Analysts and B2B managers say the willingness to handle B2B integration via mobile is there, but it just may not be time yet.
When you’re an IT department of two, with 150 outside sales folks plus 150 people in the building, everything just needs to work.
Too bad that really doesn’t happen very often. And Jesse Davis, network admin for pharmaceutical company Derma Tran (a job he recently left), was reaching the breaking point. The company was using Citrix Xen to run its virtual desktops and between needing an extra layer of virtual desktops to run the virtual apps (which was costly), the map drives failing, restrictions the company couldn’t change and endless errors the tiny IT staff was spending virtually all of their time simply putting out fires.
Davis decided it was time for a change and looked at cloud provider dinCloud and its options for virtual private data centers. Dave Graffia, vice president of sales for dinCloud, explained that simplicity, cost and performance are what brings customers looking for help. “The whole process of building a data center from scratch is just becoming overwhelming now in a way it hasn’t been before,” he said. “There are lots of moving parts, R&D, capital investment and so many layers. Our best customers end up being those people who tried it and then came to us for help.”
Cost of course is also incredibly important. Graffia pointed to a Gartner Group study that showed desktops cost companies $300 to $500 a month. Virtual desktops are of course much less pricey, and using dinCloud, companies can provision or de-provision quickly. And thanks to good tracking, the dreaded “cloud cost creep” shouldn’t be an issue.
“From an IT perspective, the virtual workspace is very easy to use,” Davis said. “Now we take maybe 10 percent of the calls we used to.” In fact, his small team is now able to tackle some other niggling jobs — like email encryption and remote device management — and eventually they’re hoping to redo the camera system around the entire compound.
Choosing a workable solution has changed everything for the IT team, Davis stressed. “It has made the world of difference for us. We’ve gone from being constantly on the phone with the cloud provider to being the Maytag Man.”
Today, midsized companies are struggling to hire tech talent, especially those with advanced DevOps and Enterprise Architect-level skills. Fortune 2000 companies have the financial clout to attract developers and IT services providers. Meanwhile, SMBs often face a tough, multi-month slog to find contract tech services providers, according to Steve Mezak, CEO of Accelerance, a company which has created a global network of development services providers.
Mezak’s assessment reminded me of interviewing Java FX expert Rob Terpilowski at JavaOne 2015 about his challenges in hiring DevOps contractors. The process of filling contract positions used to take a couple of weeks and now takes two-to-three months, said Terpilowski, software architect for Lynden Inc., a freight shipping and logistics company in Seattle.
Mezak and Andy Hilliard, Accelerance president, talked to me recently about the current developer shortage and their IT outsourcing services. We also discussed why the rise of connected applications is driving the need for enterprise architecture change and, logically, for highly-skilled enterprise architects.
Mezak and Hilliard did a lot of legwork before launching Accelerance’s IT outsourcing services. “We got on airplanes and visited and evaluated IT services companies around the world,” Mezak said. “We looked at hundreds of companies and ranked their expertise in about 40 different technology areas.” So far, Accelerance has 50 certified partners in 30 different countries.
Companies using Accelerance usually find DevOps service providers in six days or less, largely because Accelerance has a list of who has the right tech skills to offer. Just evaluating independent developers or service firms’ skillsets takes more time than most businesses have, said Mezak.
Hilliard added that the complexity of IT today increases the need for and difficulty of qualifying contractors. They need experts who can help with decisions about and implementations of continuous integration and continuous deployment processes, containerization, microservices and more. “There’s a need for a much more sophisticated level of talent, more at the architect level,” he said.
It’s unusual for companies to ask Accelerance specifically for DevOps experts, when seeking personnel. “They’re saying, ‘I just need help getting these web applications built and deployed,’” Mezak said. That can often require what is called DevOps expertise, of course.
But IT architectures are much more dynamic today than they have been it in the past. “Architectural support is critical now because how these applications are deployed and how they communicate,” Mezak said.
Virtualization and the cloud have enabled organizations to process staggering amounts of data by unbuckling server-side limitations. Today, companies handle billions of pieces of data per day, intensifying the challenge of analyzing that data. Experts predict organizations that integrate data analytics technologies with business processes and enterprise architectures will gain a competitive advantage in 2016 through the use of statistics-backed user and market information. Continued »
How hard is it to recruit tech talent in the San Francisco Bay Area? “Ridiculously hard when compared with other parts of the world,” according to Jason Smale, director of product for San Francisco-based Zendesk. While researching the challenges of hiring IT pros for our recent developer shortage survival guide, I talked with Smale, an Australian native, about his experiences with recruiting tech pros in various global markets and Zendesk’s distributed development team approach.
“San Francisco is a very different hiring landscape from Australia, Dublin, Copenhagen, Singapore or anywhere,” Smale said. “It’s always going to be competitive, because you’ve got some of the best minds in the technology space here.” This concentration of talent fosters an abundance of tech start-ups. The 2015 Startup Genome Project surveys put Silicon Valley first in number of startups, and the area has been number one for years.
Competition for software developers and engineers is intense for another reason: the number of product-oriented technology companies in the San Francisco-Silicon Valley area is also greater than in other tech hubs. “It’s a very different mix from different regions around the world,” said Smale.
In Copenhagen, Melbourne and Dublin, for example, Zendesk competes for tech talent against banks and large telecommunication companies. “There technology is a supporting function rather than the core business, which is the opposite from San Francisco,” he said. It’s easier for a technology company to recruit software engineers there, because “technologists want to come to organizations where they’re going to have the best impact.”
The tough competition for talent in Silicon Valley led Zendesk to move to distributed engineering teams. “We committed to having engineering resources outside of San Francisco,” he said. Today, 50% of its developers reside outside of San Francisco.
The benefits of the distributed team model outweigh the negative of not having face-to-face interaction. “Having a global presence has really increased the diversity inside our organization, which is really powerful to the culture,” Smale said. “Also, it removes the groupthink in many ways and has people challenge each other with alternative technical solutions. It has led to our building better product.”
Time zone issues are the biggest barrier to making distributed development work well. Teams at Zendesk had to be organized so that people working on the same piece of a project have a two-to-three hour time zone overlap.
To illustrate the problem, Smale noted that 8 a.m. in San Francisco is 9:30 p.m. in India, but 1 p.m. in San Francisco is 8 a.m. in Australia. Obviously, teaming engineers in the two latter locations makes sense.
Organizing distributed teams so that no developers have to work “crazy hours” is an example of Zendesk’s focus on retaining employees. Another example is the company’s move to offering 16 weeks of paid leave for both mothers and fathers of newborns. “We think about what policies will help our teams remain sustainable to do long-term,” said Smale.
Zendesk’s approaches to thriving during the tech talent shortage are similar to others our editors and reporters have uncovered while reporting on this topic. For more information, check out articles on worldwide developer pay trends, improving developers’ quality of life as a hiring and retention strategy and how the shortage has forced IT firms to lower new-hire requirements.
As tech writer George Lawton points out in a recent article, robotics is quickly moving from the realm of science fiction into business fact. But how hard is it for your everyday programmer to create their own robot software? Thanks to open source initiatives, it may not be so difficult. Here’s some stuff you can make. Continued »
I was in for a surprise when I sat down to interview Aga Bojko, the author of the first how-to book on using eye tracking for user experience (UX) research. Bojko immediately said that she doesn’t use eye tracking in her work as Indiegogo user research director, so much of our conversation focused on the effective, low-cost user testing approaches and tools she does use.
Eye or gaze tracking, the act of using a device to see where on the screen the user is looking, is an accepted user research method, but it’s not a practical one in a fast-paced product development environment, said Bojko. Although the prices of eye tracking tools are decreasing – consider The Eye Tribe’s $199 device – it’s not just about the tool, but the time and skills needed to use it and analyze and interpret the eye movement data correctly. “I’d never ask for an eye tracker, even though I wrote the book, and that should tell you something.”
The rise of mobile devices and Agile development also make eye tracking impractical for many businesses.
Getting standard UX research done fast enough in Agile iterations is hard, and adding eye tracking would make it harder. “It adds to the setup time, and … there’s time spent in analysis,” Bojko said. “In quick, iterative user testing, you learn just a bit more with eye tracking. Is it worth it? Not with the speed that we’re moving here.”
As for its use in mobile user interfaces (UIs), “it’s hard to do eye tracking on small screens, because eye trackers still suffer from accuracy issues,” Bojko told me. “It works much better on larger interfaces.”
Tracking users’ eye movements may not be a regular practice for Bojko and her research team, but personal contact with users is. Indiegogo’s researchers meet with customers in the headquarters in San Francisco and also go to customers’ and potential customers’ businesses and homes.
When a face-to-face meeting isn’t possible, Bojko uses remote UX testing services, like Validately and UserTesting. Such services provide videos of people using products and offer a very quick turnaround time.
Overall, Bojko enjoys the moments of revelation that come from seeing people use the product features she helps create. “There’s nothing that beats watching people achieve their goals using our product and getting their feedback on the spot,” she said. “That’s why I was glad to get a job not specifically for my eye tracking expertise, but for my user experience research expertise in general.”
This past October, HTML5 celebrated its official one-year anniversary since the standard was declared “complete” by the World Wide Web Consortium (W3C). Now we can expect HTML5 to power ads and media playback, forcefully kicking Flash right out the door it was already heading towards.
Recently I wrote a post on the issues surrounding development documentation and product specifications. At this year’s JavaOne Conference in San Francisco, I attended a session on building microservices and asked presenter Reza Rahman, Java EE Evangelist for Oracle, how microservices will impact the documentation issue.
“It’s going to get worse and worse…depending on how fine-tuned the services are,” Rahman said. “These APIs don’t simply document themselves.” The problem is exacerbated even more by the use of REST as opposed to traditional SOAP, since documentation will have to be updated to coordinate with this potentially unfamiliar language.
This is part of a series covering sessions at Oracle OpenWorld and JavaOne, 2015
After giving a brief description of what microservices are (using the words of James Lewis: “Applications that fit in your hand”), Daniel Bryant of the software and DevOps consultancy OpenCredo kicked off his instruction on building microservices by explaining some of the core decisions development teams will need to make up front.
The first “up-front decision,” Bryant explained, is picking your deployment platform. Will you use CloudFoundry? Amazon ECS? Kubernetes? Or do it yourself?