Miguel de Icaza took time out from his hectic starting-a-company schedule to chat with me this morning about his new Mono venture, Xamarin, which he and other Novell ex-pats are just getting off the ground. He announced the startup in his May 16 blog post, hot on the heels of news that Attachmate Corporation, which acquired Novell in April, had laid off virtually all members of the Mono team.
De Icaza was able to confirm those layoffs with some confidence, because he had to conduct them himself. "It really could have been worse," he said. "I had to say, hey, you're no longer employed, but then I could say, and neither am I."
De Icaza is, of course, the originator of the Mono Project, an open-source implementation of the .NET Framework based on C# and the Common Language Runtime (CLR). In 1999 he and Nat Friedman founded a company called Ximian, originally to provide Linux and Unix desktop apps based on the GNOME platform. That company became a driving force behind Mono, and it was acquired by Novell in 2003.
But Mono never truly flowered under Novell, de Icaza said, especially in the mobile space. In fact, one of the reasons he is sanguine about the layoffs is that he and his Mono mates have been trying to spinoff the technology from Novell for more than a year.
"We all thought that Mono could really shine on its own," de Icaza said. "For years we've been wrapped in this bubble of Novell, and our mobile efforts didn't align with the strategies of an infrastructure company. We pursued them because a lot of people asked us to do it, and we're passionate about it. I guess you could say that I welcome being laid off, because it gave us a reason to get together and do what we all wanted to do."
Under Novell, the Mono team did create MonoTouch, which allows developers to create C# and .NET apps for iPhone, iPad, and iPod Touch devices, and Mono for Android for developers creating apps for devices running that OS. Novell owns both of those technologies, which means that Xamarin will be building some of its initial products from scratch, including a new commercial .NET offering for Apple's iOS, and a new commercial .NET offering for Android. Xamarin will continue to contribute to the Mono project and the open source implementation of Silverlight known as Moonlight, de Icaza said, and he plans to explore "Moonlight opportunities" in the mobile space and the Mac appstore.
"At Novell we built both open source and proprietary products," de Icaza said. "The open source code is, well, open source. And the proprietary things are owned by Novell. We need to rebuild those things if we want to be in the market with similar products."
De Icaza added that he'd be happy to take over those proprietary projects if Attachmate decides not to pursue them, but he's not waiting around for that decision to be made.
"This is a very big acquisition," he says. "They have their own execution plan, and Mono isn't necessarily their top priority. So, we need to have a plan that doesn't depend on their timetable. We can't really afford to wait around."
De Icaza wouldn't offer any name or numbers, because "all the paperwork hasn't been completed yet," but he did tell me that most of the laid-off Mono team members would be joining the company.
On the speculation of the funding behind Xamarin, de Icaza says his fledgling company currently has two committed angels, neither of which is Microsoft, and he's providing some of the initial funding himself. "I'm not going to rule out that I'd like to have Microsoft as an angel investor," de Icaza told me. "We have talked to a lot of people at the company, and they do like us. But for that matter, I wouldn't mind having Google and eBay and RedHat and eTrade and Novell. We're pitching to everybody who will listen right now."
And what about that new company name?
"We've used monkey themes for many years," de Icaza said. "Ximian was a play on simian. Mono is Spanish for monkey. Xamarin comes from the tamarin monkey. And we kept the X, though to tell you the truth, I can't remember why we used it in the first place."
De Icaza expects his new company, which will be headquartered initially in Boston, to be incorporated early next week. Stay tuned.
Posted by John K. Waters on May 20, 20114 comments
"Refactor mercilessly," say the Agilistas -- especially the XPers. Good advice, but the process of changing a program's internal structure to make it easier to understand and cheaper to modify without altering its external behavior can be challenging and messy.
Enter Headway Software, a Waterford, Ireland-based company on a mission to make refactoring easier and more effective. The company's newly released refactoring tool, Restructure101, is the fulfillment of a vision, the company's CEO, Chris Chedgey, told me in an e-mail, of "nothing less than the manipulation of software files and functions with the same ease that has been available to hardware engineers for decades."
"The key is to understand the vast number of interdependencies in as much detail as necessary," Chedgey said, "without overwhelming the user. The LSM was the breakthrough we needed."
The LSM, or Levelized Structure Map, is an interactive visual model designed to allow software architects to delve into structural problems and explore solutions by directly manipulating the model in a sandbox environment, Chedgey said. LMSs "levelize" items into rows or levels. Items in the same row are not interdependent, but every item on that level depends on at least one item on the level just below it. This display technique allows developers to see the entire code base at once, including dependency information.
The maps actually exist within a simulated sandbox, so it's simple to drag-and-drop hunks of code from level to level, untangle the code and reduce complexity. Once the refactoring simulation is completed, devs export an "action list" to their favorite IDE and work through the actual changes there.
Both Java and .NET IDEs are supported. And Headway provides a Web-based repository for tracking progress on a project-wide basis.
Restructure101 provides a number of discovery and manipulation capabilities on top of the LSM. Among them: the ability to find package tangles and break them by dragging and dropping packages, classes, and methods/fields. It also supports other direct manipulations, such as creating sub-packages or inner classes (wrapping), removing package structure (flattening), and automatic repackaging.
Headway bills itself as the first independent software vendor to focus purely on software structure and architectural control and the fundamental tenets of building maintainable and extensible software. The company's products (Structure101, Structure101 Build, and Restructure101) use reverse engineering, structural analysis, and architectural mapping techniques for Java, .Net and C/C++. It also offers third-party parsers for ActionScript, PHP, SQL, SysML and UML.
Posted by John K. Waters on May 19, 20110 comments
Google sure knows how to get the attention of software developers -- 5,000+ of whom nearly blew the roof off San Francisco's Moscone Center West during the opening keynote of the search giant's annual Google I/O Conference on Tuesday when they learned they would each be getting a free Samsung Galaxy tablet.
But getting their attention and winning their hearts and minds are two different things, and the latter is absolutely essential if the company really wants to become a platform player.
This isn't the first time Google I/O attendees went home with pricey swag. Google started what has become a tradition by handing out Android phones at the first event in 2008. Redmonk analyst Stephen O'Grady, whom I ran into at the show, calls this a seeding strategy. If the success of Android phones is any indication, he said, it's a strategy that's probably working.
"This is a numbers game," he said. "There are, what, a couple thousand iPad apps? Android tablets have a long way to go to catch up to that. On the other hand, Android phones have a bigger market share collectively than the iPhone. It could be argued that giving away the hardware platform you'd like developers to target is an effective strategy."
Industry watchers at comScore reported in March that Android smartphones actually moved into the number one spot in January, with 31.2 percent of the smartphone market. In the fourth quarter of 2009 and the first quarter of 2010, Apple's iPhone OS hadmore than triple the market share of Android phones, according to Nielsen. The iPhone OS market share was 28 percent, compared to Android's measly 9 percent.
That's a hell of a harvest.
The limited-edition Galaxy 10.1 Honeycomb-based tablet (Android 3.0) is a sweet device, and the hundreds of attendees hunkered down in every nook and cranny of the conference center fondling them seemed entranced. But Google's "seeding" didn't stop there: On Wednesday the company added Verizon 4G LTE hotspots to the swag bag, along with a promise to provide every attendee with a Chrome-powered notebook when the first such devices hit the market in June. (More on this from my colleague Kurt Mackie here.)
Day One of this year's conference was all about Android. Google execs talked up the upcoming Android 2.4 release, clunkily nicknamed "Ice Cream Sandwich." No launch date was given, but a big promise was made: this release will merge Android 2.3 (Gingerbread) with Android 3.0 (Honeycomb).
"Our goal with Ice Cream Sandwich is to deliver one operating system that works everywhere, regardless of device," Hugo Barra, product management director for Google's Android group wrote on the Google Code Blog. "Ice Cream Sandwich will bring everything you love about Honeycomb on your tablet to your phone, including the holographic user interface, more multitasking, the new launcher and richer widgets."
During his conference opening keynote, Barra also promised to address the platform fragmentation problem that has plagued the Android ecosystem almost from its birth -- 300+ devices with who-knows which version of the OS; it ain't pretty. Google is partnering with a bunch of companies to coordinate the update process. The "founding team" includes Verizon, HTC, Samsung, Sprint, Sony Ericsson, LG, T-Mobile, Vodafone, Motorola and AT&T. Customers buying new smartphones from these vendors will receive Android platform updates for eighteen months after the first launch -- if the hardware supports the update, Barra said.
If these companies can get their upgrade act together, they'll improve the platform for consumers, certainly, but they'll also make life easier for Android developers (who cheered the announcement almost as loudly as they did the free tablets). And that's bound to win a few more hearts and minds.
Day One of the conference also saw the launch of the headline-grabbing Music Beta program, a cloud-based music service available now by invite only; a new Web-based movie rental service available from the Android Market ; and a new hardware/accessories support offering, dubbed Android Open Accessory. All very cool.
Day Two of the show shifted to the Chrome OS and the new Chromebooks, which prompted one attendee to opine, "It feels like Google is competing with itself with Android and Chrome."
To me Google's Chrome OS strategy seems to be the realization of a concept former Sun Microsystems' CEO Scott McNealy flogged for years: "The network is the computer." I'll be watching with interest to see if, through Google, the industry has finally caught up to that vision.
As for what one attendee called Google's "Oprah moment," I encourage the company to continue seeding the developer ecosystem with hardware platforms at its annual conference. Two words to consider as you plan next year's show: vehicle telematics.
Posted by John K. Waters on May 12, 20111 comments
When I heard that Oracle would be proposing the Hudson project to the Eclipse Foundation, a few things struck me: 1) Oracle is getting better at dancing back from stepped-on toes in lively open source communities; 2) This is the very move IBM made when it created the Eclipse Foundation; and, 3) What will this move mean for the Jenkins fork?
During the first few years of the Foundation's existence, virtually all of my conversations with executive director Mike Milinkovich included a question about "demonstrating its independence" from IBM. I asked him if he thought Oracle's move would effectively satisfy concerns about Oracle's stewardship and perceived control of the Hudson project.
"Absolutely," he told me in an e-mail. "By the time the project is officially created, the 'Hudson' trademark will be the legal property of the Eclipse Foundation. The Hudson project will be a diverse, community-led project running under the Eclipse community’s development and IP processes and rules. Actions speak louder than words, and those are some pretty tangible actions."
Several companies are going to be providing development resources for an Eclipse-based Hudson project, including Sonatype, Tasktop and VMware. VMware's participation in particular will be a boost to the project, says Jason van Zyl, founder and CTO of Sonatype, because of the SpringSource connection. (VMware owns SpringSource.)
"Many users of Hudson work in the enterprise Java space so an endorsement and provisioning of resources for the Hudson project by SpringSource is a huge vote of confidence," he said.
As for the Jenkins fork, "If this move had happened some months ago, it may have prevented [it]," Milinkovich said. "But at this point it's safe to assume that the split will remain. I am as interested as anyone to see how this will all play out."
Oracle's announcement likely came as a surprise to the Jenkins community. Oracle kept an unusually tight lid on this announcement. (I had to pinky swear to keep it under my hat until this morning to get an early briefing.) The Eclipse Foundation's director of marketing, Ian Skerrett, even tweeted an apology to @jenkinsci: "If anyone in the Jenkins community has questions about Eclipse feel free to contact me; sorry for your morning surprise."
IDC analyst Al Hilwa noted that the Jenkins community seemed to have been operating under the assumption that Oracle "was going to do some bad things with the code," and overreacted when they created the fork. "It may be time to come together again under one code base, now that it's under the Eclipse foundation," he said.
UPDATE: Hudson creator Kohsuke Kawaguchi just posted his reaction to the news today on his blog. He said the news was "quite a surprise," and added:
"[On the] one hand, I think this definitely shows the great success of the Jenkins project post divorce…. Were it not for the success of Jenkins, they wouldn’t be giving up the project…. But at the same time, I just wish Oracle saw that coming a few months earlier, while we were still seeking the middle ground. We were very interested in having the trademark moved under the custody of a neutral third party, but they were very clear that that’s not acceptable to them. And it also disappoints me that they decided not to reach out to the Jenkins community about this move, but I guess they are never really interested in working with us."
I asked Mik Kersten, CEO of Tasktop and founder of Eclipse Mylyn, what he thought this move would do for users of the popular CI tool and the growing population of plugin makers.
"For users the answer is straightforward," he said, "as Eclipse is a clear go-to place for enterprise-ready open source tools. For plug-in makers… Eclipse has very clear and hardened guidelines on APIs, which provide a significant benefit to any growing plug-in ecosystem. However, a key reason for the success of Hudson was a rapid pace of innovation in the plug-in ecosystem, which has, in part, come from Hudson’s permissive approach to committer rights." "
"The rate of innovation that comes from this kind of freedom of evolution will need to be combined with the API hardening needed by enterprise consumers," he continued. "Eclipse’s endorsement of GitHub as an extension of the Eclipse contribution ecosystem is one mechanism that could help combine the best of both worlds."
Posted by John K. Waters on May 4, 20110 comments
If you haven't been following the evolution of the Jease open-source content management system (CMS), you should. The now two-year-old project was conceived to solve some of the problems developers face when building database-driven Web applications with Java.
Jease (an abbreviation of "Java with Ease") started out as a basic CMS framework with no "system" pretentions. Back in early 2010, project founder Maik Jablonski declared his simple aim to create a framework that makes it easy for Web devs to create custom content structures (FAQs, special Web site content sections). And he built the framework on top of three open source technologies: the db4o object-oriented database, the Apache Lucene information retrieval library, and the ZK Ajax Web application framework. Jease now also supports McObject's Perst OO embedded database.
Since then, the project has bloomed with an expanded mission to create a full-blown CMS, while honoring its keep-it-simple roots. The project is evolving fast: Version 1.7 was released in February, and 1.8 was released in March.
Version 2.0, which was released this week, represents a "big leap" toward turning the Jease framework into a true CMS, Jablonski says.
"The main theme for the 2.0 release line is to continue the journey from a developer-oriented Content-Management-Framework to a user-friendly Content-Management-System which tries to make publishing in the Internet as easy as possible," he wrote in in a blog post announcing the release. "Nevertheless, Jease targets Java Web developers who want to things getting done without headaches."
The latest release adds a number of features and capabilities to the evolving CMS, including instant preview, a link checker, a redirect service, translation capabilities, and instant relocation handling of moved and/or renamed content. A full list of new features and bug fixes in Jease 2.0 is available on Jease.org.
Jease 2.0 is available under the GNU General Public License V3. There's a nice demo available here. Keep in mind that, though jease.org runs on the Apache Tomcat servlet container, this demo is runs on Jetty.
Posted on April 28, 20110 comments
Zend Technologies and Rightscale announced this week a jointly-developed platform-as-a-service (PaaS) architecture for PHP developers. The announcement was kind of overshadowed by the big VMware PaaS news, but this is a dynamic duo you should keep an eye on.
Zend, of course, is the Cupertino, Calif.-based creator and commercial maintainer of the PHP dynamic scripting language. Zend is run by Andi Gutmans and Zeev Suraski, who are key contributors to PHP and the creators of the core PHP scripting engine. RightScale is a Santa Barbara-based provider of an automated, Web-based (and eponymous) cloud management platform.
The heart of RightScale's management platform is a set of pre-built ServerTemplates for common server configurations. "It's a way to assemble a machine configuration out of building blocks," RightScale's CTO and co-founder Dr. Thorsten von Eicken explained to me last year, when his company first partnered with Zend to release a set of RightScale ServerTemplates for deploying Apache with PHP. Working together, the two companies made it possible to deploy Zend-based PHP apps across multiple clouds via the RightScale management platform.
Now they've released the RightScale Zend PHP Solution Pack, which is designed to provide massive scalability and high availability for PHP apps running in the cloud. The solution combines the RightScale management platform with the Zend Server Web app server.
Rightscale CEO and co-founder Michael Crandel told me that the new product was a direct response to customer demand.
"We've seen a real movement toward platform-as-a-service and ease-of-use for developers, who really don't want to mess with these ugly things called servers and storage volumes, but just want to upload their code and run it," Crandel said.
"Both companies wanted to create a PHP PaaS, with the characteristic elasticity of a cloud platform, but also one that would be customizable," said Kent Mitchell, Zend's director of product management. "We've combined the best practices of Zend with the best practices of RightScale to create a reference PHP architecture that you can launch basically the push of a button."
A big selling point for this solution is likely to be its flexibility -- Mitchell calls it "open PaaS" or "customizable PaaS." A platform for PHP from Zend's perspective, he said, shouldn't be tied to single cloud provider, should be application-centric, and should handle the app whole lifecycle.
"If a customer wants to move from Amazon to Rackspace, they should be able to do that," he said. "And as we move forward, people aren't going to be monitoring servers anymore; they're going to be monitoring applications. They'll want to know things like, I'm having a problem with my e-commerce application, not that server XYZ is heavily loaded. And though lots of PaaS companies handle just the deployment or test or just one part of the lifecycle, we believe that it's important to the full lifecycle: development tooling, test and QA, staging, and production deployment."
"Because of the way RightScale works, and the way Zend is very flexible," Mitchell added, "what we've got here is a solution that's a PaaS, where you can sort of lift the hood of the car, tweak the engine a little bit if you want, slam the hood back down, and now you've got your PaaS. If you don't like MySQL as the database and you want to use, say, Oracle, you just lift the hood, pull out MySQL and put in Oracle, and close the hood. Now every developer that fires it up will start with an Oracle instance."
The RightScale Zend PHP Solution Pack comes with a bunch of tools and features, including a production-ready, cloud-based environment that includes the Zend Server, the Zend Server Cluster Manager, and RightScale's Premium Onboarding, which is a step-by-step path to deploying on the cloud using best practices from the two companies.
Stephen O'Grady, Principal Analyst with RedMonk, contributed a comment for the press release that I thought was worth passing on: "Considering PHP's ubiquity on the web, it wasn't a question of if it embraced the cloud, but when," he said. "With the recently announced RightScale/Zend partnership, the two companies are offering PHP users the best of both worlds, with the time to market of Platform-as-a-Service (PaaS) and the flexibility of Infrastructure-as-a-Service (IaaS)."
More details about the solution pack are available on the Web site, where you can also sign up for an intro Webinar scheduled for April 28 at 11:00 am PDT.
Posted by John K. Waters on April 21, 20110 comments
VMware has launched an open Platform-as-a-Service (PaaS) product, which my colleague Michael Domingo reported on last week. The Palo Alto, CA-based virtualization vendor is billing its new Cloud Foundry as the industry's first open PaaS offering, a "new generation of application platform, architected specifically for cloud computing environments." Cloud Foundry is available as a cloud service operated by VMware, but also as a downloadable VM called "Micro Cloud." It's still in beta, but the source code for the project is available now on http://cloudfoundry.org/.
I heard from several people about this beta release. IDC analyst Al Hilwa, who was in Las Vegas covering the MIX11 event when the news broke, was the first to fire off an e-mail.
"Here is an infrastructure player that understands the value of application platforms and has moved to systematically assemble a cloud stack through smart acquisitions," Hilwa said. "While we have heard similar announcements from VMware a couple of times before, this one looks like the real thing."
It's "the real thing," Hilwa believes, because VMware is finally ready to put the platform into the hands of developers to play with, and plans to release the entire stack in open source "for anyone to leverage."
"While VMware's offering will be competing with Microsoft Azure, it will more directly compete for Java workloads with RedHat," he added. "VMware has set itself the ambitious goal of providing a choice of frameworks so that apps can be ported from different worlds [e.g. Java with Spring, Ruby on Rails, etc.]. This is an ambitious approach, and we should watch VMware to see if it's able to add more and more frameworks quickly enough, or generate the open source traction, to bring others into the fold."
Cloud Foundry is an important strategic move that positions VMware as "another emerging pole for Java developers," Hilwa says.
"We are living in disruptive times," Hilwa said. "And it's at exactly at times like these that players move into adjacent businesses and try to change the game. VMware clearly hopes to do that for PaaS platforms."
I also got a heads up from the indefatigable Liz Clinkenberg Christina Dalit about blog postings from the SpringSource crew on the news. It's been about 20 months since VMware acquired the chief commercial sponsor of the open source Spring Framework project. And it's still a surprisingly vocal group.
Rod Johnson, SVP and GM of VMware's SpringSource product division (and founder of SpringSource), posted what started out as a slightly canned-sounding statement about the news, but went on to offer a nice diagram of the Cloud Foundry model. He also observed: "To date, there hasn't been a strong, open PaaS destination for Java. The millions of Java developers have largely been left to fend for themselves in the cloud, with weaker options than have been available, to, say, Ruby developers. We're changing that."
He also encouraged developers to sign up for a beta account here, to download the source code and get involved in the project.
SpringSource team member Peter Ledbrook does a good job of explaining how Grails and Cloud Foundry work together on his blog "One Step Deployment with Grails and Cloud Foundry." Team member James Tyrrell talks about the integration of Cloud Foundry and Spring Roo on his blog "Roo + Cloud Foundry = Productivity in the Cloud." Great posts.
Cisco Cloud Evangelist Brian Gracely makes a stab at providing "101 Thoughts about the 'Cloud Foundry' Announcement" on his JavaLobby blog. I didn't count them, but I think he gets pretty close.
VMware has published a useful FAQ page, and you'll find video on YouTube on the CloudFoundry Channel. I liked the video by Jerry Chen, senior director of VMware's Application Platforms group, but there are others, and I'd expect more to come for a while, so you might want to subscribe.
Posted by John K. Waters on April 18, 20111 comments
What's the first big-enterprise name that pops into your head when you hear "Java community" these days? Oracle, of course, but also IBM, and lately Apple. How about Red Hat?
Red Hat doesn't always get recognized for its efforts to move Java forward. And frankly, they're kinda tired of going unnoticed.
"We often don't get the credit we're due when it comes to moving Java forward," Rich Sharples, director of product management in Red Hat's Application Platforms group, told me the other day. "Red Hat is a very active contributor to all things Java. We're the largest contributor to the OpenJDK, outside of Oracle itself. And for the last couple of releases -- Java EE 5 and Java EE 6 -- we played it pivotal role in making Java EE a much more accessible and product platform for developers."
As of today, you can add Java EE 7 to that list: Red Hat has submitted three new Java Specification Requests (JSRs) for the upcoming Java EE 7 (JSR 342). The new JSRs were submitted to create standards for Platform-as-a-Service (PaaS) and cloud deployments, Sharples explained. As the press release announcing the JSRs put it, they offer Java EE 7 users "a new consistency, flexibility," and "portability across on-premise and cloud environments, and they reflect Java EE's position as one of the most flexible and adaptable application platforms in the market."
Red Hat was responsible for the Context and Dependency Injection (CDI) spec in Java EE 6 (JSR 299), so it's not surprising see CDI 1.1 heading its list of proposed JSRs for Java EE 7. The purpose of JSR 299 was to unify the JSF managed bean component model with the EJB component model, creating a significantly simplified programming model for web-based applications. CDI 1.1 goes on to simplify and unify various layers of a Java EE application with a standard model for Dependency Injection and lifecycle management. The components are allowed to interact in a loosely coupled way, and the developer can build even simpler, more easily maintainable, and more portable apps without having to introduce non-standard, proprietary frameworks.
CDI was the brainchild of JBoss Fellow Gavin King, has become the standard for dependency injection for Java EE. The spec, formerly known as Web Beans, grew out of an open source project called Seam, an application framework on Java EE 5, which he also created.
The company is also proposing a specification for distributed data grids, which Sharples claims will allow easier and more cost-effective scaling for the data-tier, which is essential for large-scale cloud or utility computing and multi-tenant application platforms.
The third proposed JSR -- Bean Validation -- is described by the company as a means of providing "a consistent approach to implementing validation logic throughout an application, from the persistence layer to the presentation layer. The unified approach helps avoid duplication and inconsistencies, ultimately resulting in speedier development."
The new specs aren't actually JSRs yet; they still have to go through several stages within the Java Community Process. From the initial JSR Review stage, in which Red Hat's proposed specs will be evaluated and approved by the Executive Committee (EC), they will move into the Early Draft Review stage, the Public Review stage, the Proposed Final Draft stage, then Final Release, and finally Maintenance. Along the way, they could be voted down by the EC, withdrawn by the spec leads, or rendered "inactive," the fate of JSRs that have posted neither a Final Release nor a JCP milestone draft for the last 18 months.
I don't blame Red Hat for blowing its own horn about its role in the evolution of Java. The company has served on the EC, and it has been a Java spec leader or Expert Working Group member for more than 35 JSRs. The first Context and Dependency Injection spec -- SR 299 -- had a big impact on Java EE 6. The company was behind Enterprise JavaBeans (EJB) 3, JavaServer Faces, and the Java Persistence API. And it's currently is collaborating on four existing JSRs for Java EE 7: Java API for RESTful Web Services 2.0 (JSR-339); Java Message Service 2.0 (JSR-343), Java Server Faces 2.2 (JSR-344), and Java Content Repository API (JSR-333).
"I think most developers understand Red Hat's role in moving Java EE forward," Sharples said. "Our approach to delivering new technology, rather than the ivory-tower, closed-door, designed-by-committee approaches some of these other organizations take, we're on the ground, so to speak. And we do get credit from the developer community, but we certainly don't get noticed by the mainstream press."
No one would ever call me mainstream, but in the spirit of giving credit where credit is due, my hat's off to Red Hat.
Posted by John K. Waters on April 14, 20110 comments
We reported last week the release of a new version of a free and open source testing tool for iOS apps called FoneMonkey from a company called Gorilla Logic. The Broomfield, CO-based company was founded by a group of former Sun Microsystems execs in 2002 as a software services firm specializing in rich Internet applications (RIAs) and enterprise app development with Java, Adobe Flex and mobile platforms.
I had a chance last week to chat with Stu Stern, the company's president and CEO.
"We're not focused on any particular vertical," Stern told me. "We're a bunch of hard-core geeks here, so the real focus is the technology."
That bunch includes Stern, who ran the Sun Java Center, Sun's global Java professional services organization; VP of engineering Ed Schwarz, who founded the global e-Business consulting organization at Sun; and CFO Hank Harris, who directed Sun's Professional Services group, which was responsible for telecom accounts in North America.
It's not surprising that the company was initially very Java focused. But tracking trends in the industry over the last eight years led the operation into RIA development in general and Flex development in particular, for both enterprise and mobile markets. (The Flex SDK was designed for developing and deploying cross-platform RIAs based on the Flash platform.)
"One of the big deficiencies we've seen in the RIA space is automated testing support," Stern said. "We're pretty serious about Agile around here, and we believe that automated testing is critical to Agile development -- we don't see how you can do refactoring of an application if you don't have an automated test to back you up."
FlexMonkey was the company's first response to these observations. It's a free Adobe AIR application designed to record, playback, and verify Flex UI interactions. The tool also generates ActionScript-based testing scripts that can be included within a continuous integration environment.
Last year "the Gorillas" turned their attention to the iPhone, where, Stern said, the need was even more acute. "In the Flex space most of the tools were oriented toward QA specifically, and weren't well suited to the type of testing that developers do," he said. "That's what drove the development of that tool. FoneMonkey follows that same philosophy of tools that can certainly be used by QA, but are very, very well-suited to developers."
Stern is the creator of FlexMonkey and FoneMonkey, as well as a tool called Fexmonkium, a free plugin for the Selenium integrated development environment that adds FlexMonkey recording and playback functionality.
"Even though we have these testing tools, we're not a testing company, per se," Stern said. "We're very much an app development firm. The majority of the work we do is just building things for people. And the tools have come out of our needs doing that work. And we're not just a Flex and iPhone shop. We do Android, and of course, lots of Java. We're pretty much right-tool-for-the-job kinds of guys. We're not religious about any particular technology."
Stern is scheduled to lead a session at the 360 Flex Conference, underway this week at the Marriott Denver South in Colorado (April 10-11). As the name implies, the event is for Adobe Flex developers, (rumors of the death of which have been greatly exaggerated). Stern's session is called "Automating Functional Testing." He plans to talk about how a "core suite of automated functional tests" provides a solid foundation for rapidly iterating product releases "by ensuring [that] the introduction of each new feature doesn't inadvertently break pre-existing functionality," the website promo states.
I'm guessing FlexMonkey will come up in the discussion.
Given his company's investment in Flex tools, I wondered whether Stern was concerned about the apparent rise of HTML5 over Flash.
"We're not worried, first, because we're not betting the company on it," he said. "I do think Steve Jobs did a great job of spreading FUD when he wrote the infamous letter [declaring that Apple would not support Flash in Apple's mobile products]. After that letter went out, people started asking us about HTML5. But HTML5 is not really an answer to Flex. At this point in time it's an answer to Flash video -- you've got good H.265 video support in HTML5 -- but it's not a development platform for RIAs."
"We still love Flex," he added. "We just think Adobe/Macromedia nailed it in most respects in terms of making it easy to build a rich Internet application. And frankly, it doesn't seem like Apple is focused on the developer; some people around here believe that the company has downright disdain for developers. But ultimately, as hard core geeks, we don't really care. Confusion and complexity drive consulting."
And I had to ask about the future of Java under Oracle.
"Look, it's not like the future of Java is really in jeopardy at this point," Stern said. "Sun clearly dropped the ball on the user interface side (ironically, since that's where it all started), but Java still owns the backend. We are seeing some significant PHP back there, but for most hard-core back-end dev, it's still Java. And the tool support for Java is so so mature. It's an incredibly productive development platform. We think the preference for scripting languages we're all hearing about is overblown. We just don't see the leverage you get with something like Ruby."
Posted by John K. Waters on April 11, 20110 comments