Debugging Software

New Survey Results: Mobile Apps Move to HTML5 and JavaScript

Larry Lunetta — Fri, 16 Mar 2012 20:50:31 +0000

Each year in the Spring we pick a topic to do a survey and this year (no surprise given our mobile initiative), we assembled about a dozen questions covering the development of mobile app’s.

I’ll use this and the next several blogs to cover the results but, as always, the results have been very eye-opening. First, some background on the those who responded.

• 80% participate in some part of the mobile application life cycle: development, test, project management and support.
• We had a nice mix of organizations. Half the responders came from organizations under 1000 people.
• The target(s) for their mobile apps were split along the lines of consumer/customer-only (50%), internal/employee-only (20%) and both consumer and employee (30%).

Given the proliferation of device types and operating systems, one of the trends we were most interested in is how much native vs. HTML5/JavaScript development is being done. Surprisingly given the conventional wisdom that most mobile app’s are written in a native language (e.g. Objective C or Java), we found that 70% of those who responded developed either full HTML5/JavaScript or Hybrid applications.

We’re still doing the slicing and dicing on the survey results to better understand if size of organization, user target, vertical industry, etc. has more or less influence on that 70% number, but overall it validates something else we have been hearing a lot: IT organizations are really struggling to provide mobile users access to enterprise applications and services and that writing custom-code for each device is a losing battle.

Hence, the pendulum is swinging to higher-level languages like HTML5 and JavaScript, aided and abetted by mobile application platforms like JQuery, Titanium and PhoneGap. More on that in our next blog—we’ve got some data on that too.

Going Mobile

Larry Lunetta — Tue, 28 Feb 2012 19:35:47 +0000

I guess the question is “who isn’t going mobile?” When enterprises woke up on Jan 1 and found that in Q4 well over 100 million smartphones and 25 million tablets were shipped –whether they liked it or not, they were not just “going mobile” but running to catch the train.

We’ve been watching the mobile space for about a year and as recently as last summer, mobile application development for our enterprise customers was more of an horizon event than a current project. But now, we’ve clearly reached a tipping point–what a difference 6 months makes.

With smartphones and tablets crashing into enterprise IT, everything from mobile device management to mobile security to mobile application development is a number one priority. Predictably, the requirements are many and the answers are few—especially given the n-dimensional problem of hardware, operating systems, languages and networks.

This all preamble to the news that we’ve just announced two products focused squarely on mobile application development and deployment for HTML5 and JavaScript–across all device and all operating systems. Our application record and replay technology—so potent for C, C++ and Java—bridges some deep gaps in mobile application development vis a vis standard enterprise web applications. Everything from debugging to testing to security to real user monitoring and customer support reminds us of the state-of-the-art circa 1999.

In addition, there is a marvelous ecosystem of HTML5- and JavaScript-focused Mobile Application Development Platforms such as Titanium, Sybase Unwired Platform, Adobe AIR and PhoneGap—a natural home for our mobile products and where we fit seamlessly and productively. Plus Microsoft is coming with Metro.

So, we are excited about the mobile space and ReplayMOBILE and apmMOBILE are the first of a family of mobile products that will ship this year. Because of where we sit in the code execution on smartphones and tablets, we see everything from what part of the code is exercised to what works (and doesn’t) through what data flows through to “did that last dot release of the O/S just break our app?”

We’ve heard “wild west” often applied to mobile application development. Here comes the cavalry.

Check out our HTML5 and JavaScript products at

http://www.replaysolutions.com/products/replaymobile.

405 and Counting

Larry Lunetta — Wed, 02 Nov 2011 00:24:43 +0000

We do webinars about once a month on various topics across the software development lifecycle. The common theme of course is how application record and replay can dramatically and demonstrably improve defect resolution.

Usually we average between 150 and 200 registrants. However, this month’s topic has driven sign-ups absolutely off the charts. Seems like test automation is a hot button topic right now and our webinar on Wednesday, Nov 2 (Automate testing for Java EE apps – Selenium, HP ALM, LoadRunner) already has 400+ registrants with more pouring in.

I guess we shouldn’t be surprised because as organizations push more agile (and shorter) schedules, automation is crucial. For us, this trend is definitely our friend.

It starts with continuous integration and our ability to identify and explain even minor differences in test execution between code versions. After that, our ability to precisely locate regression test failures extends the ROI of that process from bug finding to bug communication, replication and resolution. We’re now even getting traction with the security folks who in addition to finding security and compliance issues via vulnerability scanners, now want to bring the issue to development in the form of our recording to accelerate remediation (after all, a vulnerability is simply just another kind of bug).

We’re excited that our automated test solutions are resonating so strongly. If you happen to miss the live webinar, check out our website where we archive all of our “greatest hits”.

Postscript: Final count 482.

LIGHTNING Strikes! Triage speed, Source-code precison

Larry Lunetta — Tue, 27 Sep 2011 13:46:38 +0000

Enterprise applications are veritable volcanoes waiting to erupt in an explosion of logs, exceptions, console messages, etc. that describe the inner workings of the code execution. Most users don’t see all that seething activity and neither do developers, testers, support or operations staff—until a problem occurs.

When an application slows down unexpectedly, malfunctions or is hacked, all those under-the-surface events take on vital importance in finding and fixing the defect. The problem is, out of the literally millions of execution events, where do you look and more importantly, if you do know where to look, how do you get the specific information that you need to diagnose the problem? Another one of those needle and haystack challenges.

Today we’re announcing the solution to this dilemma and it’s called ReplayLIGHTNING. As a real time diagnostic portal integrated with the record/replay functionality of ReplayDIRECTOR, ReplayLIGHTNING presents a graphical summary of execution behavior along with multiple views of the critical execution events that describe a defect or abnormal behavior. Think of an MRI for application execution that can be viewed in an IDE, a web UI or a dashboard in a monitoring or profiling tool.

In addition to providing highly actionable triage information, the other role for ReplayLIGHTNING is to work in conjunction with an execution recording to “data mine” for source level root cause. Because a Replay recording re-executes the binaries on playback, we can “go back in time” to ask questions that you would only know to ask in hindsight. For example, we use our knowledge of a memory leak meltdown to see when objects are allocated and if they somehow missed the GC process. So we can not only describe the memory leak at its messy end, we can also pinpoint where it all started.

ReplayDIRECTOR has always been a great “time machine” for recreating application execution with 100% fidelity. With ReplayLIGHTNING, we’ve added split-second control to dial in exactly where in that time machine you’d like to travel–with a preview of coming attractions.

Hop on board. www.replaysolutions.com/products/replaylightning

Less Filling, Taste Great

Larry Lunetta — Mon, 19 Sep 2011 12:25:18 +0000

When light beer burst on the scene a generation ago, the biggest marketing challenge was to convince consumers that brewers could actually extract calories while still maintaining acceptable flavor. OK, that was a long time ago and now we have a full range of caloric options for beer. But every marketer (and beer drinker) will remember “Less Filling, Tastes Great”.

We will be making some exciting product announcements next week that addresses a similar kind of tension—this one in the software defect resolution space. Typically if a tester, developer or support person wants to get started on diagnosing an application bug, they’ll do some screen scrapes and a log dump but what really accelerates defect resolution are source-level root cause diagnostics. With current profilers, monitors and probe products, the tradeoff between “there’s bug in your code and here’s some logs” and “you have a fatal exception and here is the offending thread and corresponding source code” has, up until now, been very difficult to reconcile.

That’s right, up until now. Given the experience we have working deep in Java code execution and the 100% visibility into program execution that our app recording technology provides, we have come up with some very clever ways to deliver both triage speed and root cause depth in the same debug UI. The quality equivalent of great tasting light beer.

Stay tuned for some truly electrifying advancements in software defect resolution. (Hint: Sept 27).

Java is BACK! See us at JavaOne

Larry Lunetta — Wed, 31 Aug 2011 21:17:11 +0000

I’m hearing from my friends in the VC community that Java is enjoying somewhat of a renaissance in terms of importance and innovation. Seems like Java’s maturity, feature set, reliability and supporting ecosystem makes it a great fit for the continuous integration, continuous deployment software processes that are fast becoming the norm.

Not surprising given that perspective, the upcoming JavaOne conference and exhibition is sold out. Now held at the same time as Oracle OpenWorld in San Francisco, JavaOne is still the place where the very latest products and technologies are on display and where new product announcements are made.

We’ll be there and we’ve got a couple of new things up our sleeve that we will be showcasing for the first time at the show. Stop by our booth #5205 Oct 3-5 and make Java code, performance and security defects a thing of the past.

Testing as Part of the Quality Process

Larry Lunetta — Tue, 16 Aug 2011 18:12:00 +0000

Given the business we are in, software quality is one of our core motivations. A recent posting on STP, the Software Test Professionals site (http://www.stpcon.com/Item/1027/?utm_source=Email&utm_medium=email&utm_content=081611-TQA&utm_campaign=NEWSLETTERS) posed the question: “Can You Really Test Quality into a Product?”. The point of the article, that “doing it right the first time” is really the only guarantee of quality, harkens back to the seminal Phil Crosby quality principles that we have talked about in previous postings.

When it comes to producing quality software, it’s hard to argue against setting (and freezing) requirements, communicating them perfectly to all constituencies and then following a well-defined and monitored development process. That’s what we all aspire to. The problem is, just like with airplanes, automobiles and pharmaceuticals, the real world of complexity, human foibles and change always intrudes. That’s why testing exists—to catch defects before they get baked into the final product.

Software testing is the safety net underneath what is hopefully sound process and practice and while no one likes bug fixing, if defects are found early enough in the cycle, properly documented and communicated with enough information to find and fix them quickly, then quality will improve. It’s not the “free” quality that Phil Crosby espoused, but with the tools and techniques that are now available to QA and Dev teams, it’s less expensive that it used to be.

Good Software Patents

Larry Lunetta — Tue, 02 Aug 2011 06:53:55 +0000

Anyone who follows the complicated and arcane world of patents understands the controversy surrounding what are known as “process” patents. Unlike patents for genuine breakthroughs like the light bulb or the transistor, process patents provide protection for weighty concepts like pricing an item on an e-commerce site or attaching a graphics file to a manufacturing procedure. Nothing wrong with this kind of innovation but not exactly rocket science and not particularly unique.

We just announced (http://www.replaysolutions.com/company/news/62) our third patent award covering various foundational elements of our application record and replay technology. This is rocket science and the technology is definitely unique.

Capturing and managing the literally thousands of non-deterministic conditions that an application can exhibit and replicating them on playback with 100% fidelity is an extremely challenging technical task. The fact that we have already been awarded three patents with nine more pending is fitting testament to the technical achievement of the Replay engineering team.
I salute our engineers for producing truly valuable inventions that have advanced the state-of-the-art for a wide range of application monitoring and diagnostic technologies. Great use of the patent system.

(Software) Quality is Free

Larry Lunetta — Mon, 25 Jul 2011 17:14:00 +0000

When the United States fell asleep at the wheel in terms of product quality, one of the “re-awakenings” came in the form of a book called Quality is Free by Phil Crosby. His core principle was that any investment in quality improvement would return more than the program’s cost in terms of organization efficiency and customer satisfaction. Hence the title of his book.
Our co-founder Jonathan Lindo is currently the featured columnist for Sticky Minds (http://www.stickyminds.com/) where he discusses the economics of software quality in the context of automated testing. Based on the experiences and feedback from our customers, Jonathan has done an excellent job of highlighting changes that are driving software testing to increased levels of automation and a “time and motion” breakdown of the costs of software defects throughout the find and fix process. Given the time and resources tied up in documenting, communicating, replicating and diagnosing a defect, there is a great deal of savings to be had in improving the process.
The article has already received very thoughtful and supportive feedback, mostly along the lines that the cost estimates may be too conservative. If that’s the case, software quality is even “free-er” than we thought.

The Value of Perfect Knowledge

Larry Lunetta — Tue, 19 Jul 2011 17:09:03 +0000

The networking folks do amazing things when they have the information they need to diagnose issues and make decisions. They use Deep Packet Inspection to inform everything from bandwidth management, equipment outages and security because DPI gives them the absolute ground truth about their network’s behavior.

We’ve just introduced ReplayDIRECTOR 3.5 and while dot releases are not typically something to shout about, this release represents the front edge of our “Deep Application Inspection” initiative that will deliver the same kind of ground truth for application execution that Deep Packet Inspection provides for the network.

RD 3.5 takes our log viewing and amplification capabilities available in the Eclipse IDE and exposes the details in our standard web UI along with information highlighting defect markers, transactions, users and even client screen shots. In upcoming releases we will be adding a wide variety of deep execution detail to the log view and exposing that data not only in our web UI, but in IDE’s and dashboards of complementary diagnostic and profiling products that provide management information across the software development lifecycle.

Because we see and replicate every method call, exception, db call, variable value, code sequence, etc. there is no more precise and comprehensive record of program execution. Deep Application Inspection is the engine for the diagnostic train and everyone from Dev to QA to Security to Operations can get on board to advance their mission of producing and maintaining high-quality, reliable business solutions.