Skype in the Enterprise? Not without major changes
The last few days I’ve seen several posts talking abut how Skype is now going to run rampant through the enterprise world. While some posts were from people I know and respect, those folks are simply off the mark.
I know we love Skype. I love Skype. For personal use. But as an enterprise architect with thirty years experience (in the large enterprise), it’s not going in my network. And ongoing discussions with my colleagues in network and security fields confirm Skype isn’t showing up in their networks either. The common response remains “not on my watch.”
The popular opinion that Skype will take over the enterprise is widely held, but I’d call it an urban myth. Popular because we love Skype. Popular because we love free. The notion is particularly popular among entrepreneurs and startup visionaries. In their business it does make good sense. But they aren’t designing, operating, maintaining or securing an enterprise business network. Many of them have never worked in that environment, and simply don’t grasp the ramifications.
There isn’t one single reason. This isn’t a problem that one change in Skype will fix.
First and foremost, the heritage of Skype will always be Kazaa. The Chief Information Security Officer (CISO) of every enterprise is going to remember that for a long time. It isn’t just P2P, it’s the DNA Skype was formed from. We’re three generations of CISOs away from embracing the core of Skype and forgetting the heritage.
Second, there are zero business controls. None. For enterprise business to adopt Skype, the whole supernode architecture will have to change dramatically. Clients will have to be pointed at a specific supernode or set of supernodes. And the client can’t ever be promoted or escalated into a supernode. In essence, the supernode of today will have to be more like a PBX with configuration and management controls that don’t exist today. And the client will have to be revamped to provide controls that also don’t exist.
Third, the encryption deployed in Skype positively precludes it from enterprise adoption. Key escrow doesn’t exist. The algorithm is a black box. Enterprise business can’t buy a magic black box. HIPAA, SOX, ITAR, and a host of other regulations require audit and configuration controls that simply don’t exist. Some organization must either have and document or submit to third party key management systems that Skype doesn’t use, support, or from all I’ve seen, even acknowledge.
There are several more sound business reasons, but I won’t prattle on endlessly. I think I’ve made my point.
Skype is a renegade telco. As consumers, we love that. We’re ebullient about it. But praise gone wild sounds like technologists deluding themselves into thinking an enterprise that isn’t using Skype is being foolhardy. The opposite is true. An enterprise business has to protect shareholder value and make sound business choices about technology.
There are two phrases that come to mind. There ain’t no such thing as a free lunch (TANSTAAFL) and you get what you pay for. Both are applicable to enterprise adoption of Skype.
Enterprise business simply can’t afford Skype, at free, or at any price in its current design. The enterprise value proposition simply isn’t there, no matter how much many of my friends and colleagues might wish it.
Don’t mistake the enterprise passing on Skype for being backwards or slighting VoIP. VoIP is big and growing in the enterprise, although not quite as quickly as industry pundits proclaim. But it’s growing globally and will continue to accelerate in deployment.
Skype, on the other hand, is at least 4-5 technology generations from being the right fit for enterprise business. And to get there will require Skype hiring the right people, and listening to the real-world requirements of the enterprise from people who design, operate, maintain and secure those large networks. I haven’t heard anyone in that space say they’ve had any discussions with Skype. And none of the leading architects, designers and developers of real enterprise scale voice services have been in the “just moved to Skype” news that I’ve seen.
on June 3rd, 2010 at 5:53 pm
[...] This post was mentioned on Twitter by Ken Camp, Skype Phone Store. Skype Phone Store said: Skype in the Enterprise? Not without major changes | Stardust …: The last few days I've seen several posts talki… http://bit.ly/aSe0lL [...]
on June 6th, 2010 at 3:48 am
Ken,
These are interesting thoughts there and I am sure you are more than correct for large enterprises.
There are two things that interest me though:
1. Where do you see Skype for SIP, which allows long distance calls to businesses by connecting their SIP phones to Skype as a long distance calls provider? That would fit larger enterprises as well.
2. Is this conclusion the same for GTalk? And if it is, then it should be the same for Google Docs as well, which would make me question Google’s strategy.
I’d love to hear your thoughts on these,
Tsahi
on June 6th, 2010 at 10:04 pm
Great questions Tsahi, and I’m not sure I have good answers or even useful opinions. Sype for SIP is interesting in concept, and certainly fantastic for small and mid-size business. I’ve yet to find a single enterprise talking about it or interested. Given a global enterprise business buys in volume from major carriers, with SLA expectations across multiple services, I honestly don’t know where Skype for SIP fits or if it does. The big organizations I’ve worked with all deploy numerous Session Border Controllers and want to manage their own gateway activity for a number of reasons.
I feel less capable of answering about Google stuff. I love it for small to mid-sized businesses. Again, I’ve simply not run into a major enterprise that takes it seriously as a contender for the enterprise business. It plays in niche space. I believe the growth of cloud computing initiatives could lead to more widespread adoption. Possibly. I also believe the corporate mindset of control, collaboration tools (Sharepoint) and intranet-like security of corporate intelligence will make large enterprise businesses mindful of security and audit requirements in ways that the small-mid business may not have to consider in quite the same ways.
There are other large enterprise issues with Google Docs type integration that tie into Active Directory integration with corporate apps, DNS and domain control, etc. And then there’s the very large world of Lotus Notes, Sametime and Mainframe/AS400 applications that simply aren’t going to move to Google (or integrate Skype).like a small startup.
I love these tools, but I’ve got this background and active current involvement in some global enterprise networks, applications and solutions where these tools not only don’t fit, but are actually opposite the mission and many business requirements.
on June 7th, 2010 at 11:56 am
I think I see your points, but I’d like to ask what then is the high level difference with something like MS Communicator? Wouldn’t e issues be the same, but the administration differenent because of AD and other tools which are known about in the enterprise settings you are more familiar with?
on June 7th, 2010 at 12:24 pm
Great question Antoine. I don’t see the comparison as an issue. Like you mentioned, the AD integration is a big deal. Not just for familiarity but for the enterprise controls. It also tightly integrates with things like Exchange, Sharepoint q d other widely deployed enterprise tools. MS has a strong history of playing to the enterprise whereas Skype doesn’t. That track record will go a long way.
on June 7th, 2010 at 2:07 pm
Makes sense. I’m growing in my knowledge in this area, and Tsahi has helped there a lot. Had that as a remaining question, thanks for you’d response.
on June 7th, 2010 at 2:47 pm
My pleasure. I agree. Tsahi is one sharp guy to pay attention. Lots of knowledge in his head for sure.