analysis This week I came back from a week’s worth of holidays to find that two of the country’s leading analyst firms had reached a complete disagreement about one of the most fascinating trends sweeping Australia’s enterprise IT landscape: Bring your own device (BYO) computing.
Firstly, a media release issued by Dimension Data trumpeted research the systems integrator had commissioned Forrester to carry out. In the release, DiData highlighted the fact that in Australia, between 10 and 18 percent of organisations recently surveyed were already providing support for employee-owned laptops, tablets and smartphones. In addition, even more local groups — between 18 and 26 percent — were planning to do so over the next two years.
As you would expect him to, given his company’s role helping organisations with such plans, DiData’s chief technology officer Ettienne Reinecke said the research showed “growing employee interest” was pushing organisations to “think differently” … perhaps an appropriate turn of phrase, given the fact that most of the tablets and many of the smartphones employees are bringing into the workplace are Apple-branded.
And yet not everyone’s so keen on the BYO device phenomenon.
A separate media release issued by Ovum took the opposite slant, claiming that “only around 10 percent” of companies it had recently surveyed said their preferred model was for employees to provide their own devices. Of course, Ovum acknowledged the BYO trend would be significant long-term, as company staff clamoured to use their iPad at work, but advised chief information officers to take a cautious approach and ensure their device management capabilities and interest in cloud computing solutions — which can aid in delivering device-agnostic services — are up to date.
Now there’s two points I want to make about these two pieces of commentary about the BYO trend in Australia. Firstly, there’s the obvious point that although they’re coming at it from different angles, both Ovum and Forrester have in fact published similar data about the adoption of BYO policies in Australia. They have both arrived at around the same figure — 10 percent — for organisations which are already supporting the trend.
But the second point is perhaps the most important one: Framing the BYO debate as a yes/no question, as both firms have done, leads to a false dichotomy.
If you read into the DiData report produced by Forrester (PDF), you’ll find that the report isn’t fundamentally about BYO computing at all; it’s about desktop virtualisation; an area much nearer and dearer to the system integrator’s heart. And in fact, the report highlighted several fascinating findings about the issue and how it’s affecting organisations.
Firstly, Forrester found that there was no doubt that organisations “of all sizes, industries and geographies” are embracing client virtualisation, with the most compelling justification for adoption of the technology being the desire to escape the “seemingly never-ending ‘rip and replace’ refresh cycle” driven by desktop infrastructure refresh projects — which are often accompanied by simultaneous operating system and application overhauls.
Secondly, the analyst firm wrote — and as I recently argued with respect to cloud computing — IT managers are embracing complex, ‘hybrid’ approaches to the issue. Very few organisations are virtualising everything on the desktop. Instead they’re taking a tactical approach, using software as a service solutions alongside thin clients, on-premise and even thick client options, depending on what approach will best fit business needs.
And lastly, and perhaps this is Forrester giving a nod to DiData, the report claimed IT managers should rely on systems integrators to support them in the process. I don’t disagree with Forrester’s argument that in many organisations, desktop virtualisation migrations are a complex procedure that neither their disparate desktop support and datacentre teams are completely prepared for; but I do think such a conclusion shouldn’t be on page 2 of a report sponsored by DiData.
At least, not if you want the report to be credible.
Way down in the report is the contention that, alongside the release of new operating systems like Windows 7, BYO strategies actually represent what Forrester describes as a “major market enabler” for desktop virtualisation. One anonymous enterprise architect’s quote sums it up well:
“About a year and a half ago, I put together an application and desktop virtualisation strategy. Unfortunately, we then sat on it for a couple of months — but virtualisation’s important re-emerged with mobile devices. We realised we needed to diversify our computing ecosystem, as we felt we had been locked into proprietary plaforms. Today we’re focused on iOS support for iPhone and iPad; next up will be Android.”
Are you noticing what’s going on here? BYO computing is not a stand-alone trend. In fact, it’s not even close to being the main issue for IT managers in administering and developing their corporate systems. But end user support for the idea — CEOs and middle-managers who want to use their iPad and iPhone at work — is providing a strong justification for IT managers and CIOs to be able to re-architect their back-end systems to meet user demand.
Without BYO computing, in short, virtual desktop business cases would have a hard problem getting off the ground.
You can see this trend in the wild, too. An article published by CIO Magazine last week highlighted Southern Health’s (located in Victoria) rollout of a BYO policy, including iPads, within its organisation. But if you delve a bit deeper (which CIO’s sister publication Computerworld did a few weeks back), you’ll find that the real activity going on at Southern Health is:
“The replacement of a number of core enterprise systems such a Patient Administration, Financial Information Management, Rostering and Payroll.” — Southern Health IT director Paul Jurman
In short, BYO computing was linked to core application replacements. iPads … were just the public face which Southern Health put on a much more systemic overhaul of its critical applications.
When you’re talking about BYO computing, it’s very easy right now to get locked into a polarised debate about whether end users should be able to bring in their own devices to workplaces. And it’s a debate which every staffer who’s locked into using a 2005-era HP desktop at work with Windows XP is interested in having — because using old technology at work, when you’ve got a 2011 MacBook Pro sitting at home — seems ludicrous.
But this is primarily an end user debate. For IT managers and CIOs, the debate is a lot more complex than ‘yes’ or ‘no’. For Australian IT managers and CIOs right now, the debate should not be “should we support BYO”, but “how can we leverage BYO to get other outcomes in the business”.
This balance between populism (“I want my iPad at work”) and pragmatic reality (“How can we architect our ageing enterprise IT systems for business agility?”) is one which IT managers will face constantly in their careers — as many other professionals will as well. Right now, with BYO computing, there’s a golden chance to balance these competing priorities out; if Australian CIOs can frame the business case right.
I love the quote from the “anonymous enterprise architect” about being locked into proprietary platforms – then saying they are going to iOS and Android (which is really proprietary as I doubt there’s many (any) businesses who would recompile the source code with their modifications and install it on their chosen hardware platform).
For anyone doing BYOD – you need to re-focus from “attaching foreign devices to the corporate network” to “presenting our applications, services and data to any platform securely and ubquitously”. If you continue the traditional wired network methodology (ie how to do I attach all these foreign devices securely) into a mobile environment you’ll waste a lot of money trying to scale managing numerous devices.
Well I interpreted the proprietary platforms comment as meaning they are going to be delivering much of their services through a web browser — which, given the right page templates, can be viewed on any device under the sun. I don’t believe many enterprises will be into installing custom software on iPads etc … the management tools to keep the software updated etc are just not there from Apple. Much easier to do the SaaS thing.
For anyone doing BYOD – you need to re-focus from “attaching foreign devices to the corporate network” to “presenting our applications, services and data to any platform securely and ubquitously”. If you continue the traditional wired network methodology (ie how to do I attach all these foreign devices securely) into a mobile environment you’ll waste a lot of money trying to scale managing numerous devices.
I agree with this.
There has been one thing that I consider why BYO would never been feasible before the web, and in fact the iPad, and that is control over the data path.
Enterprise often deals with sensitive data. Allowing employees to access it willy nilly gets you into trouble. Sensitive data could be lost, stolen, copied, or otherwise manipulated, without controls being in place to prevent it.
We have all heard the story, someone copied sensitive data onto a laptop or USB drive, and then that data was lost.
However, what the web and closed platforms like iOS afford, and this is key, is control over the data platform within a predictable and established environment. What does this mean?
Well, the web has largely become uniform over experience and security in recent years. You can expect a page to render exactly the same, and be just as secure, on multiple platforms, from the same hypertext. This means that you can have control over your data, especially if you use methods like 2-step authentication, etc.
iOS has the same practicality, with the exception of jailbroken iPads, ever single iPad will be the same, with a limited requirement for support. You can produce an Application and reasonably expect almost every iPad to support it, and be reasonably confident you have control of the data. Especially since you can implement features like Remote Wipe. Platforms like Android are dangerous, however, as their experience is non-uniform.
ThinClients and Desktop Virtualisation can provide the same benefits, which is good, however I believe with a caveat, it is an expensive way to do things in terms of resources, be in in terms of physically providing ThinClients to your employees, or allocating a significant part of your server bandwidth to Desktop Virtualisation.
This means of course I have an interesting point of view when it comes to iOS. I would love it to be an open platform, like Android, however, in reality, for it to become the dominate tablet device and actually be used for applications like Enterprise, it’s closed nature is a necessity. I do not expect Enterprise applications to be issued for Android Tablets unless the Enterprise in question has complete control over the device, which is a workable model as RIM has proven with their Blackberries, but does require significant investment by the Enterprise in question and may be limiting in terms of what employees want.
I don’t know about you, but I would much rather have a device I can do my personal work on and then quickly switch to anaylsising, say, the days trading, than to have to pull out another physical device.
You only made one good point in your post. The rest was iOS marketing.
The thing is – whether or not you can BYOD all depends on the company/industry you are working for and how concerned you are about data security.
This is why organisations like the Department of Defence only allow authorised devices to connect to the network.
Considering I don’t actually own any iOS devices or work for Apple I’m offended that you would call it “marketing”. I was merely pointing out why I think the platform is going to be more popular than Android in the BYO and Enterprise space.
Consequently I’m curious as to what the “one good point is”. I made several points not related to iOS.
Further, Android may require you to have complete control over the device because of it’s nature supporting multiple devices is a pain in the butt, and Android, by design allows you to do almost a lot with a given application, which means you need to be very paranoid about what you do and do not trust.
Doesn’t exactly preclude development for the platform, no, but it is a headache many will likely avoid.
Outside of the tablet space, Desktop Virtualisation and ThinClients I consider to be an interim solution for a lot of companies, or companies that use out-of-house products that they cannot create a web interface for. The advantage of these platforms, especially ThinClients, is you get more control over the data path.
This is why organisations like the Department of Defence only allow authorised devices to connect to the network.
They are just more paranoid. There is nothing wrong with that, considering, but security still comes down to the same basic principle: controlling the data path. The more sensitive the data is, the more control you will want.
I think it’s good to have options open but as I always say: it all depends on the evolution of users and not through sudden changes.
Comments are closed.