Bring your own device is still a grassroots rebellion for many companies, but it could be the future of the enterprise desktop Users are taking control, and there’s no stopping them from bringing their own network-connected devices to the office. Today, every phone, every e-book reader, every tablet is a gateway to a world of networked computing. Even if you tried to prevent your employees from bringing them into the building, only military-grade security could stop them.The BYOD (bring your own device) trend is variously viewed as “a world of pain for IT,” a security challenge that’s solvable, and something to be tolerated with grim resignation. And there’s a growing industry of companies who want to help you stop it, cripple it, or control it.[ Subscribe to InfoWorld’s Consumerization of IT newsletter today, then join our #CoIT discussion group at LinkedIn. | Get expert advice about planning and implementing your BYOD strategy with InfoWorld’s 29-page “Mobile and BYOD Deep Dive” PDF special report. ] But my experiences at Sun Microsystems suggest BYOD is an opportunity waiting to be grasped for enterprise IT executives — a move to management by standards rather than centrally purchased company desktops. It means selecting a basket of server-supported standard capabilities (IMAP, LDAP, PDF, HTML5, ODF, and so on) and telling people that anything that works securely with those standards is acceptable. It also offers the prospect of letting people use open source software that works with those standards, rather than having to buy everyone the same expensive proprietary software and instantly depreciating hardware, then manage them expensively until they are legacy systems.That’s not just a guess; I’ve seen it work. After a foray into a mandatory managed desktop at Sun for all staff (which instead forced people to break the rules and work around it to get their job done), we had a dual-path strategy for most of the time I was there. We had managed desktops accessed through low-cost screens for general staff; we obviously used Sunray, but contemporary options range from a bolted-down Secure Global Desktop from Oracle to the shiny delights of Google’s Chromebooks.We had what today would be called BYOD for the many outliers, with optional support packages available charged to the employee’s department if used. In the middle ground was a company-purchased laptop with a default configuration that became BYOD on delivery. I think this sort of arrangement is the obvious and inevitable destination for corporate IT today. By taking a dual path, it’s possible to offer a fully screwed-down desktop for people who have no tech skills at all, thus reducing support costs by eliminating all choice. Meanwhile, those with tech skills could look after themselves in a world of server-supported standards, again at reduced cost. If they couldn’t look after themselves, they were switched to the managed remote desktop. Switching to standardsThe key is to get away from thinking about controlling the applications and instead to controlling the standards, both protocol and file format. The cost of any desktop activity follows the maintenance and protection of the desktop state. When you have to manage statefulness at the PC desktop, you must have arms that reach to every desk and eyes that watch every screen. Attempting to do this results in the total surveillance mind-set that rules many enterprise IT policies. It also results in potentially enormous expenses, which can lead to a loss of quality and service to the end-user if budgets are cut. When you centralize the maintenance of state to a virtual desktop server, you immediately slash the costs involved. The screens from which people work can become generic, managed by a service company or even by a courier delivering new boxes to plug in when old ones break. There’s no state to manage — just furniture with power and data cords.There’s just as much software state, though, so the costs of managing the software and data used at each desktop remain the same as for desktop PCs. To control that cost, support contracts for desktop applications seem essential, and the attractions of traditional proprietary procurement are strong.But what if the employee controls the device through which they interact? What if your enterprise IT control stopped at the server and left the desktop to the discretion of its user? If it were possible to do this, your costs would be limited to the servers that maintained data and state for application consumption, and not much further. Naturally, you need strong hygiene rules — good security and authentication, for example — but you’ll need those whatever you do, along with most of the other capabilities your users will consume. But you’ll not need to manage the desktop, either at the PC/device or at the server. Open source libertyAn immediate concern that’s raised when this idea is proposed relates to software licensing. How do you ensure that your users are using properly licensed software? The answer is via a repository of approved software, populated as far as possible with open source. One of the great unsung benefits of open source software is that it comes with full permission for anyone to use it for any purpose without further actions. While there are more rules for developers, end-users are free to use open software for anything — period. No “software asset management.” No licensing or compliance police. Just full-function freedom.Do corporate IT departments need to worry about open source license compliance? Obviously, respecting authors and obeying the law are important, but for most of us the answer is probably no, there are bigger things to worry about. Open source software comes with a set of liberties commonly called “the four freedoms.” My summary: Any software under an open source license may be used, studied, modified, and distributed for any purpose, as long as the license is obeyed. I believe all the benefits of open source are the first and second derivatives of these freedoms.There are issues that companies shipping open source code as a part of products need to keep in mind, but in my view, they are no more complex and burdensome than the issues arising from shipping proprietary software. It’s important to make sure you know you have the necessary rights to everything you ship, and when you ship code made from proprietary elements, you naturally do so. Only sloppy developers fail to do this, and the Linux Foundation’s program is a fine cure for that sloppiness. Some staff will have needs that can’t be met with BYOD or with open source software. There are always exceptions. There’s no need to make your whole IT strategy revolve around them. Provide the capabilities they need, then charge them to their department cost center. By doing this, the real cost of using proprietary software or managed hardware will be reflected at a local level where local management can minimize it through local action, including training or changes to hiring practice.Big challenges are often golden opportunities in disguise. Maybe BYOD is the challenge that, properly embraced, can finally bring open source to the enterprise desktop.This article, “Is BYOD the precursor to the open source desktop?,” was originally published at InfoWorld.com. Read more of the Open Sources blog and follow the latest developments in open source at InfoWorld.com. For the latest business technology news, follow InfoWorld.com on Twitter. Open SourceCareers