But for companies seeking custom solutions or that want to
have various applications on multiple devices it comes down to investing in a mobile enterprise application development infrastructure. Many
cross-platform development schemes focus more on the client side than robust integration options, leaving some enterprises to choose between a Mobile Enterprise Application Platform (MEAP) or a native
development approach. (There are many
enterprises that don't need anything as comprehensive as a MEAP or
full-scale native development.)
Deciding whether MEAP is the way to go
Gartner's famous "The Rule of Three" maxim breaks down the MEAP decision process to the number of apps, the number of device classes, and the number of app updates per year. In each case, if they total less than three, MEAP is probably not a realistic mobile development approach for the company. Also, those that don't have to integrate with existing systems can direct their users to the apps supplied by the enterprise application vendors of the services involved.
If an enterprise decides to go the MEAP route, the list of requirements for a solution can look like this:
- Integration into existing systems.
- Ability to access other data streams on the chosen mobile platform.
- Support for Apple, Android, RIM, and Windows.
- Customization options using in-house developers.
- One platform for all in-house and external applications, with enterprise pricing for existing and new developments.
- Simplicity, especially the ease of integration and quick deployment.
- Write once and deploy anywhere.
All of those requirements can pale in comparison to the anxiety that might accompany selecting a MEAP, since they are not easy to move away from once they're deployed, according to Daniel Maycock, consultant for Slalom Consulting. As expected, not all MEAP solutions are created equal, and companies looking for these solutions often place a premium on flexibility.
Yativ draws a distinction between his company's MEAP offering and others by highlighting the idea of choice. When using Magic, development can proceed natively or in hybrid mode, which includes part native and part HTML5 or strictly in HTML5. He says the solution reduces code writing because of precompiled components that are already organized in the way they're used. Overall, he says, coding is reduced by about 50%. Less code doesn't necessarily equal less data.
Application development for mobile has its share of big data to deal with, such as running multiple apps on multiple servers and using dispersed databases, explains Yativ. Along with those challenges comes storage issues like whether to store the data locally or externally. Especially for new enterprises, says Yativ, data proliferation is one of the biggest problems. So, perhaps in reinforcing its mantra of "Outperform the Future," Magic has incorporated big data tools into its mobility offering.
Magic xpi Business Process Editor | Image credit: Magic Software (See an enlarged view of the image.)
Magic xpi Data Mapper | Image credit: Magic Software (See an enlarged view of the image.)
Breaking down the silos
For all the challenges associated with mobile development, Yativ sees treating the process in a silo as not being optimum and that the best answer is integration as part of the overall architecture, something he says is enabled with Magic because the solution delivers the whole architecture rather than just one piece. That advantage ties into the solution's performance.
In any development environment, speed and efficiency are always concerns; one area where that is a primary focus is how quickly developers can get up to speed on a platform. Speed, according to Yativ, has always been Magic's claim to fame, going back through its more than 30 years in the industry. Yativ says a programmer who has no previous experience with Magic will be able to write a small app within two to three weeks.
Yativ acknowledges the general expectation that mobile development tools should be inexpensive, and claims Magic fits that bill. What he terms as a "full studio" starting point costs about $5,000, and for those looking to develop a mid-range project the cost scales up to $50,000 - $60,000.
Duane Craig reports and writes on technology, construction, finance, food, and agriculture. He's been published in trade print magazines, the Washington Post, and widely on the web.