In one of my previous postings, I wrote about some benefits test automation may bring to software development. Automating testing processes is among software development best practices that allow for shortening release cycles, achieving better efficiency, and lowering costs. So today, I’m going to say a couple of words about common challenges test automation faces.

• Test automation is a time- and resource-consuming process. Though it helps shorten software development lifecycle in a long term, on the initial stage it will take time and require the work of a team of engineers to write necessary test scripts, sound and bug free.
• If we speak about tools for test automation, their introduction into company’s testing processes will also require a certain period of time. The time is needed for the evaluation and selection of necessary tools, whether it is Selenium, JUnit, Cucumber, or HP Mercury Load Runner. Furthermore, the solutions are often marketed as simple tools that record and repeat certain actions, but real-life automation involves much more than that, which requires the knowledge of scripts.
• Test automation doesn’t mean that there will be no need for manual testers at all anymore. There are some areas where only manual testing can help. Software development best practices are to combine both automated and manual testing applying them in the areas where they are most effective.

So, the recommendation is to combine automated and manual testing to get better testing results. Thus, quality assurance best practices recommend automating repetitive tasks humans are not very good at, as they may get tired or distracted when there is a need for repeating the same thing multiple times.

However, when it comes to investigative work, no tool can replace a human, as people are much better at noticing something strange, being suspicious, getting a funny sense about something and looking into it further. That’s why software development best practices, in terms of testing, are in taking what people do well, taking what machines do well, mixing those things together, and making overall testing incredibly powerful.

Recently, I’ve stumbled upon an article by Joe McKendrick (an information technology analyst) at ZDNet.com dwelling on “big software glitches” and their consequences we observe today. There were mentioned the latest Toyota’s troubles, the failed attempt of TD Bank to integrate two IT systems after a merge, and more. Joe also pointed out that QA tools or security middleware can help.

The sad picture shown made me again convinced that IT best practices may be get ignored even by big enterprises. In fact, large enterprises, mid-sized and small software companies are pressed today to “do more with less”. Unfortunately, this “less” sometimes means not using quality assurance best practices, which results in software failure at the wrong time. And the consequences of those failures may range from just unpleasant to horrible. With big enterprises the situation is even worse, as we speak about huge amount of information they deal with, or loads of customers using their products, etc.

The solution, however, is really simple. Implementation and use of quality assurance best practices, including unit testing, stress testing, re-testing after fixes or modifications, and test automation, which provides more accurate testing in a shorter period of time. Good management, focused on utilization of IT best practices, may help avoid the occurrence of software failures like described above, and ensure software stability in the longer term.

In a software development lifecycle testing is a critical part, on which the success of the software product depends greatly. Software development best practices recommend implementation of quality assurance as early as possible. There is a difference between finding a bug right as it appears and discovering it just before the release. And this difference costs a lot, so the sooner bugs are found, the better.

If you outsource software development, you’ve got at least three ways of applying quality assurance. You may outsource development to a third party, and test the product in-house. You may outsource the development to one provider, and testing to another. Or you may outsource the entire software development lifecycle, including testing.

The first variant, though it may seem to provide better control over the software product, makes sense only if you have elaborated quality assurance best practices and experienced testing and quality assurance experts in-house.

The second variant supposes that you outsourcing provider has software development best practices implemented, technical expertise that your product require, etc., but you want independent quality assurance done by a provider that specializes in quality assurance exclusively. This may be a wise decision to ensure the quality of outsourced software; however it may be time-consuming, because productive communicating will need to be established between you and two independent providers.

The third variant may shorten software development lifecycles to a great extent, as the processes of development and testing are interrelated. This allows test the parts of software as much as possible as early as possible, thus lessening the possibility of overlooked bugs. This variant supposes the company-outsourcer utilizes well-elaborated software development best practices and quality assurance best practices together, have already established workflows between developers and testers, etc.

You may find more about how to integrate QA testing from this software testing white paper.

When a company considers offshore software services and is about selecting a provider, one of their major considerations lies within the area of intellectual property security. Intellectual property comprising a lot of information, confidential information, sometimes marketing secrets, is, no doubt, an asset a company doesn’t wish to risk losing.

However, according to WIPO (World Intellectual Property Organization), right IP protection and due diligence procedures may help mitigate the risk of intellectual property loss.

While everyone understands the necessity of risk assessment, proper IP due diligence is something that is not always done. However, this helps to evaluate the potential partner, define potential risk areas, and, as a result, decide on which activities to leave in-house, and which of them to move offshore.

As for protective measures, they are the must as well. It’s not the question of confidence in a company providing offshore software services. It’s a question of protecting your data, valuable information you possess. Implementing electronic measures, such as limited access to information, physical security capabilities of the service provider, using protection passwords, as well as encryptions, firewalls and VPN to prevent security flaws, etc. will help to provide physical protection of your data. However, remember to protect your intellectual property legally. The measures include understanding the legal system of the country you get offshore software services from, state all IP-related issues in the contract, the same thing with licensing and code ownership, provide for information confidentiality agreements, and so on.

Yes, the things sound as you’ve-got-a-lot-of-work-to-do, but in case when offshore software services are considered, that’s a work worth doing.

Social networking tools and platforms can be useful to an enterprise while conducting their social network marketing campaign. Social network development very often faces a range of challenges developers need to overcome, in order to embody customer’ ideas in social networking tools.  Below is an example of such a challenge, and the way it was managed.

When developing a social network application, one has to decide on a platform. Typically, one would waver between Facebook, the most popular social network application development platform, and OpenSocial, a strong alternative that is featured on many popular sites, such as MySpace, Yahoo!,  Salesforce.com, etc., guaranteeing an application’s interoperability between them.

The dilemma is caused by the fact that the two APIs require completely different application architectures. Understandably, creating an application that supports both Facebook and OpenSocial is a time- and effort-consuming task.

The developers of Openface, a PHP framework designed for building applications that run on Facebook and OpenSocial platforms have found the way to solve the problem. Openface is a set of libraries enabling the same codebase to run within Facebook, Bebo, and OpenSocial containers. Using Openface is supposed to help a developer create an application that can be run on both Openface and Facebook, in the amount of time it takes to develop an application on just one of the platforms.

You may find more about effective social networking solutions  from our “Engaging Social Networks: A Review of Five Marketing Campaigns and Five Tools” white paper.

Today social networking is more than a means of socializing, finding friends, or having fun. It’s a powerful marketing instrument that helps an enterprise evaluate public opinion about the software product or the enterprise itself. It possesses good potential for both ISVs and interactive agencies, providing them with the ways of staying competitive on the market. In order to use this potential, social networks development and marketing go hand in hand. However, very often it is a challenge for advertising agencies due to the lack of technical skills.

One of the major challenges is lack of IT staff skilled in social networking development. Commonly, advertisers do marketing rather than social network development. Even if they have a LinkedIn/Facebook developer in-house, when developing a social network application, one has to be skilled across a variety of platforms. There is still no universal development platform for all social networks. In other words, different platform APIs may require completely different application architectures and technologies.

Nevertheless, implementing social media technologies into marketing is as much a challenging task for an enterprise as it is rewarding. That’s why industry leaders, as well as smaller companies, utilize social networking development to remain competitive, and do this in a variety of ways.

Find out more about how social networking development benefits marketing campaigns from our “Engaging Social Networks: A Review of Five Marketing Campaigns and Five Tools” white paper.

One of the most common mistakes of mobile application development is ignoring specific features of different mobile devices, including the differences in Internet connection, modal/ modeless forms, operating systems.

The developer must not just keep those differences in mind, but optimize the product, so that the difference between models does not become a hassle to the application functioning on any compatible device.

Here are the things to be aware of:

In terms of Internet connection mobile applications may use any of existing ways: via Wi-Fi, Bluetooth, GPRS, ActiveSync, etc.

Another thing to be aware of is that if a modeless window of a mobile application loses focus to another window, there is no opportunity for the user to retrieve it. Therefore, a mobile application absolutely should use modal forms for the main application form, messages, and every form that has input controls, in order to exclude the risk of focus being lost.

Ideally, a mobile application for an operating system of a certain version should work on all devices supporting this operating system, as long as the application is not intended to be used on mobile devices of a specific model.

Being compatible is one of the essential conditions for a mobile application to be a success. Learn more about what mistakes to avoid in mobile software development from our IT white paper.

Creating convenient and effective navigation is a challenge for mobile application developers aiming to succeed in mobile UI design. Users are capricious, normally, and difficult to please. However, to mitigate the risk of users’ dissatisfaction with a mobile application’s UI, there are some recommendations:

- One of the fundamental principles of building a mobile user interface is to minimize switching. Ideally, a user should be able to stick to just one method of navigation (out of three main means – keyboard, stylus, or fingers) all the time.

- Just as crucial as that, is to make sure that the size of controls is adequate for touch navigation. There can be nothing worse about a UI than if it causes a user to make several attempts to hit a button.

- Also, never forget about the distance between controls: it should be there to prevent mis-hits. Avoid creating large gaps, however: we recommend that spacing between controls does not exceed 10 pixels.

- Another thing that is worth considering, is implementing a night mode. Color perception is greatly dependent on light conditions; when there is little light, the human eye perceives dim colors better. In night mode, therefore, the gamma should be dimmer, which enables the user to work comfortably during nighttime.

Look through “Five Mistakes to Avoid in Creating User Interfaces for Mobile Devices” white paper to find out more.

When developing mobile devices, user interface is, probably, one of the major things to worry about. Whether the UI is convenient or not, effective of bewildering, simple and logic or difficult to understand makes half the success of mobile device software. Development of user interface thus should start from elaborate designing.

Poor structuring and ineffective usage of screen space is one of the UI designing mistakes which are better being avoided then fixed. Effective mobile user interface design should comprise:

• Thoughtfully structured controls, well-thought-out, brief texts and captions;
• Using intuitive icons instead of text in order to save screen space, however, being reasonable enough not to jam every bit of the screen with them.
• Using screen space efficiently, avoiding unnecessary graphics, borders, etc.
• Grouping related menu items into sub-menus to make the UI more convenient, however sticking to reasonable limits and avoiding endless lists of sub-menus.
• Dividing the screen into several zones of fixed height and assigning a specific function to each of them.

Check out our IT white paper describing “Five Mistakes to Avoid in Creating User Interfaces for Mobile Devices” to learn more about effective mobile user interface design.

Continue talking about multisourcing challenges that Gartner has explained recently. Gartner has also outlined key competencies that may help companies overcome “sourcing inefficiencies caused by misalignment, idle resources, unnecessary processes, overloaded operations, a heavy inventory, or a lack of focus.” I think the competencies outlined below are an important thing to consider for companies who outsource software product development. Software product  development is no exclusion.

Well, here’s what analysts recommend:

• Strategy management which aligns business needs, strategies, and governance with sourcing activities in order to ensure better ongoing business support.
• Risk management which helps prevent and mitigate, and reduce the level of risks across sourcing deals.
• Financial management – to design financial targets with the business and develop clear vision of all sourcing activities.
• Demand management – to optimize resources and skills across all sourcing activities.
• Service management – to achieve seamless, end-to-end service delivery.
• Program management which aligns portfolio and sourcing strategies to make software product development achieve desired results.

The analysts also recommend paying attention to relationship and HR management, as well as management of performance and contracts. Among the factors that turn outsourcing (multisourcing to be exact) successful were also named integration of all the providers, developed an agile sourcing environment, and targeted IT spending.