interactive as web apps, Meteor is facing a pretty healthy opportunity.

 Photo courtesy of Flickr user tonynetone

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Infrastructure Q2: Big data and PaaS gain more momentum
• Social third-quarter 2012: analysis and outlook
• The promise of software-defined networking